11 May
2004
11 May
'04
10:01 p.m.
Alright, I am awake. Several of you are heading to bed ...
I will give a short description of my project, then you can decide
for yourself which way you want to go.
Warning! Sort of longish reply ahead!
My design consist of 2 boards, one is called "Core", the other "I/O".
Both boards are almost Eurocard size. Power supply: only +5 Volts.
Features of the Core board.
---------------------------
Chips: 6809, 6821, 6850, 6264, 27128 and 27256, in words: the CPU
is the 6809 which is a 8/16 bit quite powerful CPU which has good
high-level language support (think of C), and you can read more abt
the 6809 in an other recent thread. The 27128 is 16k EPROM at memory
C000-FFFF, of which half is filled with a debug monitor and software
to control the I/O board. The 27256 is 32k -optional- EPROM at 4000-
BFFF. The 6264 is 8k RAM at 2000-3FFF. The 6821 (2x8bit I/O) forms
the buffered interface to the I/O board(s). Last, the 6850 is the ACIA,
which makes the serial (9600 Bd) connection to *any* host computer.
On board level conversion makes it a true RS-232 port.
The ACIA and the PIA are mapped in $0080-$00FF, that is legacy from
the design around the 6802 (RealConsole Mk I).
The serial port can be connected to a terminal (e.g. VT220) and you
use the monitor to debug your own (assembler) programs. Or you connect
the serial port to a PC, SUN workstation or whatever and control the
"thing" via the serial port. My application for it is to control the
I/O that drives LEDs and read switches from a home-brew console.
I have made some patches in SIMH, and I can *boot* RT11 via my console
as if were the real thing. See www.pdp-11.nl , click in the left menu
on "homebrew PDP-11". On the opening page click in the top section on
"SIMH software", "new design" and "action!" for more ...
Features of the I/O board.
--------------------------
The I/O board has 3 LS138 (1 to 8 decode chips) and octal latches
(373 and 374). One I/O board can have up to *eight* 8-bit output
ports and *eight* 8-bit input ports, thus there is on one I/O board
64 outputs and 64 input available, enough to drive a full PDP-11/70
console. The I/O board is connected to the Core board with a simple
20-wire flat cable. If what you want to control is bigger, or perhaps
totally different, you can cascade (theoretically) up to 8 I/O boards
by just running that flatcable from I/O board to I/O board.
A single jumper on the I/O board settles the addressing issues.
The practical limit of I/O boards is six (due to fan-out load issues),
but if you insist and go outside specifications I guess you could get
away with 7 I/O boards.
Applications for the Core and I/O board.
- First, the mentioned console, but *NOT* limited to PDP-xx/xx consoles!
Think of the IBM S3; not a small one, but it can be connected.
The console can be home made, but if you have a *real* console laying
around, that can be connected too! (I'd prefer a real console to go
with a real machine tough ...)
- Why limit to "blinkenlights" consoles? You can just as easy connect
it to the more modern console like the 11/34 or 11/60 ! The control of
the 7-segment displays is of course in software not with hex to 7 segment
decoder chips!
- Use some I/O to make a parallel I/O port (if you need it). I am not sure
if the speed will be sufficient to connect an IDE drive (in PIO mode 1), but
that is something I intend to try ...
- Whatever project to wanted to do! There is enough I/O capability ...
The monitor software is described on the site mentioned. If you know
the Motorola Mikbug monitor, this will look familiar.
The RealConsole software controls the I/O ports. The input ports are
read periodically in an interrupt service routine. The *debounced* data
from the switches is stored in memory locations, so
your application can
read the switches from those memory locations. Several
"hooks" (in fact
small subroutines) allow you to control any bit of any output port.
There are a few 'special' routines to handle the issues involved with
momentary push-buttons, toggles. A depressed toggle and then released
would mean that in memory that specific bit would be set for a brief
moment and then be reset. To solve that issue, you can specify via
a hook on which input port(*s*), which bit(*s*) are connected to toggles.
Those bits are processed differently. Once set (by pressing the toggle)
that bit remains set until you command (an other hook) to return to zero.
So, you can process each toggle when you want, and when that is actually
done you reset the "toggle".
Some "higher" level commands are e.g. "Axxxxx" where the "x"
is a hex
number. This command will put xxxxx on the Address LEDs. Likewise, there
is a "Dxxxx" command. You figure out what it does :-)
Project status
--------------
The project was ready to go manufacture the lay-out PCB set prototype
tpo build it and check if it does not contain errors. This week I am
working on the "upgrade" from the 6802 to the 6809 CPU, because of its
better availability. Performance improvement etc. are a nice side effect
of this move.
When the prototype is checked (within one month), the board sets are
"mass produced" in the US. Silkscreened, solder mask, etc. - professional
quality. To keep those costs low, it takes 5 weeks to get them made.
(could be done in a few days, but don't ask what the compay charges!)
You can order the board set (or more I/O Boards than Core boards) either
bare, or together with the "difficult" chips, or as a complete kit
that contains all components. I am not going to make much profit on it,
but when more people join the bigger discount we can get on the parts!
For those who rather not solder the boards, I am prepared to do that
work and test them (for a fee though).
I will write a "manual" that described the software, give a step-by-step
DYI building the boards (with fotographs), connection diagrams, multiple
I/O Board configuration plus the fan-out load calculations.
From my site you can download the software source code
and the changes
made to the current version of SIMH.
Believe me, I am dedicated to this project!
As Vince said, he did the PCB design and solved the routing issues.
We try to buy the parts as cheap as possible, so those will probably
be bought in the US and not where I Live - The Netherlands. We are
working on how to ship it all. It is a little stupid to ship many
parts from US to Holland, assemble the kits, anmd the send back some
70% of it all to the US again ...
If you have any questions regarding this project, ask!
I will try to give an answer as good as I can, as soon as I can.
I will help thinking about your specific application. For example, one
guy asked for the possibility to connect an 8-bit input port directly
to an 8-bit output port and use the combination as a bi-directional
8-bit port. The final idea was building a setup to test M-boards!
(BTW, the answer is yes with a minor patch!)
kind regards,
- Henk, PA8PDP
gooi(a)oce.nl
11 May
11 May
10:15 p.m.
On Wed, 12 May 2004, Gooijen H wrote:
As Vince said, he did the PCB design and solved the
routing issues.
We try to buy the parts as cheap as possible, so those will probably
be bought in the US and not where I Live - The Netherlands. We are
working on how to ship it all. It is a little stupid to ship many
parts from US to Holland, assemble the kits, anmd the send back some
70% of it all to the US again ...
I can probably "traffic" some of the boards and/or parts from NL to
the US (west coast), and vice versa, for those who want it.
--fred
12 May
12 May
4:07 a.m.
>> Chips: 6809, 6821, 6850, 6264, 27128 and
27256.
I was programming these parts as far back as 1979 [Ok I was using 2716/2732
eproms, but the 27128/27256 are not that much newer].
I am surprised that these parts are still in production..
My main question, however, is that since this is intended (I believe) for
long term use, is this the best hardware to base a design on. What are the
chances of getting spares for these components in 2014?
Just my thoughts.....
David.
5:16 a.m.
>> Chips:
6809, 6821, 6850, 6264, 27128 and 27256.
I was programming these parts as far back as 1979 [Ok I was using eproms, but the 27128/27256 are not that much newer].
I am surprised that these parts are still in production..
That was the main motivation for the move to the 6809, which has much
broader availability than the 6802.
My main question, however, is that since this is
intended (I believe) for
long term use, is this the best hardware to base a design on. What are the
chances of getting spares for these components in 2014?
I believe that this is a problem with virtually *any* component you could
choose. The newer stuff seems to generally have shorter lifespans. I
believe this is an outcome of having more choices -- none of the "newer"
chips (with perhaps some exceptions from Intel) has enough of a following to
be readily available down the road.
At least these old-time components have a following that has kept them
around this long...
I'm curious what everyone thinks of this issue, and what *are* the better
choices?
Vince
2 p.m.
I believe that this is a problem with virtually *any*
component you could
choose. The newer stuff seems to generally have shorter lifespans. I
believe this is an outcome of having more choices -- none of the "newer"
chips (with perhaps some exceptions from Intel) has enough of a following to
be readily available down the road.
As you said, the newer chips are worse. I've come across chips that have
gone out of production in the time between the announcement and the
distributors getting stocks of them. Hmmm....
Personally, the only 8-bit CPU I think can be depended on to stay around
is the Intel 8031 family. It's used everywhere, and many manufacturers
either make it, or make 'improved' versions. You can ignroe the
'improvements' if you like.
I thought the 6809 was getting somewhat hard to find now :-(. The Z80 and
6502 were a lot more popular and are still listed in catalogues over here
(the 6809 doesn't seem to be.)
As for SRAM and EPROMs. I am sure both will be around in 10 years time,
even if at a vastly increased capacity. And just as now we put 2764s in
place of 2716s with a few minor mods, I guess in the future we'll make up
daughterbaords to put the latest SMD EPROMs into 28 pin DIL sockets.
-tony
3:52 p.m.
On Thu, May 13, 2004 at 12:00:39AM +0100, Tony Duell wrote:
As you said, the newer chips are worse. I've come
across chips that have
gone out of production in the time between the announcement and the
distributors getting stocks of them. Hmmm....
No kidding... the project I'm working on has had to change chips at
least once, mid-stream.
Personally, the only 8-bit CPU I think can be depended
on to stay around
is the Intel 8031 family. It's used everywhere, and many manufacturers
either make it, or make 'improved' versions. You can ignroe the
'improvements' if you like.
Those certainly are popular and prevalent.
I thought the 6809 was getting somewhat hard to find
now :-(. The Z80 and
6502 were a lot more popular and are still listed in catalogues over here
(the 6809 doesn't seem to be.)
I'm personally a big 6502 fan, but I also recognize that it's not
universally popular.
As for SRAM and EPROMs. I am sure both will be around
in 10 years time,
even if at a vastly increased capacity. And just as now we put 2764s in
place of 2716s with a few minor mods, I guess in the future we'll make up
daughterbaords to put the latest SMD EPROMs into 28 pin DIL sockets.
Well... that _is_ the the sort of thing that's going to be harder to find
in 10 years... DIP parts... SMD adapters will work, but if you have to
pay for lots of them... they'll get expensive. One for an EPROM isn't
too bad, but dozens?
Fortunately, lots of us have lots of DIP parts lying around, enough to last
years and years.
-ethan
--
Ethan Dicks, A-130-S Current South Pole Weather at 13-May-2004 00:40 Z
South Pole Station
PSC 468 Box 400 Temp -67.2 F (-55.1 C) Windchill -127.2 F (-88.5 C)
APO AP 96598 Wind 14.5 kts Grid 022 Barometer 688.1 mb (10321. ft)
Ethan.Dicks(a)amanda.spole.gov http://penguincentral.com/penguincentral.html
5:35 p.m.
> I believe that this is a problem with virtually
*any* component you
could
> choose. The newer stuff seems to generally have
shorter lifespans. I
> believe this is an outcome of having more choices -- none of the "newer"
> chips (with perhaps some exceptions from Intel) has enough of a
following to
be readily
available down the road.
As you said, the newer chips are worse. I've come across chips that have
gone out of production in the time between the announcement and the
distributors getting stocks of them. Hmmm....
Personally, the only 8-bit CPU I think can be depended on to stay around
is the Intel 8031 family. It's used everywhere, and many manufacturers
either make it, or make 'improved' versions. You can ignroe the
'improvements' if you like. I thought the 6809 was getting somewhat hard to find
now :-(. The Z80 and
6502 were a lot more popular and are still listed in catalogues over here
(the 6809 doesn't seem to be.)
The 6809 can still be had from Mouser, and a few other places with much
better prices.
In this case, I think the 680x was originally chosen for the ease of
programming. I concede that the 8031 and the Z80 are both more available
(and likely to remain so). I don't see 6502 at, say, Digikey. They do seem
to be as available as the 6809, though.
As for SRAM and EPROMs. I am sure both will be around
in 10 years time,
even if at a vastly increased capacity. And just as now we put 2764s in
place of 2716s with a few minor mods, I guess in the future we'll make up
daughterbaords to put the latest SMD EPROMs into 28 pin DIL sockets.
Maybe we should all design them into everything we do, in an effort to
sustain demand for them at a level where they are considered worth providing
:-).
(BTW, for anyone who thinks that might actually work: there is a real world
out there, waiting to be discovered.)
Vince
13 May
13 May
9:13 a.m.
On 12 May 2004 at 7:16, vrs wrote:
I'm curious what everyone thinks of this issue,
and what *are* the
better choices?
The Motorola 68EC000 makes a reasonable microcontroller. 16 MB addressing
range, configurable 8- or 16-bit data bus, a range of operating speeds (far
faster than a 6809), freely available tools, still in full production, and
about $4.00 in single lots.
-- Dave
14 May
14 May
6:05 a.m.
I'm
curious what everyone thinks of this issue, and what *are* the
better choices?
The Motorola 68EC000 makes a reasonable microcontroller. 16 MB addressing
range, configurable 8- or 16-bit data bus, a range of operating speeds faster than a 6809), freely available tools, still in
full production, and
about $4.00 in single lots.
I'm not finding those. Where are you seeing/getting them? Closest that I
found were some other 68Ks that were in the $8-$10 range.
Vince
6:18 a.m.
At 08:05 AM 5/14/2004 -0700, you wrote:
(far
I've been using the MC68VZ328PV in my designs of late. It is a 68k derivative,
has a nice set of I/O, and is cheap. I did have problems locating a source
for single quantities, so I just bought a tray of 60... They are TQFP 144, so
they don't have the problems that a BGA package does for prototype work. The
MC68EZ328 is another version which is slightly slower and has less I/O. The
GCC compiler tools work well with these as they are 68k based.
--tom
I'm curious what everyone thinks of this issue,
and what *are* the
better choices?
The Motorola 68EC000 makes a reasonable microcontroller. 16 MB addressing
range, configurable 8- or 16-bit data bus, a range of operating speeds faster than a 6809), freely available tools,
still in full production, and
about $4.00 in single lots.
I'm not finding those. Where are you seeing/getting them? Closest that I
found were some other 68Ks that were in the $8-$10 range.
8:29 a.m.
> > > I'm curious what everyone thinks of
this issue, and what *are* the
> > > better choices?
> >
> > The Motorola 68EC000 makes a reasonable microcontroller. 16 MB
addressing
> > range, configurable 8- or 16-bit data bus, a
range of operating speeds
>(far
> > faster than a 6809), freely available tools, still in full production,
and
> > about $4.00 in single lots.
>
>I'm not finding those. Where are you seeing/getting them? Closest that
I
found were some
other 68Ks that were in the $8-$10 range.
I've been using the MC68VZ328PV in my designs of late. It is a 68k has a nice set of I/O, and is cheap. I did have
problems locating a source
for single quantities, so I just bought a tray of 60... They are TQFP 144,
so
they don't have the problems that a BGA package
does for prototype work.
The
MC68EZ328 is another version which is slightly slower
and has less I/O.
The
GCC compiler tools work well with these as they are
68k based.
I found those at $28 each (qty 1), but that seems pretty high for a
microcontroller. That's at least half the parts budget, for most of the
projects I am thinking of. So I still have the "where" question :-).
TQFP-144...I guess I could figure out how to attach that to my homemade
PCB...something tells me you have way nicer tools than I do.
Vince
8:40 a.m.
Vince said:
I found those at $28 each (qty 1), but that seems
pretty high for a
microcontroller. That's at least half the parts
budget, for most of the
projects I am thinking of. So I still have the
"where" question :-).
TQFP-144...I guess I could figure out how to attach
that to my homemade
PCB...something tells me you have way nicer tools
than I do.
http://www.seattlerobotics.org/encoder/200006/oven_art.htm
It works, according to my own XC95144XL (TQ100)
experience.
vax, 3900
__________________________________
Do you Yahoo!?
SBC Yahoo! - Internet access at a great low price.
http://promo.yahoo.com/sbc/
9:20 a.m.
TQFP-144...I
guess I could figure out how to attach
that to my homemade
PCB...something tells me you have way nicer tools
than I do.
http://www.seattlerobotics.org/encoder/200006/oven_art.htm
It works, according to my own XC95144XL (TQ100)
experience. 
9:26 a.m.
It was written...
oh my! I've been staying away from anything surfacemount, but this looks
like a great way to get into that! The first thing coming to mind that the
above solution gives to me.. is the ability to do the I/O board add-on for
the SBC6120 which uses a large surfacemount chip or two. Awesome!
Jay West
---
[This E-mail scanned for viruses by Declude Virus]
9:44 a.m.
Jay West wrote:
It was written...
oh my! I've been staying away from anything surfacemount, but this looks
like a great way to get into that! The first thing coming to mind that the
I started using surface mount components several years ago, and now I
*hate* having to use leaded components. The smallest spacing I have
soldered is an ICS525 with .025" lead spacing, and that really is no
problem now. Just use a small soldering tip and small solder; I use
.015". Just make sure your soldering tip is in *excellent* condition!
One of these days, I'll be able to afford a MetCal unit (about $125 on
the used equipment market or more if you want better :).)
Another advantage of surface mount components; they take up a LOT LESS
inventory space :).
12:08 p.m.
On Fri, May 14, 2004 at 01:26:03PM -0500, Jay West wrote:
It was written...
oh my! I've been staying away from anything surfacemount, but this looks
like a great way to get into that! The first thing coming to mind that the
above solution gives to me.. is the ability to do the I/O board add-on for
the SBC6120 which uses a large surfacemount chip or two. Awesome!
I've looked into the oven technique, but for my IOB6120, I just did all
the SMT by hand (but I've done it before).
-ethan
--
Ethan Dicks, A-130-S Current South Pole Weather at 14-May-2004 21:00 Z
South Pole Station
PSC 468 Box 400 Temp -74.2 F (-59.0 C) Windchill -119.1 F (-84 C)
APO AP 96598 Wind 9.9 kts Grid 026 Barometer 694.3 mb (10092. ft)
Ethan.Dicks(a)amanda.spole.gov http://penguincentral.com/penguincentral.html
9:32 a.m.
>>>> "vrs" == vrs
<vrs(a)msn.com> writes:
> > TQFP-144...I guess I could figure out how to
attach > that to my
> homemade > PCB...something tells me you have way nicer tools >
> than I do.
Nothing to it. You need a fine tip soldering iron, thin solder,
liquid flux, a good light, and a magnifying glass. Oh yes, your PCB
should have solder mask.
I've built a board with several fine pitch PQFPs and about 100 0603
capacitors, among other small stuff, using exactly those tools. Works
just fine.
paul
10:37 a.m.
On Friday 14 May 2004 13:32, Paul Koning wrote:
Even better, get a nice wide spade-tip for the soldering iron, so you
can heat multiple pins at once. Once you get the hang of it you can
"drag" the iron down along the part with the solder, and solder an IC
*very* quickly. I learned that trick in my senior design class. I
used to be afraid of surface-mount parts, but I am no longer (except
programmable things or using them in breadboards, which end up
requiring expensive adaptor sockets)
Of course, the fine tip is useful for some things, such as through-hole
devices, but you can do pretty much everything you'll need with a spade
tip.
>>>> "vrs" == vrs <vrs(a)msn.com> writes:
>>>>
> > TQFP-144...I guess I could figure out how to attach > that to
> > my
>
> homemade > PCB...something tells me you have way nicer tools >
> than I do.
Nothing to it. You need a fine tip soldering iron, thin solder,
liquid flux, a good light, and a magnifying glass. Oh yes, your PCB
should have solder mask.
I've built a board with several fine pitch PQFPs and about 100 0603
capacitors, among other small stuff, using exactly those tools.
Works just fine.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
10:54 a.m.
On Fri, 14 May 2004, Patrick Finnegan wrote:
Peter Wallace
Mesa Electronics
On Friday 14 May 2004 13:32, Paul Koning wrote:
Even better, get a nice wide spade-tip for the soldering iron, so you
can heat multiple pins at once. Once you get the hang of it you can
"drag" the iron down along the part with the solder, and solder an IC
*very* quickly. I learned that trick in my senior design class. I
used to be afraid of surface-mount parts, but I am no longer (except
programmable things or using them in breadboards, which end up
requiring expensive adaptor sockets)
Of course, the fine tip is useful for some things, such as through-hole
devices, but you can do pretty much everything you'll need with a spade
tip.
The "miniwave" process that you describe is what I've always done
I find that a small tip is more likely to cause damage to pads so I always use
a 1/16 or 3/32 spade tip. The real secret is lots of flux so the solder is
clean and retains its surface tension...
Small solder is actually worse since it tends to carry more oxide, so I
usually use .032"
>>>> "vrs" == vrs <vrs(a)msn.com> writes:
>>>>
> > TQFP-144...I guess I could figure out how to attach > that to
> > my
>
> homemade > PCB...something tells me you have way nicer tools >
> than I do.
Nothing to it. You need a fine tip soldering iron, thin solder,
liquid flux, a good light, and a magnifying glass. Oh yes, your PCB
should have solder mask.
I've built a board with several fine pitch PQFPs and about 100 0603
capacitors, among other small stuff, using exactly those tools.
Works just fine.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
10:52 a.m.
"vrs" wrote:
I'd heard of that, but haven't had the need/guts to try it yet.
I draw the line at 100 pins and use a hot air rework tool :-)
more than 100 pins and I send it to a pro (color me chicken)
-brad
TQFP-144...I guess I could figure out how to attach
that to my homemade
PCB...something tells me you have way nicer tools
than I do.
http://www.seattlerobotics.org/encoder/200006/oven_art.htm
It works, according to my own XC95144XL (TQ100)
experience.
9:53 a.m.
At 10:29 AM 5/14/2004 -0700, you wrote:
derivative,
Ouch! I think I payed $6.281 each for my tray of 60 parts, which I consider
a pretty good price for what it does.
> > >
I'm curious what everyone thinks of this issue, and what *are* the
> > > better choices?
> >
> > The Motorola 68EC000 makes a reasonable microcontroller. 16 MB
addressing
> > range, configurable 8- or 16-bit data
bus, a range of operating speeds
>(far
> > faster than a 6809), freely available tools, still in full production,
and
> > about $4.00 in single lots.
>
>I'm not finding those. Where are you seeing/getting them? Closest that
I
found were
some other 68Ks that were in the $8-$10 range.
I've been using the MC68VZ328PV in my designs of late. It is a 68k has a nice set of I/O, and is cheap. I did have
problems locating a source
for single quantities, so I just bought a tray of 60... They are TQFP 144,
so
they don't have the problems that a BGA
package does for prototype work.
The
MC68EZ328 is another version which is slightly
slower and has less I/O.
The
GCC compiler tools work well with these as they
are 68k based.
I found those at $28 each (qty 1), but that seems pretty high for a
microcontroller. That's at least half the parts budget, for most of the
projects I am thinking of. So I still have the "where" question :-). TQFP-144...I guess I could figure out how to attach
that to my homemade
PCB...something tells me you have way nicer tools than I do.
I run Cadsoft Eagle and I've used the part on a two layer board. The Eagle
tools work pretty well and they have 3 levels, including a free set that
limits you to a smaller board and fewer layers. Over the years, I eventually
upgraded to the professional version so that I could do any size board that
I want. I did not purchase the autorouter -- I think humans are better at
it.
--tom
11:21 a.m.
TQFP-144...I
guess I could figure out how to attach that to my homemade
PCB...something tells me you have way nicer tools than I do.
I run Cadsoft Eagle and I've used the part on a two layer board. The Eagle
tools work pretty well and they have 3 levels, including a free set that
limits you to a smaller board and fewer layers. Over the years, I upgraded to the professional version so that I could
do any size board
that
I want. I did not purchase the autorouter -- I think
humans are better at
it.
I use Eagle too. Spent the whole $1200. Maybe whether humans are better at
routing depends on the human. It was pretty clear to me that wasn't how I
wanted to spend my time, even if I could do a better job, so I am really
happy with the autorouter -- I let it handle the "easy" 90 percent of the
routing job.
I like to make my own PCBs when I can, though -- no solder mask there.
Maybe I will try my wings with surface mount using some of these 74f04d I
recently inherited :-)
Vince.
11:42 a.m.
I use Eagle too. Spent the whole $1200. Maybe whether
humans are better at
routing depends on the human. It was pretty clear to me that wasn't how I
wanted to spend my time, even if I could do a better job, so I am really
happy with the autorouter -- I let it handle the "easy" 90 percent of the
routing job.
Heh! Well, I didn't want to pay for it either:-)
I like to make my own PCBs when I can, though -- no
solder mask there.
If you are doing much surface mount, especially with the finer pitch parts,
you want a solder mask.
Maybe I will try my wings with surface mount using some
of these 74f04d I
recently inherited :-)
I bought a used Metcal MX1 on eBay and it is easily the best iron I've
ever used! Throw out all of the others. I also have a Bausch & Lomb Stereo
Zoom 4 microscope which makes surface mount much easier, at least for my
aging eyes.
--tom
4:31 p.m.
New subject: soldering iron & advice for magnifier
Tom wrote...
I bought a used Metcal MX1 on eBay and it is easily
the best iron I've
ever used! Throw out all of the others. I also have a Bausch & Lomb Stereo
Zoom 4 microscope which makes surface mount much easier, at least for my
aging eyes.
I'm not familiar with Metcal units. How would the MX1 compare with the
Weller WES51 I recently bought?
Also, I have a lamp magnifier... one of those lighted magnifiers on a elbow
arm you clamp to your desk. Invaluable to me for small work. Mine is missing
one of the two springs, so I'm out looking for a new one. I found a truely
nice one for $55 that has two separate flourescent bulbs on each side (I
strongly prefer flourescent lights for that rather than the cheap
incandescent units). I'm thinking these would have to be better than a
stereo microscope (which I've never used)??? Wouldn't it be hard to position
boards and such under that?
Jay West
6:53 p.m.
New subject: soldering iron & advice for magnifier
I'm not familiar with Metcal units. How would the
MX1 compare with the
Weller WES51 I recently bought?
I don't know much about the WES51, but the Metcal is an RF driven device
and it heats up immediately and maintains the temperature of the specific
tip very well. The tips are as easy to change as sliding one out and
another on in. The full model number of the unit I purchased is an
Metcal MX-500P-11, and I paid $250. My understanding is that there are
a couple of Metcal models to avoid, possibly the SP200, but I'm not
certain. I have a couple if different tips, including a very fine conical
tip and a fine J tip which is good for removing parts.
Also, I have a lamp magnifier... one of those lighted
magnifiers on a elbow
arm you clamp to your desk. Invaluable to me for small work. Mine is missing
one of the two springs, so I'm out looking for a new one. I found a truely
nice one for $55 that has two separate flourescent bulbs on each side (I
strongly prefer flourescent lights for that rather than the cheap
incandescent units). I'm thinking these would have to be better than a
stereo microscope (which I've never used)??? Wouldn't it be hard to position
boards and such under that?
They make a nice stand for the stereo microscope which allows the board to
be placed pretty much anywhere under the lens. They also have a very high
intensity lamp which runs through a flexible light pipe so that you can
position it and aim it where you need it. I don't have either of these, but
my friend does and they are both nice. I will probably have to find both
at some point.
The nice thing about the microscope is that you can really magnify down
to the nitty gritty and see what you need to see.
--tom
15 May
15 May
8 a.m.
New subject: soldering iron & advice for magnifier
Tom Uban wrote:
The WES51 is "ok" (I have one), but it's not extremely accurate temp
wise and the smallest tips are small, but not extremely sharp. I'd call
it the Volkswagon. Dependable, cheap, works well.
The Metcal is the Porsche. Nice design, nice egonometrics, very fast,
very reliable, really, really nice.
just my opinion, but I've used both.
(and I'm still kicking myself for not buying a couple of used metcal's for
pennies on the dollar a few years ago - what was thinking?!)
-brad
I'm not familiar with Metcal units. How would
the MX1 compare with the
Weller WES51 I recently bought?
I don't know much about the WES51, but the Metcal is an RF driven device
and it heats up immediately and maintains the temperature of the specific 
9:43 a.m.
New subject: soldering iron & advice for magnifier
Brad Parker wrote:
Tom Uban wrote:
The WES51 is "ok" (I have one), but it's not extremely accurate temp
wise and the smallest tips are small, but not extremely sharp. I'd call
it the Volkswagon. Dependable, cheap, works well.
The Metcal is the Porsche. Nice design, nice egonometrics, very fast,
very reliable, really, really nice.
just my opinion, but I've used both.
(and I'm still kicking myself for not buying a couple of used metcal's for
pennies on the dollar a few years ago - what was thinking?!)
-brad
I've had a Metcal MX-500 for several years, and along with a RM3E
wand, a set of Talons. The Talons are "hot tweezers" and are invaluable
for removing Rs, Cs, and SOICs. Along with this I have a Nikon SMZ-2B
microscope with a remote light source, a Taiwanese knock-off of the
Hakko 852 hot air machine (the Hakko is definitely better), and a EFD
1500 XL paste dispenser (all from eBay!). With this combination, you can
do just about everything except BGA. I'm not the least bit afraid of
SMT (actually, I've gotten to really hate through-hole, and prefer
surface mount), but BGA is serious voodoo. It's the board layout
equivalent of doing buried vias.
I first "discovered" Metcals 12+ years ago. I was an software
engineer at a company, and dabbled in hardware. We had a small
manufacturing line with about 6 people on it. Vaugh Equipment Sales
called on us, and somehow wound up talking to me. They came in for a
demo, and brought a Metcal. I was impressed, and became an immediate
advocate. Every company I've worked at has either had Metcals already,
or been converted after some campaigning. There's just no comparison.
I've used the Hakkos, the Wellers, some other brands, and none even come
close to the quality and usability of a Metcal.
eBay is a great place to find these. Sometimes they go for too
much, and sometimes you can steal one. You want a MX-500, a RM3E wand,
a stand, something like a STTC-025 tip (small "normal" tip, for
through-hole, connectors, etc), a STTC-040 (30 degree angle fine tip,
for SMT), a STTC-117 (similiar to a Weller 100 watt gun tip. Good for
heavy ground planes, etc), and one of the hoof tips for drag soldering
(it's in a box that hasn't been unpacked from the move, can't remember
the number). If you're lucky, and stumble across a Metcal RF Power
Meter for under $100, buy it. It's an in-line wattmeter, and can tell
you a lot about the health of the base, the tip, and your workload.
I've bought tips off of eBay, too. Find someone selling an
assortment for of 5 or so used tips for $35. So far, everyone I've
dealt with has been accurate and honest about the condition of the
tips. The photos should be up-close and clear.
The non-MX-500s should generally be avoided. They're getting old,
and if they break, Metcal does not repair them anymore. And trust me,
you CANNOT fix them yourself. They require specialized tuning
equipment. If you repair a Metcal and it gets de-tuned, you'll fry the
output drivers, and it'll be truly dead. Sure, you can fix a power
switch or something like that, but that's not generally what fails.
The SP-200 should also be avoided. It does not use the same tips as
the MX series, and the output power is about half (15 watts vs 35).
They're not that cost effective to buy, although they're generally
better than something like Weller WTC-201 (or whatever is close to that
number. WT-something).
Metcal has a good article on why these irons are so good. The tip
temperature is very precise, and because they're low-mass tips, the
respond extremely quickly. They're one of the only irons that are
considered to have "self-cleaning" contacts. For MilSpec work, most
irons require a calibration procedure any time the tip is changed. This
usuallyinvolves cleaning where the tip goes in to the handle, and then
checking it on a temperature measuring device. Because the tip of a
Metcal is basically a section of transmission line, removing it from the
handle cleans the contact area. And last time I checked, Metcal tips
were not required to have a temperature check between changings (but
they do require a weekly test).
One of the other things I really like about a Metcal is how fast you
can change a tip. A tip heats up to working temperature in about 10-15
seconds. Turning the iron off, you hold the tip against the sponge for
about 10 seconds, and you can remove it with your fingertips. If 10
seconds is too long, a silicon pad comes with a the wand, which allows
you to remove the tip while it's still scorching hot. Also, the wand
cable is silicon, so you don't have to worry about damaging the cable if
you nail it with the iron. They don't, however, take to being slammed
in drawers well (this is the single biggest cause of failure, says
Metcal). The cable is flexible transmission line, and a crimp creates
an impedance bump in the line that causes the iron to be unable to
measure the tip temperature correctly.
--jc
I'm
not familiar with Metcal units. How would the MX1 compare with the
Weller WES51 I recently bought?
I don't know much about the WES51, but the Metcal is an RF driven device
and it heats up immediately and maintains the temperature of the specific
17 May
17 May
4:49 a.m.
New subject: soldering iron & advice for magnifier
>>>> "Jay" == Jay West
<jwest(a)classiccmp.org> writes:
Jay> Tom wrote...
> I bought a used Metcal MX1 on eBay and it is
easily the best iron
> I've ever used! Throw out all of the others. I also have a Bausch
> & Lomb Stereo Zoom 4 microscope which makes surface mount much
> easier, at least for my aging eyes.
Jay> I'm not familiar with Metcal units. How would the MX1 compare
Jay> with the Weller WES51 I recently bought?
I've used a Weller WTCPR (Curie point temperature control) iron for
many years. The "PTS" tip is sharp and slender, perfect for fine SMD
work. It's harder to find than most other tips. Extra thin solder
and a lamp + magnifier completes the toolkit.
Jay> Also, I have a lamp magnifier... one of those lighted magnifiers
Jay> on a elbow arm you clamp to your desk. Invaluable to me for
Jay> small work. Mine is missing one of the two springs, so I'm out
Jay> looking for a new one. I found a truely nice one for $55 that
Jay> has two separate flourescent bulbs on each side (I strongly
Jay> prefer flourescent lights for that rather than the cheap
Jay> incandescent units). I'm thinking these would have to be better
Jay> than a stereo microscope (which I've never used)??? Wouldn't it
Jay> be hard to position boards and such under that?
I'd say this is a matter of personal preference. I picked my tools
based on what was available along with the advice of a professional --
a lady who routinely assembles prototype boards full of 240 lead PQFPs
in no time flat. She uses one of those magnifiers with a circular
fluorescent tube wrapped around it. I didn't have one of those so I
made do with a Sunnex light plus magnifier combo, not quite as good
but sufficient.
The stereo microscope at the office did come in handy for
post-assembly inspection.
paul
14 May
14 May
10:43 a.m.
On Friday 14 May 2004 12:29, vrs wrote:
I found those at $28 each (qty 1), but that seems
pretty high for a
microcontroller. That's at least half the parts budget, for most of
the projects I am thinking of. So I still have the "where" question
:-).
Arrow.com has MC68VZ328's in stock for about $10 each, single quantity.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
11:07 a.m.
At 02:43 PM 5/14/2004 -0500, you wrote:
On Friday 14 May 2004 12:29, vrs wrote:
Heh! I had always done my search with the PV suffix and that always
showed no stock. No reason why the wider temperature range part wouldn't
be fine.
--tom
I found those at $28 each (qty 1), but that seems
pretty high for a
microcontroller. That's at least half the parts budget, for most of
the projects I am thinking of. So I still have the "where" question
:-).
Arrow.com has MC68VZ328's in stock for about $10 each, single quantity.
10:47 a.m.
___________________________________________________________________
TQFP-144...I guess I could figure out how to attach that to my homemade
PCB...something tells me you have way nicer tools than I do.
Vince
Hand soldeing TQFP* is easy and only requires a decent soldering iron, and
lots and lots of flux.. (surface tension make it easy)
Peter Wallace
12:45 p.m.
On Thu, May 13, 2004 at 02:13:05PM -0400, J. David Bryan wrote:
On 12 May 2004 at 7:16, vrs wrote:
I would certainly be in favor of anything 68000-based... I've been writing
68K assembler since 1985, and have a collection of 68K hardware tools
I'm curious what everyone thinks of this
issue, and what *are* the
better choices?
The Motorola 68EC000 makes a reasonable microcontroller. 16 MB addressing
range, configurable 8- or 16-bit data bus, a range of operating speeds (far
faster than a 6809), freely available tools, still in full production, and
about $4.00 in single lots. from my days with Software Results (Northwest
Instruments bus analyzer,
Fluke 9000A w/68K pod...)
I realize that these days the 64-pin DIP package is not in favor, but I
do happen to have several hundred if not a small number of thousand CPUs
in that format. In any case, the instruction set is the real win here,
more than the package.
-ethan
--
Ethan Dicks, A-130-S Current South Pole Weather at 14-May-2004 21:40 Z
South Pole Station
PSC 468 Box 400 Temp -75.5 F (-59.8 C) Windchill -122.5 F (-85.90 C)
APO AP 96598 Wind 10.3 kts Grid 022 Barometer 694.2 mb (10096. ft)
Ethan.Dicks(a)amanda.spole.gov http://penguincentral.com/penguincentral.html
7413
days inactive
7418
days old
31 comments
16 participants
participants (16)
-
ard＠p850ug1.demon.co.uk -
brad＠heeltoe.com -
dickset＠amanda.spole.gov -
dvcorbin＠optonline.net -
GOOI＠oce.nl -
jcwren＠jcwren.com -
jdbryan＠acm.org -
jwest＠classiccmp.org -
marvin＠rain.org -
pat＠computer-refuge.org -
pcw＠mesanet.com -
pkoning＠equallogic.com -
uban＠ubanproductions.com -
vax3900＠yahoo.com -
vrs＠msn.com -
waltje＠pdp11.nl