3 Jun
2007
3 Jun
'07
10:10 a.m.
If you are doing a design with on-board regulators in the 21st
century, why not use integrated switching rather than linear
regulators?
Heat dissipation will be MUCH lower if you do this.
3 Jun
3 Jun
10:39 a.m.
Al Kossow wrote:
If you are doing a design with on-board regulators in the 21st
century, why not use integrated switching rather than linear
regulators?
Heat dissipation will be MUCH lower if you do this.
True, but with the advent
modern chips like CMOS I see that
power consumption less of a problem. Is the proto-board
for 8 bit S-100 or 16 bit S-100 as I remember some where the buss
was upgraded for 16 bit processors? ( Can you say INTEL ? )
If it where not for that fact, having De-codeing and buffer chips
laid out as that would save some space rather than in the pro-type
area.
1:15 p.m.
I remember the16bit option too.
That's not hard to add since I was thinking about including the decode
logic to save space.
Thanks for the tip.
=Dan
[ My Corner of Cyberspace http://ragooman.home.comcast.net/ ]
woodelf wrote:
Is the proto-board
for 8 bit S-100 or 16 bit S-100 as I remember some where the buss
was upgraded for 16 bit processors? ( Can you say INTEL ? )
If it where not for that fact, having De-codeing and buffer chips
laid out as that would save some space rather than in the pro-type
area.
11:43 a.m.
On 3 Jun 2007 at 10:10, Al Kossow wrote:
If you are doing a design with on-board regulators in
the 21st
century, why not use integrated switching rather than linear
regulators?
Heat dissipation will be MUCH lower if you do this.
Al, I confess to not having a good handle on this matter of what's
"vintage" and what's not. I was looking at 78xx linear regulators as
being "authentic". I mean, why bother with S-100 at all? There are
plenty of better busses now, right?
I plead ignorance at knowing where to draw the line on this stuff.
Cheers,
Chuck
12:17 p.m.
Chuck Guzis wrote:
On 3 Jun 2007 at 10:10, Al Kossow wrote:
Al, I confess to not having a good handle on this matter of what's
"vintage" and what's not. I was looking at 78xx linear regulators as
being "authentic". I mean, why bother with S-100 at all? There are
plenty of better busses now, right?
I was hoping somebody would ask that. There are really nicer busses, PCB
formats, connectors available now. (VME ? Not talking about the bus, but
the Euro connector & format is nice and cheap)
Cheers
3:51 p.m.
On Jun 3, 2007, at 2:43 PM, Chuck Guzis wrote:
There's nothing "authentic" or even remotely elegant about 78xx
regulators in my opinion. They're a current product, have been for
decades, and probably will be for another two decades.
If you are
doing a design with on-board regulators in the 21st
century, why not use integrated switching rather than linear
regulators?
Heat dissipation will be MUCH lower if you do this.
Al, I confess to not having a good handle on this matter of what's
"vintage" and what's not. I was looking at 78xx linear regulators as
being "authentic". I mean, why bother with S-100 at all? There are
plenty of better busses now, right?
"Better" is subjective...Better for what application? If you're
designing a small system for, say, industrial automation, do you
really want the expense and extreme complexity of something like PCI
or VME? Having done PCI interfacing with an FPGA, I can say that
getting something working on an S-100 bus involves a whole lot less
hair pulling.
-Dave
--
Dave McGuire
Port Charlotte, FL
5:11 p.m.
On 3 Jun 2007 at 18:51, Dave McGuire wrote:
> There's nothing "authentic" or
even remotely elegant about 78xx
regulators in my opinion. They're a current
product, have been for
decades, and probably will be for another two decades.
"Authentic"--yes. They were on almost all original S-100 boards.
Which was my point. There are lots of great regulators now--
particularly the buck/boost models, but they wouldn't be "authentic"
to the S-100 era.
"Better" is subjective...Better for what
application? If you're
designing a small system for, say, industrial automation, do you
really want the expense and extreme complexity of something like PCI
or VME?
At the time of the S-100 bus, there was already a better bus--
Multibus. It outlived S-100 by a very long time.
If you want simple, try STD bus. But then, even ISA/PC-104 will do
just fine for S-100 speeds.
Cheers,
Chuck
11:50 p.m.
On Jun 3, 2007, at 8:11 PM, Chuck Guzis wrote:
Yes, that's why I quoted "authentic". There's nothing (IMO) about
the presence or absence of a 7805 that makes an S-100 board any more
or less authentic...that's the point I was trying to get across.
Also true. I'd bet, however, that the number of S-100 products is
twice that of Multibus and STD bus combined, even when taking into
account stuff like Multibus-based Cisco routers and related stuff. I
wonder why that is...is it just a matter of having been developed
sooner?
-Dave
--
Dave McGuire
Port Charlotte, FL
>
There's nothing "authentic" or even remotely elegant about 78xx
regulators in my opinion. They're a current
product, have been for
decades, and probably will be for another two decades.
"Authentic"--yes. They were on almost all original S-100 boards. Which was my point. There are lots of great
regulators now--
particularly the buck/boost models, but they wouldn't be "authentic"
to the S-100 era.
Very true.
"Better" is subjective...Better for what application? If you're
designing a small system for, say, industrial automation, do you
really want the expense and extreme complexity of something like PCI
or VME?
At the time of the S-100 bus, there was already a better bus--
Multibus. It outlived S-100 by a very long time.
If you want simple, try STD bus. But then, even ISA/PC-104 will do
just fine for S-100 speeds. 
4 Jun
4 Jun
12:28 a.m.
Dave McGuire wrote:
Also true. I'd bet, however, that the number of
S-100 products is
twice that of Multibus and STD bus combined, even when taking into
account stuff like Multibus-based Cisco routers and related stuff. I
wonder why that is...is it just a matter of having been developed sooner?
I don't have any experience from which to draw when trying to come up
with a number for the amount of S-100 product out there, but I can say
without hesitation that there is a metric buttload of STD-bus stuff out
there, but only in specific arenas. You can get an STD-bus board (or
boardset) to do pretty much anything you can think of in industrial
control/manufacturing robotics. It's a very popular bus in that arena.
Peace... Sridhar
5:24 p.m.
On Jun 4, 2007, at 3:28 AM, Sridhar Ayengar wrote:
Yes I've heard that, and it's no coincidence that the three or
four STD-bus cards that I've seen in my life happened to be motor
controllers. :-)
I really like the form factor of those cards...I wish they were
easier to come by; I'd love to build up a little STD-bus "hack
machine" around a Z80 or 6809.
-Dave
--
Dave McGuire
Port Charlotte, FL
Also true.
I'd bet, however, that the number of S-100 products
is twice that of Multibus and STD bus combined, even when taking
into account stuff like Multibus-based Cisco routers and related
stuff. I wonder why that is...is it just a matter of having been
developed sooner?
I don't have any experience from which to draw when trying to come
up with a number for the amount of S-100 product out there, but I
can say without hesitation that there is a metric buttload of STD-
bus stuff out there, but only in specific arenas. You can get an
STD-bus board (or boardset) to do pretty much anything you can
think of in industrial control/manufacturing robotics. It's a very
popular bus in that arena.
7:28 p.m.
Dave McGuire wrote:
On Jun 4, 2007, at 3:28 AM, Sridhar Ayengar wrote:
Yes I've heard that, and it's no coincidence that the three or four
STD-bus cards that I've seen in my life happened to be motor
controllers. :-)
I really like the form factor of those cards...I wish they were easier
to come by; I'd love to build up a little STD-bus "hack machine" around
a Z80 or 6809.
It would probably be easiest to build a "hack machine" around the 6502.
There are a significant number of 6502-based CPU boards for talking to
said motor controllers over the STD-bus. I've Z80-based ones too, but
Also
true. I'd bet, however, that the number of S-100 products is
twice that of Multibus and STD bus combined, even when taking into
account stuff like Multibus-based Cisco routers and related stuff. I
wonder why that is...is it just a matter of having been developed
sooner?
I don't have any experience from which to draw when trying to come up
with a number for the amount of S-100 product out there, but I can say
without hesitation that there is a metric buttload of STD-bus stuff
out there, but only in specific arenas. You can get an STD-bus board
(or boardset) to do pretty much anything you can think of in
industrial control/manufacturing robotics. It's a very popular bus in
that arena. from what I've seen the 6502 ones are more common.
Peace... Sridhar
11:37 a.m.
On 4 Jun 2007 at 2:50, Dave McGuire wrote:
Yes, that's why I quoted "authentic".
There's nothing (IMO) about
the presence or absence of a 7805 that makes an S-100 board any more
or less authentic...that's the point I was trying to get across.
Well, I don't know much about the nuances of this collecting thing.
Does replacement of 74LS logic with, say, 74HCT logic lower the value
of a vintage object?
Also true. I'd bet, however, that the number
of S-100 products is
twice that of Multibus and STD bus combined, even when taking into
account stuff like Multibus-based Cisco routers and related stuff. I
wonder why that is...is it just a matter of having been developed
sooner?
That's an interesting question, considering the variety of boards
possible using the Multibus daughterboard system. Given the long
life of Multibus and its deployment in industrial systems (as well as
some Sun hardware), I wonder if the total production volume of
Multibus boards isn't greater than that of S-100. Yet Multibus sems
to get very little notice from the vintage collectors. Really a
shame, since MB is much more of a well-thought-out design.
At Durango, we had an MDS-800, but we also went "on the cheap" and
purchased a Multibus cage and cards, some floppies and a PSU and
built our own MDS-compatible (wooden) box as a backup. Not nearly as
pretty or expandable, but it booted from the same ISIS-II floppy as
did the MDS.
Cheers,
Chuck
1:52 p.m.
At 10:37 AM 6/4/2007, you wrote:
On 4 Jun 2007 at 2:50, Dave McGuire wrote:
Someone built one of my Altair replica PCBs and went to great lengths to
make sure the date code on the ICs was older than 1975. They also sourced
the old brown cylindrical resistors.
I provide the exact same parts that were on the origional BOM, but don't
try to match date codes! : )
I guess everyone has their own idea of what is vintage enough.
Yes, that's why I quoted
"authentic". There's nothing (IMO) about
the presence or absence of a 7805 that makes an S-100 board any more
or less authentic...that's the point I was trying to get across.
Well, I don't know much about the nuances of this collecting thing.
Does replacement of 74LS logic with, say, 74HCT logic lower the value
of a vintage object?
2:15 p.m.
Grant Stockly wrote:
Someone built one of my Altair replica PCBs and went to great lengths to
make sure the date code on the ICs was older than 1975. They also
sourced the old brown cylindrical resistors.
I provide the exact same parts that were on the
origional BOM, but don't
try to match date codes! : )
I guess everyone has their own idea of what is vintage
enough.
I draw the line at Vintage smoke. Since most people want to run
a classic computer first rather display it I think the stuff on the BOM
is fine.
2:38 p.m.
Grant Stockly wrote:
> Someone built one of my Altair replica PCBs and went to great lengths to
> make sure the date code on the ICs was older than 1975. They also
> sourced the old brown cylindrical resistors.
I've been known to do that, but for much more trivial examples - when
I built my 1976-design Elf, I did the best I could on chips (my CPU
and memory were of the period, but I have not yet tracked down a 74L00
for the clock divider circuit), but I was happy to have a bin of
vintage brown cylindrical resistors for the pullups. I had to
compromise on the regulator - the oldest one I could find in my junk
bin was an LM340T.
Obviously, the computer doesn't care, but I wanted something that
would have looked as if I had made it when I first got into computers
as a kid. A complete indulgence to nostalgia.
-ethan
4:48 p.m.
On Jun 4, 2007, at 2:37 PM, Chuck Guzis wrote:
I guess that depends on the collector. I still have a hard time
considering S-100 stuff "collectible" because every piece of S-100
gear I own is something I've had since it was current technology. I
thought, though, we were originally talking about *brand new* S-100
boards.
Agreed. But it just doesn't seem to be anywhere near as common.
Very odd.
Yes,
that's why I quoted "authentic". There's nothing (IMO) about
the presence or absence of a 7805 that makes an S-100 board any more
or less authentic...that's the point I was trying to get across.
Well, I don't know much about the nuances of this collecting thing.
Does replacement of 74LS logic with, say, 74HCT logic lower the value
of a vintage object? Also true.
I'd bet, however, that the number of S-100 products is
twice that of Multibus and STD bus combined, even when taking into
account stuff like Multibus-based Cisco routers and related stuff. I
wonder why that is...is it just a matter of having been developed
sooner?
That's an interesting question, considering the variety of boards
possible using the Multibus daughterboard system. Given the long
life of Multibus and its deployment in industrial systems (as well as
some Sun hardware), I wonder if the total production volume of
Multibus boards isn't greater than that of S-100. Yet Multibus sems
to get very little notice from the vintage collectors. Really a
shame, since MB is much more of a well-thought-out design. At Durango, we had an MDS-800, but we also went
"on the cheap" and
purchased a Multibus cage and cards, some floppies and a PSU and
built our own MDS-compatible (wooden) box as a backup. Not nearly as
pretty or expandable, but it booted from the same ISIS-II floppy as
did the MDS.
Oh that sounds cool. :)
-Dave
--
Dave McGuire
Port Charlotte, FL
9:02 p.m.
On 4 Jun 2007 at 19:48, Dave McGuire wrote:
Oh that sounds cool. :)
Wood was very "in" for prototyping back then. :) Looked even funkier
if you plastered wood-grain vinyl shelf paper over it.
Anyone remember the Atari Race Car game at Andy Capp's around 1976?
IIRC wasn't it just brown-stained plywood?
Cheers,
Chuck
3 Jun
3 Jun
1:13 p.m.
On such a small board, I'm not sure it's practical yet. The pcb real
estate + parts cost vs. the wattage might not be cost effective. The
LM2940 series regulator is a basic yet improved with minimum overhead.
When you need to convert down to something such as 3.3v, 2.5v or 1.8v,
an onboard switcher will help. If there's any switching power supply
being used, it would be sitting at the bottom of the backplane rather
than on each board. It's more cost effective this way. Then you can just
remove the onboard regulator and bypass it with a jumper wire.
=Dan
[ My Corner of Cyberspace http://ragooman.home.comcast.net/ ]
Al Kossow wrote:
If you are doing a design with on-board regulators in the 21st
century, why not use integrated switching rather than linear
regulators?
Heat dissipation will be MUCH lower if you do this.
6410
days inactive
6412
days old
17 comments
9 participants
participants (9)
-
aek＠bitsavers.org -
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
emu＠e-bbes.com -
ethan.dicks＠gmail.com -
grant＠stockly.com -
mcguire＠neurotica.com -
ploopster＠gmail.com -
ragooman＠comcast.net