3 Nov
2011
3 Nov
'11
6:07 a.m.
A friend said that he has a collection of NOS 4000 series CMOS ICs.
If there is any interest I will get them so they don't go in the trash.
--
Michael Thompson
3 Nov
3 Nov
9:35 a.m.
Michael Thompson wrote:
A friend said that he has a collection of NOS 4000
series CMOS ICs.
If there is any interest I will get them so they don't go in the trash.
If he has a large supply he could sell/donate them to someone like
http://www.unicornelectronics.com/
Otherwise small shops like mine would be somewhat interested - but we
don't use a lot of these in our coin-operated game repairs.
John :-#)#
--
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
Call (604)872-5757 or Fax 872-2010 (Pinballs, Jukes, VideoGames)
www.flippers.com
"Old pinballers never die, they just flip out"
4 Nov
4 Nov
12:24 a.m.
would any of them be good with the prom program that came with the intel
mds225
On Thu, Nov 3, 2011 at 11:35 AM, John Robertson <pinball at telus.net> wrote:
Michael Thompson wrote:
A friend said that he has a collection of NOS
4000 series CMOS ICs.
If there is any interest I will get them so they don't go in the trash.
If he has a large supply he could sell/donate them to someone like
http://www.unicornelectronics.**com/ <http://www.unicornelectronics.com/>
Otherwise small shops like mine would be somewhat interested - but we
don't use a lot of these in our coin-operated game repairs.
John :-#)#
--
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9 Call
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, VideoGames)
www.flippers.com "Old pinballers never die, they just flip
out"
5:38 a.m.
On Nov 3, 2011, at 12:35 PM, John Robertson wrote:
Michael Thompson wrote:
Jameco, also; I've gotten quite a few older ICs to fix my Defender boards
(particularly 4116 RAM chips, which seem to be hard to find for less than an arm and a leg
each anywhere else, unless I'm looking in the wrong places) from them. CMOS ICs are
in a slightly different class, but he might be able to get better than scrap for them.
I've very little use for them myself.
- Dave
A friend said that he has a collection of NOS
4000 series CMOS ICs.
If there is any interest I will get them so they don't go in the trash.
If he has a large supply he could sell/donate them to someone like
http://www.unicornelectronics.com/ 
10:42 a.m.
On 4 Nov 2011 at 8:38, David Riley wrote:
Jameco, also; I've gotten quite a few older ICs to
fix my Defender
boards (particularly 4116 RAM chips, which seem to be hard to find for
less than an arm and a leg each anywhere else, unless I'm looking in
the wrong places) from them. CMOS ICs are in a slightly different
class, but he might be able to get better than scrap for them. I've
very little use for them myself.
Don't a lot of 4000-series ICs have 74HCxxxx equivalents? I've still
got a bunch of old 4000-series CMOS from when they first started
appearing. ISTR that they were very slow, unless one used high
(+15V?) supplies. I think I still have my Solid State Scientific
databook from that time.
--Chuck
3:42 p.m.
On 4 Nov 2011 at 8:38, David Riley wrote:
The common ones that don't have indentical funcionality in 74xx TTL seem
to. For example there's a 74HC4017 10-state counter IIRC. However, when
there is the same functionality but differnet pinouts (e.g 7400 .vs.
4011), there isn't a driect equivalent in 74HCxxx, the TTL pinout takes
precedence. So there is a 74HC00, there isn't a 74HC4011
And there are quite a few 'odd' fucntions in the 4000 series CMOS range
that were not commonly used and never made it into 74HCxxx, My older
Philips 4000 series data book lists an HPIB talker.listner state machine
IC, for example.
And I think (without checking the databooks) that 4000 series will stand
a higher supply voltage thea 74HCxxx
Jameco, also; I've gotten quite a few older
ICs to fix my Defender
boards (particularly 4116 RAM chips, which seem to be hard to find for
less than an arm and a leg each anywhere else, unless I'm looking in
the wrong places) from them. CMOS ICs are in a slightly different
class, but he might be able to get better than scrap for them. I've
very little use for them myself.
Don't a lot of 4000-series ICs have 74HCxxxx equivalents? I've still got a bunch of old 4000-series CMOS from when they
first started
appearing. ISTR that they were very slow, unless one used high
(+15V?) supplies. I think I still have my Solid State Scientific
databook from that time.
I know I have by 4000 series databooks...
They are slow, and thus are not commonly used in classic computers. The
odd oens do turn up, particualrly where speed is not very important but
he function was (IIRC ,the BCD interface for my HP9815 calculator has
some odd 4000 series shift registers in it, there is no direct TTL
equivalent,
The also, of course, turn up where low pawer consuption is important.
-tony
4:42 p.m.
On Nov 4, 2011, at 6:42 PM, Tony Duell wrote:
And I think (without checking the databooks) that 4000
series will stand
a higher supply voltage thea 74HCxxx
My recollection is that the 74HCxxxxx are nominal 5V parts (the HCT ones specifically run
with TTL switching thresholds). CD4000 series is full range, I think usually 3.3-15v or
thereabouts, with the switching threshold usually at 0.5Vdd.
74HC was generally considerably slower than 74F for the same part (though also took WAY
less power), 74HCT slower still.
They are slow, and thus are not commonly used in
classic computers. The
odd oens do turn up, particualrly where speed is not very important but
he function was (IIRC ,the BCD interface for my HP9815 calculator has
some odd 4000 series shift registers in it, there is no direct TTL
equivalent,
And I sure love the 405x series of (basically) analog multiplexers and 4016 analog
switches. By now there are a number of better analog multiplexers and switches out there,
but there weren't any straight TTL muxes that I remember which operate in both
directions (because they were typically combinational logic instead of pass gates).
- Dave
5 Nov
5 Nov
2:12 p.m.
On Nov 4, 2011, at 6:42 PM, Tony Duell wrote:
I thought that 74HCT were 5V-only parts (somethign like a 5.5V maximum).
74HC was wider (up to 6 or 7V, and down to about 2V). The 4000 series
are a lot wider.
And I think (without checking the databooks) that
4000 series will stand
a higher supply voltage thea 74HCxxx
My recollection is that the 74HCxxxxx are nominal 5V parts (the HCT
ones specifically run with TTL switching thresholds). CD4000 series is
full range, I think usually 3.3-15v or thereabouts, with the switching
threshold usually at 0.5Vdd. 74HC was generally considerably slower than 74F for
the same part
(though also took WAY less power), 74HCT slower still.
I didn;t think there was taht much speed difference between 74HC and
74HCT. I will have to check the data books agian. And yse, 74F (and
74AS) is the fast one...
400 series CMOS will go up to 15 or 18V (depending on the version) I
think. This led to a lot of car-related projected in UK magazines which
ran 400 series CMOS chips straight off the car battery.
They are slow, and thus are not commonly used in
classic computers. The
odd oens do turn up, particualrly where speed is not very important but
he function was (IIRC ,the BCD interface for my HP9815 calculator has
some odd 4000 series shift registers in it, there is no direct TTL
equivalent,
And I sure love the 405x series of (basically) analog multiplexers and
4016 analog switches. By now there are a number of better analog
multiplexers and switches out there, but there weren't any straight TTL
muxes that I remember which operate in both directions (because they
were typically combinational logic instead of pass gates).
3:53 p.m.
On Nov 5, 2011, at 5:12 PM, Tony Duell wrote:
Well, yes. I like that they're just switches. Buffers have their place, of course,
but sometimes it's nice to have a little DIP IC that'll switch off your headphone
port without clicking too much. They're fundamentally different devices.
I actually found the 4052 great for muxing a UART channel 4 ways because you didn't
have to use 2 chips; the Rx and Tx were treated equally because they were just
transmission gates. It would only ever be good if you were on the same PCB (or at least
not going too far), but that was what I was doing, so it was great.
They're also not particularly fast, but at least for what I was using them for, they
didn't need to be. As with anything, it's a matter of understanding the
limitations. :-)
- Dave
And I sure
love the 405x series of (basically) analog multiplexers and
4016 analog switches. By now there are a number of better analog
multiplexers and switches out there, but there weren't any straight TTL
muxes that I remember which operate in both directions (because they
were typically combinational logic instead of pass gates).
True (although IIERC some of these exist in 74HC etc).
But rememebr a digital multiplexer (like the 74x151) performs a buffering
function too -- the input is 1 TTL load, the output will drive at least
10 simialr loads. This is obviously not true of the analogue
multiplexers, which are trasnmission gates. This can be important.
And it's why you can't use 4016s, etc as bidirecitonal data bus buffers.
They perform no buffering at all.
6 Nov
6 Nov
11:40 a.m.
But rememebr a
digital multiplexer (like the 74x151) performs a buffering
function too -- the input is 1 TTL load, the output will drive at least
10 simialr loads. This is obviously not true of the analogue
multiplexers, which are trasnmission gates. This can be important.
And it's why you can't use 4016s, etc as bidirecitonal data bus buffers.
They perform no buffering at all.
Well, yes. I like that they're just switches. Buffers have their
place, of course, but sometimes it's nice to have a little DIP IC
that'll switch off your headphone port without clicking too much.
They're fundamentally different devices.
I actually found the 4052 great for muxing a UART channel 4 ways
because you didn't have to use 2 chips; the Rx and Tx were treated
I usspect you could have fitted the logic into a PAL if you ant a 1-chip
digital solution :-). Or you could have used a pair of chips ('153 and
'139, perhaps) and multiplexed a couple of flow cotnrol singals too.
More seriously, there is no reason not to use a 4052 here. It does the
job. There is no problem with speed or fanout.
equally because they were just transmission gates. It
would only ever
be good if you were on the same PCB (or at least not going too far),
but that was what I was doing, so it was great.
They're also not particularly fast, but at least for what I was using
True, but as you rightly say, that's not always important. You'd not use
one to switch the bus lines on a multi-MHz processor. That doens't mean
they have no uses.
them for, they didn't need to be. As with
anything, it's a matter of
understanding the limitations. :-)
Of courese. I 100% agree with you. As I have said many times, it's a
mistake to think that because soemthing can't solve a particular problem,
it can't solve any problem. It's an equal mistake ot assuem that becase
(a) is better than (b) in one particualr circumstance, it must always be
better.
4000 series CMOS is an example of this. It can't solve every logic
problem, and there are families that are better in some cases. This does
not make 4000 series CMOS useless.
-tony
7:36 p.m.
I actually found the 4052 great for muxing a UART channel 4 ways
because you didn't have to use 2 chips; the Rx and Tx were treated
4 Nov
4 Nov
6:02 p.m.
On 4 Nov 2011 at 22:42, Tony Duell wrote:
And there are quite a few 'odd' fucntions in
the 4000 series CMOS
range that were not commonly used and never made it into 74HCxxx, My
older Philips 4000 series data book lists an HPIB talker.listner state
machine IC, for example.
And I think (without checking the databooks) that 4000 series will
stand a higher supply voltage thea 74HCxxx
That is true--74HCxx are generally rated at +7V max. I believe that
the "middle" voltage (i.e. recommended) voltage for 4000 CMOS was
10V. At 5V, the speed dropped to less than half that at 10V. Back
then, 10 or 15V was commonly thought of as "CMOS logic levels". If
someone uses that term today, it's likely he means 3.3V or less.
National also had a parallel series of CMOS chips, with 4000
equivalents, the MM56xx series.
I still have the pile of 4010 buffers I bought new. I was using
them in a too-clever PMOS-to-TTL conversion setup (I don't recall the
details or even if it worked). They have 1976 date codes.
For a time, it seemed as if 4000 CMOS couldn't find itself. Lots of
counters and shift registers. Did anyone actually use a 4581 ALU in
anything? I've always wondered if anyone built up a CPU from 4000
CMOS...
4000 series CMOS is still in current production.
--Chuck
7:07 p.m.
Chuck Guzis wrote:
National also had a parallel series of CMOS chips,
with 4000
equivalents, the MM56xx series.
Wow, never heard of those. They had MM74Cxxx parts, which were
functionally equivalent to the other 74xx parts but with CD4000
equivalent electrical performance and timing.
Did anyone actually use a 4581 ALU in anything?
I've always
wondered if anyone built up a CPU from 4000 CMOS...
RCA had a bit-slice part in one of their 1970s COSMOS databooks, with an
app note in the back, but I've never been able to find that particular
COSMOS databook again, so if it went into production at all it must not
have lasted very long. My recollection is that it was not as fancy as
the Am2901.
RCA did make some CMOS parts intended for building a DSP processor. I
think I've got a slim "user manual" for them somewhere. Intended for
military applications.
8:30 p.m.
Chuck Guzis wrote:
These were common in Yaesu and Kenwood HAM radios
National also had a parallel series of CMOS
chips, with 4000
equivalents, the MM56xx series.
Wow, never heard of those. They had MM74Cxxx
parts, which were
functionally equivalent to the other 74xx parts but with CD4000
equivalent electrical performance and timing. 
9:07 p.m.
On 2011 Nov 4, at 7:07 PM, Eric Smith wrote:
Chuck Guzis wrote:
The CD40181 (and 40182) is mentioned in some RCA COS/MOS databooks
National also had a parallel series of CMOS
chips, with 4000
equivalents, the MM56xx series.
Wow, never heard of those. They had MM74Cxxx parts, which were
functionally equivalent to the other 74xx parts but with CD4000
equivalent electrical performance and timing.
Did anyone actually use a 4581 ALU in anything?
I've always
wondered if anyone built up a CPU from 4000 CMOS...
RCA had a bit-slice part in one of their 1970s COSMOS databooks,
with an app note in the back, but I've never been able to find that
particular COSMOS databook again, so if it went into production at
all it must not have lasted very long. My recollection is that it
was not as fancy as the Am2901. from the 70s.
Exactly what the name suggests: a
CMOS version of the 74181 4-bit ALU.
RCA did make some CMOS parts intended for building a
DSP processor.
I think I've got a slim "user manual" for them somewhere. Intended
for military applications.
Fairchild made some funky bit-slice ALU, register stack, etc.,
devices in the 70s (I think we have spoken of them here before).
5 Nov
5 Nov
12:16 a.m.
Chuck Guzis wrote:
Did anyone actually use a 4581 ALU in anything?
I've always
wondered if anyone built up a CPU from 4000 CMOS...
I wrote:
RCA had a bit-slice part in one of their 1970s COSMOS
databooks,
with an app note in the back, but I've never been able to find that
particular COSMOS databook again, so if it went into production
at all it must not have lasted very long. My recollection is that it
was not as fancy as the Am2901.
Brent Hilpert wrote:
The CD40181 (and 40182) is mentioned in some RCA
COS/MOS
databooks from the 70s. Exactly what the name suggests: a
CMOS version of the 74181 4-bit ALU.
Same with the 4581/4582. Possibly different part numbers assigned by
different vendors for effectively the same part.
I was referring to a more integrated bit-slice part, though since I
can't find any docs on it now, I have no idea what the part number was
nor the exact functions included.
Fairchild made some funky bit-slice ALU, register
stack, etc.,
devices in the 70s (I think we have spoken of them here before).
The Fairchild Macrologic parts (9401-9410 bipolar, 4701-4710 CMOS) are
reasonably well known, but I'm still looking for datasheets for their
much earlier 3800 and 3804 MOS ALU chips.
3:52 p.m.
At 7:07 pm -0700 2011/11/04, Eric Smith wrote:
RCA had a bit-slice part in one of their 1970s COSMOS
databooks, with an
app note in the back, but I've never been able to find that particular
COSMOS databook again, so if it went into production at all it must not
have lasted very long. My recollection is that it was not as fancy as
the Am2901.
I have a 1985 RCA databook titled "High-Reliability Integrated
Circuits" that sounds very similar. The parts listed as part of the
'EPIC' chip set are
- GP001 8-Bit General Processor Unit
- GP501 Emulating Controller
- GP502 Microprogram Sequencer [2910 clone]
- GP503 8-Bit x 8-Bit Multiplier
- GP511 Voltage-Level Converter and Buffer
- GP301 512 x 8-Bit ROM
- GP302 256 x 16-Bit ROM
They list as "planned",
- GP514 Double Address Select Unit
- GP515 Double Register Select Unit
- GP516 Bus Interface Unit
- GP517 Interrupt Control and Timing Unit
The related app notes, all written by one K Karstad, are
- ICAN-7202 An Introduction to the Use of the General-Processor Unit, GP001
- ICAN-7211 Introducing and Applying the 8 x 8 CMOS/SOS Multiplier, GP503
- ICAN-7226 A Primer on Microprogramming and Bit-Slice Architecture
Exemplified in Systems Using Members of the EPIC CMOS/SOS Family
- ICAN-7241 A 16-Bit Microprogrammable Bit-Slice Computer Design Using
the GP001 CPU and GP502 Controller from the EPIC Family
- ICAN-7259 A Guide to the Emulating Microprogram Controller GP501 --
With Programming Examples
- ICAN-7281 The GP501 - A Flexible and Powerful Microprogram Controller
for Emulation in the Control Section of High-Performance
Microcomputers
- ICAN-7283 A Comparison of EPIC CMOS/SOS Microprogram Controllers GP501
and GP502
--
Kevin Schoedel <schoedel at kw.igs.net> VA3TCS
4 Nov
4 Nov
6:53 p.m.
Tony Duell wrote:
And I think (without checking the databooks) that 4000
series will stand
a higher supply voltage thea 74HCxxx
Yes. 4000B series are rated for operation at from 3Vto 18V (abs max
20V), while 74HC series are rated for operation from 2V to 6V (abs max
7V). Also 4000B are generally *much* slower (at same voltage) than
74HCxxx parts. For instance, CD4011 prop delay at 5V and 25C is rated
125ns typ, 250ns max, while 74HC00 at 4.5V and 25C is rated 9 ns typ and
18ns max.
Each family still has applications where it can't be replaced by the
other, which is why (some parts of) both families are still in production.
Nowadays my favorite 74xx families are 74ABT and 74LVC. The 74ABT
family is great for buffers that need high output drive. The 74LVC
family runs on 3.3V but has 5V-tolerant inputs, with low input leakage
even when Vdd=0V. I wish there were more 74LVC parts offered in DIP
packages for ease of prototyping, but DIP is going the way of the dodo.
Ces't la vie. At least the 74LVC240, '245, '373/573, and '374/574 are
offered in DIP. I use those when I have to interface 5V inputs to a
3.3V I/O system (i.e., FPGA, microcontroller) that does not have
5V-tolerant I/O.
In the other direction, 3.3V CMOS outputs can directly drive 5V chips
that have TTL-compatible inputs, but when it is necessary to drive 5V
parts that actually expect 5V CMOS input levels, the 74HCT, 74ACT, or
74ABT series is useful.
9 p.m.
On 4 Nov 2011 at 18:53, Eric Smith wrote:
In the other direction, 3.3V CMOS outputs can directly
drive 5V chips
that have TTL-compatible inputs, but when it is necessary to drive 5V
parts that actually expect 5V CMOS input levels, the 74HCT, 74ACT, or
74ABT series is useful.
Do there exist any split supply (3.3V/5V) level converter chips in
DIP? When you're building a circuit where anybody or their cousin
can tie something to your interface, you'd like the chip most likely
to get fried by a careless user to be in an easily replaceable form.
I've got tubes and tubes of RS422 transceiver chips in DIP. I've
thought about using those on a one-off project as level shifters.
Just use one leg of the differential pair.
--Chuck
5 Nov
5 Nov
7:24 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
Chuck Guzis wrote:
Do there exist any split supply (3.3V/5V) level
converter chips in
DIP?
I don't think so. That's why I use 74LVC parts for the 5V to 3.3V
direction, and 74HCT for the 3.3V to 5V direction, if 5V CMOS levels are
actually needed. Usually they aren't.
A bidirectional bus where actual 5V CMOS levels are needed on the 5V
side is the annoying case; then I end up using two separate buffers, one
in each direction.
When you're building a circuit where anybody or
their cousin
can tie something to your interface, you'd like the chip most likely
to get fried by a careless user to be in an easily replaceable form.
Yes.
8:12 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 5 Nov 2011 at 19:24, Eric Smith wrote:
Chuck Guzis wrote:
If it's a 3.3v microcontroller on the other side, it's a good reason
for selecting one with 5V tolerant inputs, I guess. At least you can
use plain old HCT in both directions.
--Chuck
Do there exist any split supply (3.3V/5V) level
converter chips in
DIP?
I don't think so. That's why I use 74LVC parts for the 5V to 3.3V
direction, and 74HCT for the 3.3V to 5V direction, if 5V CMOS levels
are actually needed. Usually they aren't.
8:42 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
Chuck Guzis wrote:
If it's a 3.3v microcontroller on the other side,
it's a good reason
for selecting one with 5V tolerant inputs, I guess.
Sadly those seem to be going away. A lot of the new stuff is not 5V
tolerant.
Some of the Xilinx FPGAs (e.g., Virtex-6) aren't even 3.3V tolerant!
9:34 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 5 Nov 2011 at 20:42, Eric Smith wrote:
Sadly those seem to be going away. A lot of the new
stuff is not 5V
tolerant.
It feels like I've come full circle...wasn't RTL 3V logic? All
they'd have to do is start offering logic in TO-100 packages again
(we still have flatpacks; they just have more leads).
--Chuck
10:02 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 5, 2011, at 11:42 PM, Eric Smith wrote:
Chuck Guzis wrote:
There are a whole boatload of problems with the Virtex-6, and that's just one of them.
The DDR3 waveforms coming out of them are just disgusting (though they do work) and some
of the errata we've found are truly horrifying. Let's not start on the voltage
sequencing issues (which, curiously, the Virtex-5 was free of; you could power that on in
any order you damn well pleased).
I was never much of a Xilinx guy before, but the V6 made me seriously question ever using
them again (though the Spartan 6 may have kept me from swearing off them entirely).
- Dave
If it's a 3.3v microcontroller on the other
side, it's a good reason
for selecting one with 5V tolerant inputs, I guess.
Sadly those seem to be going away. A lot of the new stuff is not 5V tolerant.
Some of the Xilinx FPGAs (e.g., Virtex-6) aren't even 3.3V tolerant!
10 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 5, 2011, at 11:12 PM, Chuck Guzis wrote:
On 5 Nov 2011 at 19:24, Eric Smith wrote:
Yeah, a 3.3v CMOS output will still be putting out higher than many 5v TTL chips output. A
surprising number of modern low-voltage micros still have 5v-compatible inputs. I worked
with a seriously irritating TI DSP that, among other insults, only took 1.8v inputs (no
higher, no lower). We had 12 voltage rails on that one AMC card (about 2"x5",
highly integrated).
We did, however, like the 74AVC4T245 for voltage translation. That almost fits your
dual-supply bill, only the AVC parts don't go up to the full 5v. For that, your best
bet would be in the LVC family; they have some dual-rail '245 and similar equivalents
(e.g. the SN74LVC8T245). Nothing in DIP, though, so if you really wanted that you'd
have to mount them to surfboards (I really like the ones by Bellin at
http://www.beldynsys.com/; not cheap, but pretty nice protoboards the do mount up
nicely).
- Dave
Chuck Guzis wrote:
If it's a 3.3v microcontroller on the other side, it's a good reason
for selecting one with 5V tolerant inputs, I guess. At least you can
use plain old HCT in both directions. Do there exist any split supply (3.3V/5V) level
converter chips in
DIP?
I don't think so. That's why I use 74LVC parts for the 5V to 3.3V
direction, and 74HCT for the 3.3V to 5V direction, if 5V CMOS levels
are actually needed. Usually they aren't.
6 Nov
6 Nov
10:46 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 6 Nov 2011 at 1:00, David Riley wrote:
> We did, however, like the 74AVC4T245 for voltage
translation.
That
almost fits your dual-supply bill, only the AVC parts
don't go up to
the full 5v. For that, your best bet would be in the LVC family; they
have some dual-rail '245 and similar equivalents (e.g. the
SN74LVC8T245).
Nothing in DIP, though, so if you really wanted that
you'd have to mount them to surfboards (I really like the ones by
Bellin at http://www.beldynsys.com/; not cheap, but pretty nice
protoboards the do mount up nicely).
Those are fine for prototypes, but you can get pretty much the same
thing cheaper at Futurlec:
http://futurlec.com/SMD_Adapters.shtml
And even those are somewhat overpriced. I picked up a small pile of
100-pin adapters for PIC32 from TeamETT that have spots for all of
the close-in components (mostly chip caps and some resistors) as well
as a header for an ICD2 programmer hookup for about $0.50 each and
they even came in a cloth drawstring bag. I figure that the people
in Thailand being flooded out could do with a little help. Better
than sending your money to China.
http://www.etteam.com/
Some very fun stuff there.
I'm after a level-shifter that I can put in a socket on a production
board, not on a prototype, so it can be easily replaced by a
relatively unskilled person.
--Chuck
8 Nov
8 Nov
2:34 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/06/2011 01:00 AM, David Riley wrote:
Yeah, a 3.3v CMOS output will still be putting out higher than many
5v TTL chips output. A surprising number of modern low-voltage micros
still have 5v-compatible inputs. I worked with a seriously
irritating TI DSP that, among other insults, only took 1.8v inputs
(no higher, no lower). We had 12 voltage rails on that one AMC card
(about 2"x5", highly integrated).
We did, however, like the 74AVC4T245 for voltage translation. That
almost fits your dual-supply bill, only the AVC parts don't go up to
the full 5v. For that, your best bet would be in the LVC family;
they have some dual-rail '245 and similar equivalents (e.g. the
SN74LVC8T245). Nothing in DIP, though, so if you really wanted that
you'd have to mount them to surfboards (I really like the ones by
Bellin at http://www.beldynsys.com/; not cheap, but pretty nice
protoboards the do mount up nicely).
For level conversion, I've just used a really neat chip in two
designs, and it seems to work very well. It's the TI TXS0108E. It's a
bidirectional converter that has separate Vcc inputs for each side.
-Dave
--
Dave McGuire
New Kensington, PA
I don't think so. That's why I use 74LVC
parts for the 5V to
3.3V direction, and 74HCT for the 3.3V to 5V direction, if 5V
CMOS levels are actually needed. Usually they aren't.
If it's a 3.3v microcontroller on the other side, it's a good
reason for selecting one with 5V tolerant inputs, I guess. At
least you can use plain old HCT in both directions.
4:47 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 8 Nov 2011 at 17:34, Dave McGuire wrote:
For level conversion, I've just used a really
neat chip in two
designs, and it seems to work very well. It's the TI TXS0108E. It's
a bidirectional converter that has separate Vcc inputs for each side.
And SMT, so not user-replaceable. Maybe I should use some
CD4010s... :)
--Chuck
4:56 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/08/2011 07:47 PM, Chuck Guzis wrote:
Hey, I'm a user! ;) (they're a hell of a lot easier to desolder than
the ARM chips I'm connecting them to!)
-Dave
--
Dave McGuire
New Kensington, PA
For level
conversion, I've just used a really neat chip in two
designs, and it seems to work very well. It's the TI TXS0108E. It's
a bidirectional converter that has separate Vcc inputs for each side.
And SMT, so not user-replaceable. Maybe I should use some
CD4010s... :)
5:08 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 8 Nov 2011 at 19:56, Dave McGuire wrote:
Hey, I'm a user! ;) (they're a hell of a
lot easier to desolder
than the ARM chips I'm connecting them to!)
Yeah, but the problem is that you and I have had some practice at
this and know a few tricks. That's not all users.
(Just finished soldering a PIC32 TQFP-100 using a Weller TCP. Piece
o' cake... )
--Chuck
7:43 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
Yeah, but the problem is that you and I have had some
practice at
this and know a few tricks. That's not all users.
But the only way for people who (like me) are no good at soldering
surface-mount to get better is to practice.
I recently worked with a colleague building some circuitry. I was
surprised at how bad he was at soldering - he must have not done much
of it or something; he doesn't seem to be especially low dexterity in
most respects.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
9 Nov
9 Nov
10:13 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/08/2011 08:08 PM, Chuck Guzis wrote:
Well, true. But my logic on that is, if I can learn to do it, anyone
can.
Hey,
I'm a user! ;) (they're a hell of a lot easier to desolder
than the ARM chips I'm connecting them to!)
Yeah, but the problem is that you and I have had some practice at
this and know a few tricks. That's not all users. (Just finished soldering a PIC32 TQFP-100 using a
Weller TCP. Piece
o' cake... )
Ahh yes. Dragging?
-Dave
--
Dave McGuire
New Kensington, PA
10:39 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 1:29 PM, Chuck Guzis wrote:
On 9 Nov 2011 at 13:13, Dave McGuire wrote:
Must not be RoHS, then. That stuff is awful; it doesn't wick and it always looks like
a cold solder joint.
Anyone had any experience doing BGAs with hot air jets? The tutorials on SparkFun make it
look at least doable, and the machines are less than $200. Very tempting.
- Dave
Ahh yes. Dragging?
Yup, with the broadest chisel tip that I have. Cleanup with a bit of
solder wick. Works like a charm.
10:56 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/09/2011 01:39 PM, David Riley wrote:
Must not be RoHS, then. That stuff is awful; it doesn't wick and it
always looks like a cold solder joint.
I'm right there with you; I hate that stuff. I have a nice stock of
REAL solder that should last me a good long time.
Ahh yes. Dragging?
Yup, with the broadest chisel tip that I have. Cleanup with a bit
of solder wick. Works like a charm. Anyone had any experience doing BGAs with hot air
jets? The
tutorials on SparkFun make it look at least doable, and the machines
are less than $200. Very tempting.
I haven't tried it, but I'm bound to before too long. I have a hot
air machine here. I also have a small X-ray machine for post-soldering
inspection of the inner balls, but haven't quite worked out the imaging
end of it.
-Dave
--
Dave McGuire
New Kensington, PA
11:31 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 1:56 PM, Dave McGuire wrote:
I imagine acquiring the machine is probably the harder part of that endeavour. :-)
- Dave
Anyone had any
experience doing BGAs with hot air jets? The
tutorials on SparkFun make it look at least doable, and the machines
are less than $200. Very tempting.
I haven't tried it, but I'm bound to before too long. I have a hot air machine
here. I also have a small X-ray machine for post-soldering inspection of the inner balls,
but haven't quite worked out the imaging end of it.
12:27 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/09/2011 02:31 PM, David Riley wrote:
I imagine acquiring the machine is probably the harder part of that endeavour. :-)
Not if you have a friend who repairs dental office equipment for a
living. =)
-Dave
--
Dave McGuire
New Kensington, PA
Anyone had any experience doing BGAs with hot air
jets? The
tutorials on SparkFun make it look at least doable, and the machines
are less than $200. Very tempting.
I haven't tried it, but I'm bound to before too long. I have a hot air machine
here. I also have a small X-ray machine for post-soldering inspection of the inner balls,
but haven't quite worked out the imaging end of it.
11:05 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 9 Nov 2011 at 13:39, David Riley wrote:
Must not be RoHS, then. That stuff is awful; it
doesn't wick and it
always looks like a cold solder joint.
I like lead and have stockpiled enough lead-bearing solder to last
the rest of my days on this vale of tears. I'll keep using it until
the envrio-cops break down my door and confiscate it.
Given the level of electronics recycling today and the reclaimability
of tin-lead solder, I think the RoHS measure was far too premature.
Even worse, the US appears to be signing onto the program...
--Chuck
11:57 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 2:05 PM, Chuck Guzis wrote:
On 9 Nov 2011 at 13:39, David Riley wrote:
Agreed. We only worked with RoHS because so many of our customers have to see
internationally. What RoHS mostly targets, I thought, was the folks who just toss broken
or outdated electronics in the trash (which, let's face it, is probably most of the
people we know because no one has ever told them otherwise).
What worries me more, actually, is the mandate to get rid of incandescent bulbs before a
suitable alternative that's not pumped full of mercury gas is available. I like CFLs
as a light source (well, some of them, anyway), but I've yet to meet anyone else who
doesn't just throw them in the bin when they're dead.
- Dave
Must not be RoHS, then. That stuff is awful; it
doesn't wick and it
always looks like a cold solder joint.
I like lead and have stockpiled enough lead-bearing solder to last
the rest of my days on this vale of tears. I'll keep using it until
the envrio-cops break down my door and confiscate it.
Given the level of electronics recycling today and the reclaimability
of tin-lead solder, I think the RoHS measure was far too premature.
Even worse, the US appears to be signing onto the program...
12:57 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
What worries me more, actually, is the mandate to get
rid of
incandescent bulbs before a suitable alternative that's not pumped full
I don't beleive there is a suitable alternatic for _all_ applciations. I
liek CFLs, I was usign them long before they were commonplace. But there
are at least 4 locations where I use good old filament lamps :
1) Over my lathe (I am worried about flicker, perhaps unnecessarily)
2) In my copyling stand (You can't get a continouls spectrum from any CFL
that I've seen, so using them a a light source for colour photogrpahy is
a non-staeter)
3) In the white light source in the darkroom (CFLs have a long afterglow
that is easily bright enough to fog film. Iwant to turn of the ligth and
be able to handle light sensitive materials)
4) In the balaset unit I sue when repairing SMPSUs (you don't honestly
think that CFLs have the right V-I characteristic for this, do you?)
of mercury gas is available. I like CFLs as a light
source (well, some
And yet they mad the use fo mercury in all sorts of other things like
barometers, thermometers, switches, etc. ARGH!
of them, anyway), but I've yet to meet anyone else
who doesn't just
throw them in the bin when they're dead.
You mean you dont' tklae them apart and estract any still-good components
from the base?
-tony
1:39 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 3:57 PM, Tony Duell wrote:
Well, I'd consider it. Most of my family and friends just toss them in the garbage,
along with dead alkaline cells and all sorts of other things that probably shouldn't
get very near our water supply.
- Dave
What worries
me more, actually, is the mandate to get rid of
incandescent bulbs before a suitable alternative that's not pumped full
of mercury gas is available. I like CFLs as a light source (well, some
of them, anyway), but I've yet to meet anyone else who doesn't just
throw them in the bin when they're dead.
You mean you dont' tklae them apart and estract any still-good components
from the base? 
5:02 p.m.
New subject: Proper disposal of CFLs - Re: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 09/11/11 4:39 PM, David Riley wrote:
On Nov 9, 2011, at 3:57 PM, Tony Duell wrote:
Well, I'd consider it. Most of my family and friends just toss them in the garbage,
along with dead alkaline cells and all sorts of other things that probably shouldn't
get very near our water supply.
- Dave
Then a hefty (refundable) disposal deposit should be built into the price.
--T
What
worries me more, actually, is the mandate to get rid of
incandescent bulbs before a suitable alternative that's not pumped full
of mercury gas is available. I like CFLs as a light source (well, some
of them, anyway), but I've yet to meet anyone else who doesn't just
throw them in the bin when they're dead.
You mean you dont' tklae them apart and estract any still-good components
from the base?
10 Nov
10 Nov
11:53 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
[Disposing of dead CFLs]
Well, I'd consider it. Most of my family and
friends just toss them
in the garbage, along with dead alkaline cells and all sorts of other
things that probably shouldn't get very near our water supply.
Yes, I ahve to admit it's common to do that over here too, for a variety
of reasons :
1) The polution involved in taking them to a recycling centre is possibly
more than the polution caused by dumping the device in the landfill (even
if you walk or cycle, you will still exhale more CO_2 doing that than if
you were sitting arround, don't give me the nonsense about those methods
of transprt causing 0 polution).
2) We pay our taxes to have our household rubbish collected...
3) Some, if not most, recycling seems to be a big con. Some areas of
London now have public litter bins with one side for recyclable waste
(paper, cans, bottles, etc) the other side for other rubbish. I have seen
such bins being empteid -- the containes from the 2 sides are emptied
into the same part of the same dustcart. So people now tend not to waster
their time soritng their litter.
-tony
3:28 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/10/2011 12:53 PM, Tony Duell wrote:
We pay our taxes to have our household rubbish collected...
3) Some, if not most, recycling seems to be a big con. Some areas of
London now have public litter bins with one side for recyclable waste
(paper, cans, bottles, etc) the other side for other rubbish. I have seen
such bins being empteid -- the containes from the 2 sides are emptied
into the same part of the same dustcart. So people now tend not to waster
their time soritng their litter.
After watching the 1st season of Dr Who, it is not even safe to put out the
trash over there. :)
-tony
3:51 p.m.
New subject: City recycling - Re: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 10/11/11 2:53 PM, Tony Duell wrote:
[Disposing of dead CFLs]
This implies you should vote be voting for the candidate who will set up
a household recycling service (we have one in Toronto, but it has been
threatened by the latest lumpen-Mayor).
If neither candidate offers the policy you want -- well, isn't that an
interesting problem...
--T
Well, I'd consider it. Most of my family and
friends just toss them
in the garbage, along with dead alkaline cells and all sorts of other
things that probably shouldn't get very near our water supply.
Yes, I ahve to admit it's common to do that over here too, for a variety
of reasons :
1) The polution involved in taking them to a recycling centre is possibly
more than the polution caused by dumping the device in the landfill (even
if you walk or cycle, you will still exhale more CO_2 doing that than if
you were sitting arround, don't give me the nonsense about those methods
of transprt causing 0 polution).
2) We pay our taxes to have our household rubbish collected...
3) Some, if not most, recycling seems to be a big con.
Some areas of
London now have public litter bins with one side for recyclable waste
(paper, cans, bottles, etc) the other side for other rubbish. I have seen
such bins being empteid -- the containes from the 2 sides are emptied
into the same part of the same dustcart. So people now tend not to waster
their time soritng their litter.
-tony
11 Nov
11 Nov
11:57 a.m.
New subject: City recycling - Re: 3.3V/5V level shifting (was Re: 4000 Series
This implies you should vote be voting for the
candidate who will set up
a household recycling service (we have one in Toronto, but it has been
threatened by the latest lumpen-Mayor).
We get some such recycling over here, in that paper/card/glass/plastci
bottles are collected separately, etc. But there is no special collection
for electrical stuff/CFLs.
If neither candidate offers the policy you want -- well, isn't that an
interesting problem...
We have many more than 2 candidates to choose between (3 main political
parties, several minor parties, and independant candidates).
-tony
9 Nov
9 Nov
8:10 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
I don't beleive there is a suitable alternatic [to
incandescents] for
_all_ applciations.
Of course not.
1) Over my lathe (I am worried about flicker, perhaps
unnecessarily)
Probably. I'm reasonably sure modern fluorescents, including CFLs,
flicker at multiple KHz, not at 50/60 or 100/120 Hz.
2) In my copyling stand (You can't get a
continouls spectrum from any
CFL that I've seen, so using them a a light source for colour
photogrpahy is a non-staeter)
Well, possibly excepting some special applications, yeah.
3) In the white light source in the darkroom (CFLs
have a long
afterglow [...]).
Just recently I got some white LEDs. They're supposedly
high-brightness, and they aren't kidding; I (conservatively) hooked it
up to 5V with a 1K resistor and got a "that's _bright_" light. Then I
masured and found it was pulling about 2.2mA - and it's rated for some
ten times that. Running it at about 21mA, it was on the bench pointing
up, and in the middle of the day there was a clear area of brightness
visible on the ceiling, some six feet above the bench. This wasn't
full direct sunlight, but it wasn't a cavern, either; holy crap that
little bugger is bright. It was burning about 58 mW and probably
turning almost all of it into light.
Its spectrum was pretty close to continuous, based on a DVD used as a
diffraction grating. It topped out in the high blue, a little bit
before the visible violet fades out into the long ultraviolet, but
aside from that it looked about as good as possible given how crude the
measurement instrument was.
4) In the balaset unit I sue when repairing SMPSUs
(you don't
honestly think that CFLs have the right V-I characteristic for this,
do you?)
I have a stash of incandescents which I'm keeping specifically for uses
where their electrical characteristics are important, such as that.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
10 Nov
10 Nov
12:14 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
I don't beleive there is a suitable
alternatic [to incandescents] for
_all_ applciations.
Of course not.
1) Over my lathe (I am worried about flicker,
perhaps unnecessarily)
Probably. I'm reasonably sure modern fluorescents, including CFLs,
flicker at multiple KHz, not at 50/60 or 100/120 Hz.
2) In my copyling stand (You can't get a
continouls spectrum from any
CFL that I've seen, so using them a a light source for colour
photogrpahy is a non-staeter)
Well, possibly excepting some special applications, yeah. 3) In the white light source in the darkroom
(CFLs have a long
afterglow [...]).
Just recently I got some white LEDs. They're supposedly
high-brightness, and they aren't kidding; I (conservatively) hooked it
up to 5V with a 1K resistor and got a "that's _bright_" light. Then I
from the local 'pound store' (said light
contaiend 24 such LEDs, I've
never seen them at anything like that price in an
electronics catalogue).
Anyaay, I took one of the LEDs, touche the appropraite wire on one side
of a 9V battery and completed the circuit by holding the other ELD wire
in one hand touching the other battery terminal with my other hand. In
other words my had-to-had resistance was the LED series reisotr. The
result was a definite glow from the LED. Not bright, but certainly visible.
[1] This was for a classic-computer applciation. I'd rewound the motors
in an HP9125A plotter (for the 9100 calculator) and wanted to check the
voltage-seped characteristics.
Its spectrum was pretty close to continuous, based on
a DVD used as a
Interesting. That might negate what I said above. I am going to have to
try thenm as a photographic light source.
-tony
2:01 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Thu, 10 Nov 2011 20:14:12 +0000 (GMT)
ard at p850ug1.demon.co.uk (Tony Duell) wrote:
Since 'white' LEDs are phosphor-bnased too,
they presumably have a
strange spectrum too, and are not suitable as a light source for colour
(film) photography.
Depends on the LED. LED lights for macro photography are quite
common
these days.
--
\end{Jochen}
\ref{http://www.unixag-kl.fh-kl.de/~jkunz/}
11 Nov
11 Nov
11:53 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
Depends on the LED. LED lights for macro photography
are quite common
these days.
Yes, but a lot of thsoe are used with digital cameras, and I suspect the
response to a non-continous spectrum of a filtered CCD array is differnt
from that of colour film.
-tony
1:28 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Fri, 11 Nov 2011 19:53:08 +0000 (GMT)
ard at p850ug1.demon.co.uk (Tony Duell) wrote:
I suspect the response to a non-continous spectrum of
a filtered
CCD array is differnt from that of colour film.
When it comes to a non-continous
spectrum, even the response of
different color films is different. The difference from a CCD to a film
isn't that much different as from one color film to an other. Especially
if you compare so different things as negative film and slide film or
E6 / C41 to the now extinct Kodachrome...
Film and CCDs (or CMOS sensors, basicly a light sensitive DRAM) use the
same principle of spliting up the color space into read, green and
blue. If there is a narrow spectral line inbetween (e.g. sodium lamps
for street lighting) booth will get the same problems.
Ahh, I forgot video camera tubes. They are bitten by the same problem...
--
\end{Jochen}
\ref{http://www.unixag-kl.fh-kl.de/~jkunz/}
9 Nov
9 Nov
1:02 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Wed, Nov 9, 2011 at 2:57 PM, David Riley <fraveydank at gmail.com> wrote:
Agreed. We only worked with RoHS because so many of our customers have to
see internationally. What RoHS mostly targets, I thought, was the folks
who just toss broken or outdated electronics in the trash (which, let's
face it, is probably most of the people we know because no one has ever
told them otherwise).
also, Telecom, Military and Medical hardware are exempted from RoHS. As
such, we never had to follow RoHS in our Switching Systems, I'm sure you've
heard - Lead-Free is unreliable for Hi-Rel Systems due to shorts created by
crystalline growths - ie. tin whiskers. Lead was initially added in the
early part of the 20th century to mitigate this known problem.
Unfortunately, I just read that Servers and Network systems have recently
been removed from this exemption since 2010 - oh great.
Consumer items are made like junk anyway - they're meant to be thrown away
:) But now we have this new recycling law which you are forced to pay a
recycler to dispose of consumer items. The trashman is no longer allowed to
haul away your junk. Now think what will really happens to the junk, spewn
in some remote area of the woods
What worries me more, actually, is the mandate to get rid of incandescent
bulbs before a suitable alternative that's not pumped full of mercury gas
is available. I like CFLs as a light source (well, some of them, anyway),
but I've yet to meet anyone else who doesn't just throw them in the bin
when they're dead.
LED lamps are getting cheaper
10 Nov
10 Nov
11:47 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Wed, Nov 9, 2011 at 2:57 PM, David Riley <fraveydank at gmail.com> wrote:
In the UK, at least, it appers the RoHX regulations only apply when the
device is offered for sale, which means that prototypes and
home-constructed units are excempt [It's crazxy IMHO that Maplin, who
sell mainly to hobbyists only sell lead-free solder now...]. And you can
sue leaded solder for repairs on things that were oriignally made with
leaded solder (like classic computers).
This is one reason I much prefer things to come as solder-yourself kits.
I can then use whatever solder I choose.
SOme connectors, particuallrly Mil-spec types seem to only exist in
non-RoHS versions (this may be cadmium plating, rather than lead). This
has the curuous feater that while I, or anyone else, can by the
conenctors to connect to an HP9830 pirnter port, or the input of an
HP9866 pritner, and while it is perfectly legal for anyone make up cables
and interaces for their own use, it is illegal for me to sell such cables
(or even kits to make such cables I think). Go figure.
Agreed. We only worked with RoHS because so many of our customers have to
see internationally. What RoHS mostly targets, I thought, was the folks
who just toss broken or outdated electronics in the trash (which, let's
face it, is probably most of the people we know because no one has ever
told them otherwise).
also, Telecom, Military and Medical hardware are exempted from RoHS. As such, we never had to follow RoHS in our Switching
Systems, I'm sure you've
heard - Lead-Free is unreliable for Hi-Rel Systems due to shorts created by
crystalline growths - ie. tin whiskers. Lead was initially added in the
early part of the 20th century to mitigate this known problem.
Indeed.
Unfortunately, I just read that Servers and Network
systems have recently
been removed from this exemption since 2010 - oh great.
Consumer items are made like junk anyway - they're meant to be thrown away
This doens't mean you _have_ to throw them away. You know me well enough
by now to know that I'll have a go at repairing them..
:) But now we have this new recycling law which you
are forced to pay a
recycler to dispose of consumer items. The trashman is no longer allowed to
haul away your junk. Now think what will really happens to the junk, spewn
in some remote area of the woods
In other words, these regulatios have almost certainly increased
polution. As usual...
-tony
11 Nov
11 Nov
5:37 a.m.
New subject: Disposal of CFL's - was: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
--- On Thu, 11/10/11, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
Yes, but of course. This is human nature. There is a similar problem with old tires around
here. The garbage man won't take them, and disposal costs a few dollars. Therefore,
it's quite common to see them discarded into woods, tossed into the median on
highways, or in the river.
An acquaintance of mine, a couple years ago, faced with problems disposing of a massive
pile of Compaq Deskpro Pentium 1's, was not about to pay the ten bucks apiece the
"recycling" company wanted. The trash men where he lived wouldn't take them,
and nobody else wanted them. Therefore, there are now three Ford Taurus loads of old
PC's at the bottom of the river, deposited there under cover of darkness.
While I don't approve of his methods, I can understand the problem. Disposal of some
things is difficult. Old refrigerators, tires, computers, batteries - they have to go
somewhere. And nobody wants to pay to get rid of something. I think it's getting
better, but it still has a long way to go. And unfortunately, these new CFL's and
other "greeny" devices aren't helping. For years we have thrown our burned
out incandescent light bulbs in the trash. They're harmless and inert - glass and
metal. But CFL's contain mercury and other chemicals. Probably not much of them, so
it's likely not as big an issue now... but it could be in the future. In my
experience, CFL's really don't last too long - about the same as a normal
incandescent - and when they do go, they tend to burn and melt the base. Nobody is going
to bother hanging onto them, waiting for some kind of option for recycling. They go in the
trash.
Of course, the real problem is that we have a society that produces massive amounts of
electronics with a very short usable life. Rather than addressing the symptoms, they
really need to focus on the cause - and produce devices that don't *need* to be
replaced every six months. With every step towards being "green", we take two
steps back, by creating more and more broken or unusable stuff. RoHS regulations have
introduced unreliable lead-free solder, and that's introduced massive piles of junk
laptop computers and game consoles into the garbage. I can't count the number of
machines I've seen rendered irreparable due to faulty solder joints on large BGA
chips. They're starting to get the hang of it, and newer machines seem to be slightly
better, but still. It's all politics, and very little of it actually has anything to
do with realistically being efficient or polluting less.
Another side effect.. as they remove more and more materials from the computers, they get
less reliable. They ALSO get less attractive to recycle, because there isn't enough
material to extract to make it worthwhile.
-Ian
:) But now we
have this new recycling law which you
are forced to pay a
recycler to dispose of consumer items. The
trashman is
no longer allowed to
haul away your junk. Now think what will really
happens to the junk, spewn
in some remote area of the woods
In other words, these regulatios have almost certainly
increased
polution. As usual...
6:34 a.m.
New subject: Disposal of CFL's - was: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 11, 2011 8:42 AM, "Mr Ian Primus" <ian_primus at yahoo.com> wrote:
--- On Thu, 11/10/11, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
Yes, but of course. This is human nature. There is a similar problem with
old tires
around here. The garbage man won't take them, and disposal costs
a few dollars. Therefore, it's quite common to see them discarded into
woods, tossed into the median on highways, or in the river.
:) But
now we have this new recycling law which you
are forced to pay a
recycler to dispose of consumer items. The
trashman is
no longer allowed to
haul away your junk. Now think what will really
happens to the junk, spewn
in some remote area of the woods
In other words, these regulatios have almost certainly
increased
polution. As usual...
An acquaintance of mine, a couple years ago, faced with problems
disposing of a
massive pile of Compaq Deskpro Pentium 1's, was not about to
pay the ten bucks apiece the "recycling" company wanted. The trash men
where he lived wouldn't take them, and nobody else wanted them. Therefore,
there are now three Ford Taurus loads of old PC's at the bottom of the
river, deposited there under cover of darkness.
This law is another example of corporate slight of hand trick. There was a
time when you get a deposit return on most items, from sodapop bottle to
tires. Now these recyclers make their money coming in and on the items they
refurbished that's going out. You and I are paying for their stock
inventory to produce their products - what kind of logic is that. Now there
is still the local recyclers centers in several cities which are free, but
you have to motivate yourself - which unfortunately most people don't -to
bring the items which are sorted right there into separate recycle bins.
And we already have recycle trucks in many cities that pickup glass/plastic
bottles - so why not electronics ?
=Dan
9 Nov
9 Nov
2:50 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 9 Nov 2011 at 14:57, David Riley wrote:
Agreed. We only worked with RoHS because so many of
our customers
have to see internationally. What RoHS mostly targets, I thought, was
the folks who just toss broken or outdated electronics in the trash
(which, let's face it, is probably most of the people we know because
no one has ever told them otherwise).
By the time it's ready to be permanently retired, I'll probably have
re-soldered every joint in my 90's Volvo "Turbobrick". First it was
the relays (bad joints on the small PCBs inside of them), then the
radio, then the dashboard instrument PCB. All nicely lead-free and
given to turning any joint into an open or intermittent circuit.
On the Turbobrick forum, there are a few guys who resolder all of the
relay PCBs when they acquire a car.
What worries me more, actually, is the mandate to get
rid of
incandescent bulbs before a suitable alternative that's not pumped
full of mercury gas is available. I like CFLs as a light source
(well, some of them, anyway), but I've yet to meet anyone else who
doesn't just throw them in the bin when they're dead.
White LED lamps seem to have about the same phosphor lifetime as
regular fluorescent lamps. I have some LED nightlights in my halls
and I've observed that the output is slowly declining as they age.
Given the initial cost of LED incandescent replacements, I'm a bit
skeptical about the useful life claims being made.
Around here, a lot of the big box stores are offering a credit
(usually $5) toward exchanging a string of incandescent Christmas
tree lights for the LED variety. Horrible, flickering harsh things.
--Chuck
8:13 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
White LED lamps seem to have about the same phosphor
lifetime as
regular fluorescent lamps.
You mean they're not white LEDs, but narrow-spectrum LEDs exciting a
phosphor such as is used in a fluorescent tube?
I was wondering how they got the flat spectrum....
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
8:37 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 11:13 PM, Mouse wrote:
Yup, that's the genius behind them. They're basically fluorescent lamps using an
LED as the UV source instead of mercury vapor. Though I'm told there are additional
tricks they employ to even out the spectrum now as well. Never bothered to look into it.
LEDs in general tend to emit an astonishingly narrow bandwidth, so getting white out of
them is quite an achievement.
- Dave
White LED
lamps seem to have about the same phosphor lifetime as
regular fluorescent lamps.
You mean they're not white LEDs, but narrow-spectrum LEDs exciting a
phosphor such as is used in a fluorescent tube?
I was wondering how they got the flat spectrum....
10 Nov
10 Nov
11:49 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Wed, 9 Nov 2011 23:37:18 -0500
David Riley <fraveydank at gmail.com> wrote:
[whilte LEDs]
They're basically fluorescent lamps using an LED
as the UV
source instead of mercury vapor.
Not exactely. Whilte LEDs are based on blue LEDs
and a fluorescent that
converts part of this blue light to longer wavelengts. I sugest you go
to the site of a major LED manufacturer like Cree or Nichia, donload
some data sheets and look at the spectrums. Usualy the spectrum has a
peek at around 450 nm, but not that narrow like a blue LED without
fluorescent. Often there is a gap at around 500 nm and then a broad
"hill" at 550 nm to 600 nm depending on color temperature. So the
spectrum is quite smoth.
Most whilte LEDs have a Color Rendering Index of 70%..80%, where
sun light is 100% and incandescent lamps are close to this. There are
high CRI LEDs with a CRI of around 95%. One of those is lightening my
living room at this very moment. It needs 14 W and produces as much
light as a 60 W incandescent lamp. It has a color temperature of 5000 K,
that fits to the 5500 K calibration of computer screens much better
then incandescent lamps. (2700 K or 3200 K for halogen)
Incandescent replacements based on LEDs are a poor choice. Incandescent
lamps work better the hotter they get. So the fixtures for incandescent
lamps are made in a way that isolates as much heat as possible. This is
absolutely contrary to the needs of LEDs. LEDs need to stay cool. They
need a loot of cooling to keep the junction temperature in a reasonible
range. They decay fast when hot and the efficieny drops. Unfortunately
real LED lapms, that are constructed according to the needs of LEDs,
are rare. Thats why I build LED lamps myself, as I am the geek, that I
am. ;-)
Ahh, low to mid power, high efficient LEDs can produce about 10 times
more light then an incandescent lamp! Though, most high power LEDs with
acceptable to good CRI are around factor 4 to 5. That is about the
same efficiency as CFLs.
There are UV LEDs. But they are not as efficient as blue LEDs and they
decay quite fast due to the UV light they produce. Therefore they are
not used for white LEDs.
--
\end{Jochen}
\ref{http://www.unixag-kl.fh-kl.de/~jkunz/}
1:20 p.m.
New subject: White LEDs (was: Re: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?))
On 10 Nov 2011, at 11:49 AM, Jochen Kunz wrote:
Not exactely. Whilte LEDs are based on blue LEDs and a
fluorescent that
converts part of this blue light to longer wavelengts. I sugest you go
to the site of a major LED manufacturer like Cree or Nichia, donload
some data sheets and look at the spectrums. Usualy the spectrum has a
peek at around 450 nm, but not that narrow like a blue LED without
fluorescent. Often there is a gap at around 500 nm and then a broad
"hill" at 550 nm to 600 nm depending on color temperature. So the
spectrum is quite smoth.
I wouldn't be surprised to find that individual white emitters function in that
fashion.
Most whilte LEDs have a Color Rendering Index of 70%..80%, where
sun light is 100% and incandescent lamps are close to this. There are
high CRI LEDs with a CRI of around 95%. One of those is lightening my
living room at this very moment. It needs 14 W and produces as much
light as a 60 W incandescent lamp. It has a color temperature of 5000 K,
that fits to the 5500 K calibration of computer screens much better
then incandescent lamps. (2700 K or 3200 K for halogen)
I have a bunch (on the order of 100) CREE LR6 modules installed which have a 92+ CRI (not
that CRI means much). The visible spectrum seems quite flat and I've been very
pleased with the results with color film.
I'd sort of expected that these things worked as suggested -- by irradiating a
phosphor screen -- but I had one arrive damaged in shipment that put an end to that
notion. What I thought was a phosphor screen was just a diffusor lens; the guts of the
thing was a mix of colored SMD LEDs mixed with white XR-Es, so for (these admittedly
bulky) things they seem to have smoothed the spectrum by playing with the emitter mix.
Incandescent replacements based on LEDs are a poor choice. Incandescent
lamps work better the hotter they get. So the fixtures for incandescent
lamps are made in a way that isolates as much heat as possible. This is
absolutely contrary to the needs of LEDs. LEDs need to stay cool. They
need a loot of cooling to keep the junction temperature in a reasonible
range. They decay fast when hot and the efficieny drops. Unfortunately
real LED lapms, that are constructed according to the needs of LEDs,
are rare. Thats why I build LED lamps myself, as I am the geek, that I
am. ;-)
The LR6 modules are huge, bulky things that are a fairly tight fit into 6" IC cans
with most of the bulk being heat sink. Despite the 50K hour lifetime on the things I was
worried about temperature rise, but stuffing a thermocouple into the housing showed a
delta of only a few degrees C over ambient regardless of power setting (the things are
dimmable as well).
Bottom line is you can get LED replacements for certain classes of incandescent lighting
that has good color performance, but it's not cheap.
--
Dr. Christian Kennedy
chris at mainecoon.com AF6AP
http://www.mainecoon.com PGP KeyID 108DAB97
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
"Mr. McKittrick, after careful consideration..."
2:29 p.m.
New subject: White LEDs (was: Re: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?))
On Thu, 10 Nov 2011 13:20:13 -0800
Christian Kennedy <chris at mainecoon.com> wrote:
I'd sort of expected that these things worked as
suggested -- by irradiating a phosphor screen -- but I had one arrive damaged in shipment
that put an end to that notion. What I thought was a phosphor screen was just a diffusor
lens; the guts of the thing was a mix of colored SMD LEDs mixed with white XR-Es, so for
(these admittedly bulky) things they seem to have smoothed the spectrum by playing with
the emitter mix.
Thats a common way to get a good CRI. What you describe as
"phosphor
screen" is called "remote phosphor".
The LR6 modules are huge, bulky things
Huge and
bulky = large surface = good cooling...
Bottom line is you can get LED replacements for
certain classes of
incandescent lighting that has good color performance, but it's
not cheap.
Yes.
--
\end{Jochen}
\ref{http://www.unixag-kl.fh-kl.de/~jkunz/}
10:42 p.m.
New subject: White LEDs (was: Re: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?))
Zitat von Jochen Kunz <jkunz at unixag-kl.fh-kl.de>:
Thanks for that. Most people don't notice how hot those things get.
So cooling is important !
The LR6
modules are huge, bulky things
Huge and bulky = large surface = good cooling...
23 Nov
23 Nov
8:36 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
At 01:49 PM 11/10/2011, Jochen Kunz wrote:
Incandescent replacements based on LEDs are a poor
choice. Incandescent
lamps work better the hotter they get. So the fixtures for incandescent
lamps are made in a way that isolates as much heat as possible. This is
absolutely contrary to the needs of LEDs. LEDs need to stay cool.
A greenie friend of mine recently put an LED replacement bulb in
her garage, not far from the remote control garage door opener.
It began to malfunction in odd ways, sometimes working, mostly not.
Eventually a electrician figured out the LED bulb was causing enough
radio interference to disrupt the remote control receiver.
- John
24 Nov
24 Nov
1:02 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
Eventually a electrician figured out the LED bulb was
causing enough
radio interference to disrupt the remote control receiver.
However, I seem to remembner reading that old filament bulbs can cause TF
interference too. Some valve radios over here used a low-powered ligth
bulb as the dropping resisotr in the series heater chain [1] and when
the bulb got old, yuo'd get interference on the output.
Nothing to do with an intermitany break in the filament, IIRC the effect
was somethign to do with the evapourated metal from the filament on the
inside of the glass (which is why it only occured with old, blackened
bulbs) collecting electrons from the filament. But there was more to it
than that. I will see if I can find the articles.
For some reason the term 'Barkhausen effect' is coming to mind. But that
may be a collision ib my hash tables.
[1] One UK magazine actually published a design for a combined radio and
table lamp (!).. Yes, the lamp bulb was the heater dropper resistor.
-tony
3:40 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 24 Nov 2011 at 21:02, Tony Duell wrote:
For some reason the term 'Barkhausen effect'
is coming to mind. But
that may be a collision ib my hash tables.
Isn't that a magnetically-induced thing related to crystal
dislocations under a changing magnetic field? i.e., Barkhausen
noise.
I also think of early HF oscillators produced by placing the grid of
a triode at a positive potential relative to the cathode and the
plate at a negative potential. Called "Barkhausen" or "electron
orbit" oscillators.
Also related are oddball oscillations in tetrode HF power amplifiers,
I think.
Or it could be the Thanksgiving Day (today in the US) liquor
talking...
--Chuck
25 Nov
25 Nov
12:59 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 24 Nov 2011 at 21:02, Tony Duell wrote:
I think that's right. A simple-ish demonstration of this (which I have
done) is to take a mild steel bar, wind a couil on it and connect said
coil to the input of an audio amplifier. Then magnetise the bar in the
traditional way by stronking it with one pole of a permanent magnet. You
get the Barkhausen noise as you stroke the magnet until the bar is fully
magnetised. Of course if you tehn remagnetise it the opposit way (by
strokign it with the other pole of the magnet) you can get the noise again.
For some reason the term 'Barkhausen
effect' is coming to mind. But
that may be a collision ib my hash tables.
Isn't that a magnetically-induced thing related to crystal
dislocations under a changing magnetic field? i.e., Barkhausen
noise.
I also think of early HF oscillators produced by placing the grid of
a triode at a positive potential relative to the cathode and the
plate at a negative potential. Called "Barkhausen" or "electron
orbit" oscillators.
I wnder if that's the effect. The metal film depositied on the inside of
the glass could become negatively charged due to electronics hitting it.
Also related are oddball oscillations in tetrode HF
power amplifiers,
I think.
There's the dynatron oscilalator (nothign to do with the UK radio company
of that name AFAIK) which depends on the negative resistance part of the
Ia .vs. Va curve for a normal tetrode. The beam tetrode was sometimes
called a 'kinkles tetrode' over here, because it doesn't have this dip in
the chracteristic curve, and the 'KT' in valve numebrs like KT66 and KT88
stands for 'Kinklrss Tetrode'.Needless to say those can't be used as
dynstron oscillators.
-tony
2:16 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 25 Nov 2011 at 20:59, Tony Duell wrote:
There's the dynatron oscilalator (nothign to do
with the UK radio
company of that name AFAIK) which depends on the negative resistance
part of the Ia .vs. Va curve for a normal tetrode. The beam tetrode
was sometimes called a 'kinkles tetrode' over here, because it doesn't
have this dip in the chracteristic curve, and the 'KT' in valve
numebrs like KT66 and KT88 stands for 'Kinklrss Tetrode'.Needless to
say those can't be used as dynstron oscillators.
Found a Wiki reference to "Barkhausen-Kurz" tube:
http://en.wikipedia.org/wiki/Barkhausen-Kurz_tube
and found a mention of same in an old (pre-war) electronics text.
Apparently if you add some tuned circuits to a Barkhausen oscillator,
it's called a "Gill-Morell" oscillator. Same Barkhausen as the
Barkhausen effect, however.
Said text also mentioned that the best oscillation was obtained from
triodes with circular (versus oval or rectangular) elements, which
figures. Even has a practical circuit using type 76 triodes.
--Chuck
12:32 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
At 03:02 PM 11/24/2011, you wrote:
The symptoms drove my friend a bit crazy... the garage door opener would
work when approaching the house (before the light was on) but wouldn't
work to close the garage door once inside (and the light was on).
- John
Eventually a
electrician figured out the LED bulb was causing enough
radio interference to disrupt the remote control receiver.
However, I seem to remembner reading that old filament bulbs can cause TF
interference too.
10 Nov
10 Nov
12:18 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
You mean
they're not white LEDs, but narrow-spectrum LEDs exciting a
phosphor such as is used in a fluorescent tube?
I was wondering how they got the flat spectrum....
Yup, that's the genius behind them. They're basically fluorescent
lamps using an LED as the UV source instead of mercury vapor. Though I'm told there are additional tricks they employ
to even out the
spectrum now as well. Never bothered to look into it.
Interesting. I wonder how continuous it really is.
LEDs in general tend to emit an astonishingly narrow bandwidth, so
Which is not suprising when you consider how they work.
getting white out of them is quite an achievement.
Well, it would be if they'd managed it without cheating (using the
phosphor) :-)
-tony
noon
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 9 Nov 2011 at 14:57, David Riley wrote:
> Agreed. We only worked with RoHS because so many of our customers
> have to see internationally. What RoHS mostly targets, I thought, was
> the folks who just toss broken or outdated electronics in the trash
Well, it's a pity said regulations make life difficult for people who
actualyl want a lasting prodcut with soldered joints that don't fall
apart or grow tin wiskers.
> (which, let's face it, is probably most of
the people we know because
> no one has ever told them otherwise).
My friends are moslty like me. They repair old electronics, and if they
can't do that they extract every useful bit from the device.
White LED lamps seem to have about the same phosphor
lifetime as
Right...
regular fluorescent lamps. I have some LED
nightlights in my halls
and I've observed that the output is slowly declining as they age.
I see, so the claim that I've seen made several times that 'LEDs don't
burn out' is downright false (not that I ever thought it was true, after
all, I've replaced plenty of dead LEDs over the years...)
Around here, a lot of the big box stores are offering
a credit
(usually $5) toward exchanging a string of incandescent Christmas
tree lights for the LED variety. Horrible, flickering harsh things.
I haev used LEDs in Newtonsday decorations, but running them off a DC
supply (so they don;'t flicker). The reson was that LEDs are easy to
drive from logic circuitry, so I'v ehad Newtonsday decorations controller
by a PDP11, PERQ, HP41, HP71 and HP48 over the years.
-tony
1:41 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 10 Nov 2011 at 20:00, Tony Duell wrote:
I see, so the claim that I've seen made several
times that 'LEDs don't
burn out' is downright false (not that I ever thought it was true,
after all, I've replaced plenty of dead LEDs over the years...)
I'm sure that after hundreds of thousands of hours, a white LED will
still have a visible glow, even though visible light output will have
dimmed and color-shifted away from what it used to be.
Re: spectrum--I seem to recall reading a couple of months ago in one
of the trade rags that someone (Cree?) is putting red LEDs together
on the same fixture as their high-power white ones.
I haev used LEDs in Newtonsday decorations, but
running them off a DC
supply (so they don;'t flicker). The reson was that LEDs are easy to
drive from logic circuitry, so I'v ehad Newtonsday decorations
controller by a PDP11, PERQ, HP41, HP71 and HP48 over the years.
I've run mine off a full-wave bridge and that seems to handle the
flickering, but the light quality is still very harsh. I'm used to
the warm glow of incandescents, not the ultra-pure point-source
colors of LEDs. They're supposed to remind you of candles, after
all, not someone doing TIG welding.
--Chuck
2:09 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Thu, 10 Nov 2011, Chuck Guzis wrote:
I've run mine off a full-wave bridge and that
seems to handle the
flickering, but the light quality is still very harsh. I'm used to
the warm glow of incandescents, not the ultra-pure point-source
colors of LEDs. They're supposed to remind you of candles, after
all, not someone doing TIG welding.
Chuck, you might want to look into "warm white" LEDs. The ones I've
used
for edge lit panels appear to have a color temp right around 2400-2600K.
If memory serves a 40W incandescent bulb is right around 2400K.
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.
ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://www.scarletdme.org - Get it _today_!
Political correctness is a doctrine, fostered by a delusional, illogical
minority, and rabidly promoted by an unscrupulous mainstream media, which
holds forth the proposition that it is entirely possible to pick up a turd
by the clean end.
9 Nov
9 Nov
12:46 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 9 Nov 2011 at 13:39, David Riley wrote:
As have I. I was lucky to pick up some NOS reels of 60/40 for a bargain
price...
AFAIK the major suppliers over here still list leaded solder, they just
warn that it's not HoHS-compliant. I've not tried to buy any, but I've
bought other non-compliant parts (like Mil-Spec connectors) with no
problems.
Must not be RoHS, then. That stuff is awful; it
doesn't wick and it
always looks like a cold solder joint.
I like lead and have stockpiled enough lead-bearing solder to last Given the level of electronics recycling today and the
reclaimability
of tin-lead solder, I think the RoHS measure was far too premature.
It was claimed that hte extra energy reqired due to the higher
temperature used for laad-free solder totally negates the environmental
benefits of the latter...
Even worse, the US appears to be signing onto the
program...
My (somewhat cynical) view is that htis was broguht in at the request of
large conumer electronics companies who wnated ot use a oslder that fell
apart after a few years...
-tony
11:50 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Wed, Nov 9, 2011 at 1:39 PM, David Riley <fraveydank at gmail.com> wrote:
Anyone had any experience doing BGAs with hot air jets? The tutorials on
SparkFun make it look at least doable, and the machines are less than $200.
Very tempting.
I wouldn't trust a hot air machine with an expensive bga part. I see the
irda machines are getting cheaper - not sure as to the quality of these new
machines. But that method is much preferred. I see several repair houses
already using them - so a small assembly line is also possible.
=Dan
10:42 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/9/2011 1:29 PM, Chuck Guzis wrote:
On 9 Nov 2011 at 13:13, Dave McGuire wrote:
Do you guys use a magnifier with light for SMT work? Any
brands/recommendations? Ballpark prices? Sally Beauty supply trick?
Thanks
Keith
Ahh yes. Dragging?
Yup, with the broadest chisel tip that I have. Cleanup with a bit of
solder wick. Works like a charm.
--Chuck
11:02 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/09/2011 01:42 PM, Keith Monahan wrote:
Do you guys use a magnifier with light for SMT work?
Any
brands/recommendations? Ballpark prices? Sally Beauty supply trick?
For bigger stuff, I use one of the big magnifiers with a circular
fluorescent light around it, mounted on an articulated arm. For smaller
stuff, I use a Bausch & Lomb binocular microscope configured for 30x
magnification, with an integral focused illuminator.
-Dave
--
Dave McGuire
New Kensington, PA
11:09 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 9 Nov 2011 at 13:42, Keith Monahan wrote:
Do you guys use a magnifier with light for SMT work?
Any
brands/recommendations? Ballpark prices? Sally Beauty supply trick?
I use three things--a 10 power binocular loupe, a small jeweler's
loupe and a low-power (inspection) stereomicroscope. The latter is
really valuable.
But I've got old eyes, so don't take my recommendations seriously.
--Chuck
10:47 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/09/2011 01:29 PM, Chuck Guzis wrote:
Sounds good. Liquid flux? (I use a flux pen)
-Dave
--
Dave McGuire
New Kensington, PA
Ahh yes.
Dragging?
Yup, with the broadest chisel tip that I have. Cleanup with a bit of
solder wick. Works like a charm.
11:27 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 9 Nov 2011 at 13:47, Dave McGuire wrote:
Sounds good. Liquid flux? (I use a flux pen)
Heh. Bass rosin disssolved in ethanol. Allows me to use those
little chunks when they get too small to stroke a bow with.
--Chuck
11:17 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Wed, Nov 9, 2011 at 1:29 PM, Chuck Guzis <cclist at sydex.com> wrote:
On 9 Nov 2011 at 13:13, Dave McGuire wrote:
I worked with the old VGA chipsets with 208 qfp packages when they were the
rage. I would pre-coat the pads with some no clean flux solder, not much,
just enough to make a connection. Then squirt some liquid flux across the
pads - using the bottle w/needle applicator. Position the chip, using the
corner pins as alignment - and tacking them with the iron. And then I would
lightly press on the rest of the pins with a blade tip, about 4/5 pins at a
time. Never any solder bridges or solder-wick cleanup needed.
=Dan
Ahh yes. Dragging?
Yup, with the broadest chisel tip that I have. Cleanup with a bit of
solder wick. Works like a charm.
11:57 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 8 Nov 2011 at 17:34, Dave McGuire wrote:
Wasn't the 4010 superceded by the 4050 (and the 4009, the inverting
version, by the 4049)? I've used them as level shifters from 12V CMOS to TTL.
And since when haev SMD parts not been user-replaceable? Well, OK, if I
am replacing an chip (SMD or otherwise), presumably I am not able to use
the product at the same time. But other than that I don't see the problem.
You need to get a better class of users :-)
-tony
For level conversion, I've just used a
really neat chip in two
designs, and it seems to work very well. It's the TI TXS0108E. It's
a bidirectional converter that has separate Vcc inputs for each side.
And SMT, so not user-replaceable. Maybe I should use some
CD4010s... :)
7:25 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/8/2011 5:34 PM, Dave McGuire wrote:
For level conversion, I've just used a really neat
chip in two designs,
and it seems to work very well. It's the TI TXS0108E. It's a
bidirectional converter that has separate Vcc inputs for each side.
I'm anxious to build a PCB that is based around a pair of these
http://www.ti.com/product/sn74cbtd3384
10-Bit FET Bus Switches With Level Shifting
for a level conversion app.
I sampled some, although there was only the .635 pitch 24 SSOP/QSOP
version available.
Keith
8:08 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 09/11/2011, at 11:25 PM, Keith Monahan wrote:
On 11/8/2011 5:34 PM, Dave McGuire wrote:
I vaguely remember looking at that series. They're 5V tolerant, but won't output
5V from 3.3V, right?
Scott.
For level conversion, I've just used a really
neat chip in two designs,
and it seems to work very well. It's the TI TXS0108E. It's a
bidirectional converter that has separate Vcc inputs for each side.
I'm anxious to build a PCB that is based around a pair of these
http://www.ti.com/product/sn74cbtd3384
10-Bit FET Bus Switches With Level Shifting
for a level conversion app.
I sampled some, although there was only the .635 pitch 24 SSOP/QSOP version available.
8:35 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 11:08 AM, Scott Kevill wrote:
It's just a bunch of FETs. There's a good app note by TI on using them to convert
both directions, actually. IIRC it uses an open-collector arrangement of sorts, so
it's not super-fast (though that depends on what your metric for fast is), but
it's great for I2C level conversion. Not bad for other stuff, either.
- Dave
I'm
anxious to build a PCB that is based around a pair of these
http://www.ti.com/product/sn74cbtd3384
10-Bit FET Bus Switches With Level Shifting
for a level conversion app.
I sampled some, although there was only the .635 pitch 24 SSOP/QSOP version available.
I vaguely remember looking at that series. They're 5V tolerant, but won't output
5V from 3.3V, right?
10:34 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/9/2011 11:35 AM, David Riley wrote:
IIRC it uses an
open-collector arrangement of sorts, so it's not super-fast (though
that depends on what your metric for fast is), but it's great for I2C
level conversion. Not bad for other stuff, either.
Dave,
Once they are enabled (which can take max 7ns), the normal propogation
delay appears to be 250ps max. Doesn't sound too slow for most of my
applications.
The voltage drop is mosfet switch drop(5v-4v) + internal diode(4v to
3.3v) to do 5v to 3.3v. (I barely know what I'm talking about here)
@Scott: Right, the 3.3v passes through untranslated. Based on my limited
understanding, this should be ok. With Vih of 5v TTL being 2v, you're
good there. Over the Vt of 5v CMOS, but just under Vih.
This app note explains it well.
http://www.ti.com/lit/an/scda003b/scda003b.pdf
Thanks,
Keith
9:09 p.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 10/11/2011, at 12:35 AM, David Riley wrote:
On Nov 9, 2011, at 11:08 AM, Scott Kevill wrote:
Thanks, I'm still very new to the electronics side (I'm a software guy).
I have a Spartan 6 FGPA dev board that has 48 5V-tolerant I/Os (via CBT3244ABQ bus
switches) and 20 non-5V-tolerant I/Os. I'm trying to figure out the easiest way to
create a general daughterboard that would allow me to test/play-with a bunch of different
DIP ICs (eg. microcontrollers, CPUs, ROMs, etc.) with minimal manual rewiring for each IC
(remapping with constants in the HDL instead).
All the I/Os output 3.3V, which is fine for most of the signals I need, but some ICs (eg.
Z80+DMA+PIO+SIO, 8021) need higher. Plus being able to output 5V would mean it could
provide pin-configurable Vcc for most of the other ICs.
Making the requirements more tricky, some of the ICs require other supply voltages outside
that range. (eg. FD1771 needs -5V & +12V; FD1793 needs +12V; 8021 needs +15V to dump;
M57630 ROM needs -5V & +12V). I don't need to drive these oddball voltages from
FPGA I/Os, so I don't mind wiring those manually.
It would be nice if I could also use this to play with my Votrax SC-01A, but that's
even weirder with minimum input voltages of more like +6 to +8V (according to the SC-01
datasheet, assuming the SC-01A is similar). So I'd ignore this one and use a separate
circuit for it.
So..
- a huge 64-pos (MC68000 and some microcontrollers) universal ZIF socket
- alternatively a 48-pos universal ZIF and a regular 64-pos socket
- a way to output 5V from all 64 I/Os
- additional 5V tolerance for 16 of the non-5V-tolerant FPGA I/Os (48 are already 5V
tolerant)
- either additional -5V to +15V tolerance for all I/Os or a way to manually isolate
per-I/O from the manually wired supply voltages above (eg. SPDT dip switches?) The FPGA
I/Os do support tri-stating, but I'm not sure that would be safe.
- two rows of 32 header pin sockets? to allow a flying lead to supply any IC pin
- highest signal rate needed would probably be 8MHz for the MC68000, although hmm.. if I
were sampling or measuring IC timing, perhaps higher for input signals (eg. 32 MHz)
This does leave 4 unused I/Os that may or may not be helpful.
I would be creating a PCB for this, but through-hole would be preferred as I have no
experience (yet) with SMT.
Thoughts? Suggestions? Or is this crazy/na?ve? :)
Scott.
I vaguely remember looking at that series.
They're 5V tolerant, but won't output 5V from 3.3V, right?
It's just a bunch of FETs. There's a good app note by TI on using them to
convert both directions, actually. IIRC it uses an open-collector arrangement of sorts,
so it's not super-fast (though that depends on what your metric for fast is), but
it's great for I2C level conversion. Not bad for other stuff, either.
10:46 p.m.
New subject: How much a brazilian mac clone is valued?
This is just a curiosity, since I'm not willing to sell my gem
But how much can be considered "fair value" for this (very very very
rare) Brazilian mac clone?
http://www.tabalabs.com.br/apple/mac_512/index.htm
Thanks
Alexandre Souza
(no, I'm not selling it)
---
Enviado do meu Motorola PT550
Meu site: http://www.tabalabs.com.br
11:40 p.m.
New subject: How much a brazilian mac clone is valued?
No picture of the front?
/P
10 Nov
10 Nov
12:53 a.m.
New subject: How much a brazilian mac clone is valued?
Strange, there should be one...
Some here: http://chester.blog.br/mac512.html
Mine is the one-drive type. This is the two-drives. I'll update the site
as soon as possible
---
Enviado do meu Motorola PT550
Meu site: http://www.tabalabs.com.br
----- Original Message -----
Subject: Re: How much a brazilian mac clone is valued?
No picture of the front?
/P
6:27 a.m.
New subject: How much a brazilian mac clone is valued?
I vaguely remember reading somewhere that Brazil had really odd rules
concerning computers for a while. In the 1980's, in an effort to create a
local computer industry, the government banned imports of computers. This
made it somewhat difficult to get the industry started. There were all
kinds of clones and knock-offs being produced for a while.
I don't think many of those Unitrons where made, they would probably be
worth a bit of money to a completest collector, but not much to anyone else.
12:15 p.m.
New subject: How much a brazilian mac clone is valued?
Jason McBrien wrote:
I vaguely remember reading somewhere that Brazil had
really odd rules
concerning computers for a while. In the 1980's, in an effort to create a
local computer industry, the government banned imports of computers.
That is an extremely popular misconception, even among those who
actually lived through the infamous "reserved market" policy. Due to the
huge foreign debt crises, all imports were banned in Brazil at that
time: from cars to chocolate bars. What was unique about computers was
that in addition to imports being illegal, foreign companies were
forbidden from making micro or minicomputers locally (either for local
sales or for exports). Long established companies such as IBM and
Burroughs continued to make their mainframes.
This
made it somewhat difficult to get the industry started. There were all
kinds of clones and knock-offs being produced for a while.
It wasn't the industry that I would have liked to see, but since the
whole world standardized on PC clones not as much damage was done as I
had feared. In retrospect, I see that the lack of venture capital and
the huge economic turmoil created a situation where all of the
reasonably funded companies had people from outside the industry in
charge. If you want to imagine how bad that can be, think of what result
you would have if you made someone from Pepsi or Nabisco president of a
US computer company ;-)
Another negative factor was that the rules were made for minicomputers
and didn't take a third party software industry into account. Copyright
was only extended to software in late 1987. With nothing but piracy
(importing software was illegal, but copying it was not) a clone has a
tremendous advantage over an original design (like the ones I worked
on).
Brazil had a very weak semiconductor industry, with the bulk of it being
one small factory by Texas Instruments and another one being built by
Ford. The government killed that off, in exchange for promises of
private investment in locally owned factories. That didn't work out well
(Brazil has never had a semiconductor factory since) but at least the
"no import" rule was never extended to components.
But in terms of revenues, the local industry did grow from scratch to an
impressive size by the end of the 1980s. A lot of companies were killed
off in the increasing economic mess, so practically all that were still
around when the reserved market policy ended in 1992 are still with us
(another popular misconception is that in 1993 the foreign companies
swept in and killed all the locals).
I don't think many of those Unitrons where made,
they would probably be
worth a bit of money to a completest collector, but not much to anyone else.
Even most collectors are probably not interested. As for numbers,
Unitron never got the permission to sell since Apple applied government
to government pressure to keep that from happening. There is one Unitron
partner who likes to give interviews saying they had a warehouse with
300 machines ready to ship. I really doubt that and have indications
that the actual number of prototypes was below 60. But since this was an
active development project, there were many variations.
One was the "Turbo Mac" which used the PAL equations I designed (not
reveresed engineered as many texts claim) instead of the Apple ones
(cloned through reverse engineering), as I described in
http://www.merlintec.com/lsi/mac512.html
There were variations with two internal floppies (like the link
Alexandre gave), but most had just one like the original Mac. Some
floppies didn't have the mechanical eject button, but most did. A few
had black keycaps and mouse button, while others are closer to the
original in color. Almost all have EPROMs copied from the Apple ROMs,
but a few have an expansion board with RAM that is loaded at boot time
and which runs the software that Rainer Brockerhoff was developing. All
of the Unitron boards, as far as I know, have a socket for the SCSI chip
(like a Plus) but it is unlikely that more than one or two prototypes
had the actual chip and the software to use it. I saw one version which
had an external 9 inch monitor and looked like a tiny PC.
When the analog board of my Unitron machine went bad, I had to take it
apart anyway and so took some pictures of the inside. You can see those
and other material related to the project as the files with "unitron" in
their name at
http://www.merlintec.com/download/
-- Jecel
3:39 p.m.
New subject: How much a brazilian mac clone is valued?
On 10/11/11 9:27 AM, Jason McBrien wrote:
I vaguely remember reading somewhere that Brazil had
really odd rules
concerning computers for a while. In the 1980's, in an effort to create a
local computer industry, the government banned imports of computers. This
made it somewhat difficult to get the industry started. There were all
kinds of clones and knock-offs being produced for a while.
A watered down form persists in the form of high import tariffs -- a
laudable and necessary idea, except for loopholes which Dell and such
seem to exploit. Brazil doesn't have a corresponding ban on
foreign-owned local manufacturing, or if it does, it's incompletely,
ahem, enforced.
Also, these policies have not helped the Windows problem: Brazil is one
of the worst-infested countries on the planet, by numbers (and hence
also in terms of malware and the black hat industry that has grown
around it there).
--T
I don't think many of those Unitrons where made,
they would probably be
worth a bit of money to a completest collector, but not much to anyone else.
11 Nov
11 Nov
9:11 a.m.
New subject: How much a brazilian mac clone is valued?
Toby,
A watered down form persists in the form of high
import tariffs -- a
laudable and necessary idea, except for loopholes which Dell and such
seem to exploit. Brazil doesn't have a corresponding ban on
foreign-owned local manufacturing, or if it does, it's incompletely,
ahem, enforced.
The big change in 1992 was exactly to encourage foreign companies to set
up local factories, as Dell did. And even Foxconn is expanding its
factory in Brazil to make Apple iPads locally (though the decision to go
through with this changes every two weeks). I don't have an opinion on
whether this strategy is good or not, but the results are certainly
those that were planned.
Also, these policies have not helped the Windows
problem: Brazil is one
of the worst-infested countries on the planet, by numbers (and hence
also in terms of malware and the black hat industry that has grown
around it there).
On the other hand, there is a strong open source movement and you can
see Linux based computers being sold in supermarkets. The ministry of
education only buys machines with open source for the students. The
impact on the Windows situation you described is slight, but it is
growing as time goes by.
-- Jecel
4:51 p.m.
New subject: Tech imports/open source in Brazil - Re: How much a brazilian mac clone is valued?
On 11/11/11 12:11 PM, Jecel Assumpcao Jr. wrote:
That is good to know. I've worked with several Brazilians who are very
enthusiastic about open source, and I look forward to them changing the
situation over time!
--Toby
Toby,
Yes, I'm aware Dell has a factory. I assumed it was a loophole; but
thanks for clarifying. Seems it's a loophole in the spirit of the
tariffs that is large enough to sail a container ship through...
A watered down form persists in the form of high
import tariffs -- a
laudable and necessary idea, except for loopholes which Dell and such
seem to exploit. Brazil doesn't have a corresponding ban on
foreign-owned local manufacturing, or if it does, it's incompletely,
ahem, enforced.
The big change in 1992 was exactly to encourage foreign companies to set
up local factories, as Dell did. And even Foxconn is expanding its
factory in Brazil to make Apple iPads locally (though the decision to go
through with this changes every two weeks). I don't have an opinion on
whether this strategy is good or not, but the results are certainly
those that were planned.
Foxconn isn't very well regarded as an employer. I hope this isn't
merely a barometer of desperation. I also hope they don't hike the
worker suicide rate as they have done, without giving a damn, in China.
Also, these policies have not helped the Windows
problem: Brazil is one
of the worst-infested countries on the planet, by numbers (and hence
also in terms of malware and the black hat industry that has grown
around it there).
On the other hand, there is a strong open source movement and you can
see Linux based computers being sold in supermarkets. The ministry of
education only buys machines with open source for the students. The
impact on the Windows situation you described is slight, but it is
growing as time goes by.
-- Jecel
10 Nov
10 Nov
6:33 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 10, 2011, at 12:09 AM, Scott Kevill wrote:
So..
- a huge 64-pos (MC68000 and some microcontrollers) universal ZIF socket
Do they make those? The 64-pin DIPs are even wider than the 40-pin ones; I think it's
900mil (almost an inch).
- alternatively a 48-pos universal ZIF and a
regular 64-pos socket
That might be more doable; I see 40-pin ones, at least, on most ROM burners.
- a way to output 5V from all 64 I/Os
You probably don't actually need to output 5v. TTL (the logic family you're
looking to deal with) generally didn't actually *output* anywhere over 3v, so the
switching threshold is considerably lower than that (generally 2.0v for TTL). Thus 3.3v
CMOS logic (which does swing the full 3.3v) should be fine as long as your inputs can take
occasional excursions above 3.3v.
- additional 5V tolerance for 16 of the
non-5V-tolerant FPGA I/Os (48 are already 5V tolerant)
Bear in mind that for a lot of the giant DIPS (64 pins, etc) there are multiple power and
ground pins. So you may not need them all to be tolerant.
- either additional -5V to +15V tolerance for all I/Os
or a way to manually isolate per-I/O from the manually wired supply voltages above (eg.
SPDT dip switches?) The FPGA I/Os do support tri-stating, but I'm not sure that would
be safe.
Tristating a pin will not save it; all it does is turn off a pass transistor or two, and
if that transistor's breakdown voltage is exceeded, it will conduct (usually
catastrophically). You could use external switches (such as a CD4016 or similar, which
will go up to 15v), but this one will take some more thought.
- highest signal rate needed would probably be 8MHz
for the MC68000, although hmm.. if I were sampling or measuring IC timing, perhaps higher
for input signals (eg. 32 MHz)
Give yourself some margin and build for about 20MHz bandwidth. It'll save you a lot
of signal integrity headaches.
I would be creating a PCB for this, but through-hole
would be preferred as I have no experience (yet) with SMT.
If you can fit it, great! Through-hole is also generally a lot more mechanically stable
(I can't count the number of times I've ripped off an SMT pin header because
I've been clumsy or not thinking straight; usually takes the pads off, too).
Thoughts? Suggestions? Or is this crazy/na?ve? :)
It's not a bad idea; it sounds like what a lot of ROM burners/readers do, just
expanded. If you're looking to program ROMs, you might want to keep in mind that a lot
of older EPROMs and similar devices (even older PICs and such) use a high voltage
(typically 12v) on a pin to signify that you're ready to program. You might want a
way to switch that in.
Scott.
SIde note: Are you the Scott Kevill responsible for GameRanger? If so, thank you, it made
my high school years a lot more fun (back when it first came out, around the time of Quake
2 for Mac). If not, never mind.
- Dave
9:58 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 10/11/2011, at 10:33 PM, David Riley wrote:
Unfortunately, I do. For example, the Z80A (and support chips) require the CLK input to be
>= +4.4V (ie. Vcc - 0.6V) even though 3.3V is plenty for all the other inputs. With the
8021, the T1, XTAL1, XTAL2, and RESET inputs are 3.8V min.
Aside from that, being able to drive +5V means that most chips can be 100% mapped by the
HDL with no manual rewiring. For example, I could tie an I/O to high to provide Vcc for
the chip.
I don't need all of them to be tolerant at the same time, right, but as a generic
board that can work with any chip I insert, I cannot make assumptions on the layout of
which socket pins will need to be power or ground.
Thanks, I suspected that might be the case, and somewhere deep in the datasheets would
probably have the specifics, too.
Another thought might be a two dual-rows of header pins, which would all have
jumpers/straps across them (allowing flow between FPGA I/O and chip). Then for the one or
two custom voltages, I'd remove the jumper and slide header pin sockets attached to
flying leads manually.
Noted. The FPGA will have no problem reaching those speeds and much higher. I will just
need to make sure that any level shifting or tolerance additions don't slow things
down too much.
That's nice to know. It's just that it seems to be getting harder to find newer
parts with DIP style.
Yeah, the idea would be that I could place almost any type of chip there, and then put any
kind of digital "host" / life-support system around it. That would make it a
great tool to help with testing chips, replicating chips, measuring their timing. Or even
verifying that an FPGA replication of a vintage computer works perfectly with each
original hardware component.. one-by-one substituting your replicated HDL component with
the real component. So many things you could do with this.
Haha, the very same! Thanks for the kind words. I'm still doing GameRanger. It now
supports PC as well as Mac, and as a result has *way* more users, as you might imagine.
I suppose you could thank me by helping me with this crazy plan. ;)
Scott.
On Nov 10, 2011, at 12:09 AM, Scott Kevill wrote:
I thought they did, but the pitch wasn't 0.1". Non-universal 64-pos ZIFs are
available, but just a plain socket would probably do for that. Yeah, the MC68000 I have
here looks to be about 0.9" wide. The 48-pos universal ZIF looks pretty common for
about $4.
So..
- a huge 64-pos (MC68000 and some microcontrollers) universal ZIF socket
Do they make those? The 64-pin DIPs are even wider than the 40-pin ones; I think
it's 900mil (almost an inch).
- alternatively a 48-pos universal ZIF and a
regular 64-pos socket
That might be more doable; I see 40-pin ones, at least, on most ROM burners. - a way to
output 5V from all 64 I/Os
You probably don't actually need to output 5v. TTL (the logic family you're
looking to deal with) generally didn't actually *output* anywhere over 3v, so the
switching threshold is considerably lower than that (generally 2.0v for TTL). Thus 3.3v
CMOS logic (which does swing the full 3.3v) should be fine as long as your inputs can take
occasional excursions above 3.3v. - additional
5V tolerance for 16 of the non-5V-tolerant FPGA I/Os (48 are already 5V tolerant)
Bear in mind that for a lot of the giant DIPS (64 pins, etc) there are multiple power and
ground pins. So you may not need them all to be tolerant. - either
additional -5V to +15V tolerance for all I/Os or a way to manually isolate per-I/O from
the manually wired supply voltages above (eg. SPDT dip switches?) The FPGA I/Os do support
tri-stating, but I'm not sure that would be safe.
Tristating a pin will not save it; all it does is turn off a pass transistor or two, and
if that transistor's breakdown voltage is exceeded, it will conduct (usually
catastrophically). You could use external switches (such as a CD4016 or similar, which
will go up to 15v), but this one will take some more thought. - highest
signal rate needed would probably be 8MHz for the MC68000, although hmm.. if I were
sampling or measuring IC timing, perhaps higher for input signals (eg. 32 MHz)
Give yourself some margin and build for about 20MHz bandwidth. It'll save you a lot
of signal integrity headaches. I would be
creating a PCB for this, but through-hole would be preferred as I have no experience (yet)
with SMT.
If you can fit it, great! Through-hole is also generally a lot more mechanically stable
(I can't count the number of times I've ripped off an SMT pin header because
I've been clumsy or not thinking straight; usually takes the pads off, too). Thoughts?
Suggestions? Or is this crazy/na?ve? :)
It's not a bad idea; it sounds like what a lot of ROM burners/readers do, just
expanded. If you're looking to program ROMs, you might want to keep in mind that a lot
of older EPROMs and similar devices (even older PICs and such) use a high voltage
(typically 12v) on a pin to signify that you're ready to program. You might want a
way to switch that in. From what I remember, for EPROMs I only need to hold
12V in those cases to put it in program mode, right? The actual I/O-driven signalling is
on other pins at normal voltage levels. In which case, I'd be having to allow for that
anyway for the Western Digital FD chips that require a +12V source. The 8021 needs a +15V
source to dump its ROM.
Ideally, it'd be nice to eliminate all hand wiring. The FPGA would need to specify
what voltage to use on each pin without requiring 4 I/Os per pin, since 256 I/Os would be
a bit steep.
Hmm, that sounds a bit like a serial DAC. I wonder if something along those lines would be
the better way to do this? Or some other form of custom "I/O expander"? 64 pins,
each with a selectable level of -5V, 0V, +5V, +12V, and +15V (maybe more). Or if the
expander were smart enough, it could remember the "high" voltage setting
configuration per pin. Since it'll only need to toggle between 0 and "V".
Inputs back to the FPGA would only need to be high/low.
If it's fast enough, with that kind of design change, it could be extended to any
number of pins, and any number of sockets at once. The datasheet says the Spartan 6 can do
"Up to 1,080 Mb/s data transfer rate per differential I/O". Interesting...
Imagine the possibilities!
Scott.
SIde note: Are you the Scott Kevill responsible for GameRanger? If so, thank you, it
made my high school years a lot more fun (back when it first came out, around the time of
Quake 2 for Mac). If not, never mind.
9 Nov
9 Nov
11:03 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On 11/09/2011 07:25 AM, Keith Monahan wrote:
I'm anxious to build a PCB that is based around a
pair of these
http://www.ti.com/product/sn74cbtd3384
10-Bit FET Bus Switches With Level Shifting
for a level conversion app.
I sampled some, although there was only the .635 pitch 24 SSOP/QSOP
version
available.
Note that these aren't active translators. The "level shifting" in the
5V to 3.3V direction is just the pass FET being turned off, which means
that the 3.3V you get out can be slightly squirrelly. It's generally
fine for CMOS digital inputs, There is no level shifting at all in the
3.3V to 5V direction, they just pass the 3.3V signal through. There's
nothing wrong with that as long as you're aware of it.
What these parts do is not any different than other commonly available
"QuickSwitch" parts, such as those from IDT and Pericom, and you can get
those in many packages, including SO. Most of those do not have the
diode internally, so this TI part does save one component, but that
wouldn't excite me enough to switch from SO to SSOP if I didn't already
want SSOP for some other reason.
Eric
11:54 a.m.
New subject: 3.3V/5V level shifting (was Re: 4000 Series CMOS Parts?)
On Nov 9, 2011, at 2:03 PM, Eric Smith wrote:
On 11/09/2011 07:25 AM, Keith Monahan wrote:
Exactly. We used the QuickSwitch parts (with a 1N4148 on the power rail to drop the
threshold) on lots of PCI cards to FPGAs which won't take 5v (which is most of
'em). Worked a charm. It was really more of a safety thing anyway, as I'm given
to understand that most modern PCs use 3.3v signaling anyway, but since the cards were
universally keyed, we wanted to be safe.
Of course, the same board also had to interface to PC-104, which definitely is 5v in most
implementations. The QuickSwitches came in handy there, too. :-) SSOP came in handy,
because having both interfaces ended up using a whole lotta chips in a pretty small
space.
- Dave
I'm anxious to build a PCB that is based
around a pair of these
http://www.ti.com/product/sn74cbtd3384
10-Bit FET Bus Switches With Level Shifting
for a level conversion app.
I sampled some, although there was only the .635 pitch 24 SSOP/QSOP version available.
What these parts do is not any different than other commonly available
"QuickSwitch" parts, such as those from IDT and Pericom, and you can get those
in many packages, including SO. Most of those do not have the diode internally, so this
TI part does save one component, but that wouldn't excite me enough to switch from SO
to SSOP if I didn't already want SSOP for some other reason.
4 Nov
4 Nov
3:39 p.m.
On 11/03/2011 09:07 AM, Michael Thompson wrote:
A friend said that he has a collection of NOS 4000
series CMOS ICs.
If there is any interest I will get them so they don't go in the trash.
I'll take them if nobody else needs them.
-Dave
--
Dave McGuire
New Kensington, PA
4559
days inactive
4581
days old
98 comments
27 participants
participants (27)
-
ard＠p850ug1.demon.co.uk -
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
chris＠mainecoon.com -
dkelvey＠hotmail.com -
emu＠e-bbes.com -
eric＠brouhaha.com -
fraveydank＠gmail.com -
geneb＠deltasoft.com -
hilpert＠cs.ubc.ca -
ian_primus＠yahoo.com -
jbmcb1＠gmail.com -
jecel＠merlintec.com -
jfoust＠threedee.com -
jkunz＠unixag-kl.fh-kl.de -
keithvz＠verizon.net -
mcguire＠neurotica.com -
michael.99.thompson＠gmail.com -
mouse＠Rodents-Montreal.ORG -
pinball＠telus.net -
pontus＠Update.UU.SE -
pu1bzz.listas＠gmail.com -
ragooman＠gmail.com -
schoedel＠kw.igs.net -
scott＠kevill.com -
tdk.knight＠gmail.com -
toby＠telegraphics.com.au