25 Jan
1999
25 Jan
'99
4:50 a.m.
List:
Last night we found, in of all places the attic,
some missing literature - my given up for lost original
Altair BASIC Referance manual, complete with the little
MITS tty test printout ( ?SIN(3.14159/6) \ .5 \ ok ), an
Atari BASIC quick referance guide, ad for an Atari MPP
modem, and most interesting "Bugbook III", 1975 which
discusses a MARK 80 (8080) Microdesigner of E&L instruments.
This is somewhat interesting as it consists of a backplane
with a breadboard area complete with bus connections for
interfacing experiments. Bus is called SK-10 and cards plug
into dual wide DEC edge connectors.
Anybody have a MARK 80? Is it any relation to the 8008
Mark-8?
Chuck
cswiger(a)widomaker.com
25 Jan
25 Jan
9:03 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 970 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990125/7d4f505c/at…
11:09 a.m.
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
Alas
the ROM is missing. Some kind person here sent me the source/hex
dump of it, so one day I'll get round to burning a 1702...
-tony
Hi I'm looking fro 1702's, not 1702A's which are quite common
but 1702's. You wouldn't happen to have any of these would
you. I'll trade you 1702A's for any you might have.
Dwight
12:10 p.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 719 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990125/6e94b23b/at…
1:06 p.m.
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
What's the real difference between a 1702 and a 1702A? I must admit that
I am not a great one for historical correctness, so I tend to use
'updated' parts if I happen to have them to hand
> Dwight
-tony
Hi Tony
1702's use a different programming algorithm than the
1702A's. Otherwise they work the same. Only a few
1702 were made, making them quite rare. Most have been
melted down for the gold that Intel used in
early packaging.
I'm looking for the non-A types because I have an
old 4004 development system that has a programming
board on it. Of course, the board was designed to do
1702's and not 1702A's. It had a selector to select
another PROM type that looked like it would work but
the signal levels still were not correct. I suspect that
they intend it to work but didn't understand the changes
the chip designers were up to. It does part of the needed
changes but not all.
Thanks
Dwight
1:19 p.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 349 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990126/74bfecea/at…
1:49 p.m.
At 11:10 PM 1/25/99 +0000, Tony wrote:
The 1702's had 10% of the program duty cycle of the 1702A, so took 10x
longer to
program. The voltages were the same. I don't know what actually was revised on
the chip, larger output/program transistors?
-Dave
Hi I'm looking fro 1702's, not
1702A's which are quite common
but 1702's. You wouldn't happen to have any of these would
you. I'll trade you 1702A's for any you might have.
What's the real difference between a 1702 and a 1702A? I must admit that
I am not a great one for historical correctness, so I tend to use
'updated' parts if I happen to have them to hand
26 Jan
26 Jan
7:10 a.m.
New subject: 1702A programming (was Bugbooks - Mark 80 )
Dave Dameron <ddameron(a)earthlink.net> wrote:
At 11:10 PM 1/25/99 +0000, Tony wrote:
The 1702's had 10% of the program duty cycle of the 1702A, so took 10x
longer to
program. The voltages were the same. I don't know what actually was revised on
the chip, larger output/program transistors?
-Dave
Hi Dave
The 1702A's also required that the address be complemented before
the Vdd and Vgg were brought low. The 1702's didn't seem to need
this and Vdd and Vgg were left at there low states the entire time.
This also is a problem because the programmer I have doesn't
switch Vdd and Vgg.
Dwight
Hi I'm looking fro 1702's, not
1702A's which are quite common
but 1702's. You wouldn't happen to have any of these would
you. I'll trade you 1702A's for any you might have.
What's the real difference between a 1702 and a 1702A? I must admit that
I am not a great one for historical correctness, so I tend to use
'updated' parts if I happen to have them to hand
2:07 p.m.
New subject: 1702A programming (was Bugbooks - Mark 80 )
At 10:10 AM 1/26/99 -0800, Dwight wrote:
Thanks,
The Intel booklet I have has a 1702A programmer, and it implies it is
compatable with the 1702 with a lower duty cycle. So the 1702 _could_ work
with
the address complement step and switched Vdd, Vgg, but the 1702A requires
these.
I looked in my collection and found only 1702A's. Have been thinking about
building a programmer for when I ever build a 8008 system, but the Intel
one is fairly involved. There was a simpler one in the Feb. 1978 issue of
Popular Electronics. Sone things I prefer about the second, it used a
counter and shift register to control the timing instead of monostables.
-Dave
Dave Dameron <ddameron(a)earthlink.net> wrote:
revised
on
At 11:10 PM 1/25/99 +0000, Tony wrote:
> The 1702's had 10% of the program duty cycle of the 1702A, so took 10x
> longer to
> program. The voltages were the same. I don't know what actually was the chip,
larger output/program transistors?
-Dave
Hi Dave
The 1702A's also required that the address be complemented before
the Vdd and Vgg were brought low. The 1702's didn't seem to need
this and Vdd and Vgg were left at there low states the entire time.
This also is a problem because the programmer I have doesn't
switch Vdd and Vgg.
Dwight
27 Jan
27 Jan
12:54 p.m.
New subject: 1702A programming (was Bugbooks - Mark 80 )
Programming the 1702A's
Well in some sence I have a little egg on my face.
The programmer I have was infact designed for
1702A's but doesn't do it following
the specifications.
First 1702 versus 1702A's. As was mentioned, the
1702's couldn't take the same duty cycle by 10X.
It isn't true that you can take a 1702 programmer
and program 1702A but you can take a 1702A programmer
and slow down the pulse rate and program 1702's.
The difference is that the 1702A's require an address
complement phase at the start of the cycle. This phase
doesn't effect the 1702's so the only issue is repeat
rate.
Now to the issue with my MP7-03 board. The first PROM
I tried was either originally blown or I blew it out.
The programming voltage was about 53 volts. I cranked
it down to 47 volts and tried another PROM and it now
works fine.
There is still the part I don't understand. The specifications
for programming the 1702A show a setup time for the
complement address at the start of 25 microseconds. The
MP7-03 is violating this by quite a bit. The Vgg
and Vdd are never pulsed to Vcc. The sequence is:
step0: time before zero
Vcc 5V
Prog 5V
Vgg -9v
Vbb 5V
Vdd -9v
address lines true and 0/TTL levels
Data lines true and 0/TTL levels
step1: time 0
Vdd to Vcc-47volts
Prog to Vcc
Vgg to Vcc-35volts
Vbb to Vcc+12volts
Address lines to Vcc and Vdd levels but complemented
Data lines true and Vdd/Vcc levels
step2: time 50 microseconds
Adrress lines to true, Vdd/Vcc levels
step3: time 150 microseconds
Prog to Vdd
step4: time 2.5 milliseconds
Prog to Vcc
step5: time 2.9 milliseconds
Levels to the same as Step0
Repeat every 15 milliseconds for 30 cycles.
If you'll plot this out, you'll see that there is zero
complement address setup time but the transition from
14 volts Vcc-Vdd to 47 volts Vcc-Vdd may compensate for
this. Without understanding the internals, I couldn't say
but it is not like the specifications.
Part of the reason I'm posting the sequence used for this
tester is that it is a little simpler than following the exact
specification and may help some home made programmer. I
also ran some AMD 1702A's through it without problems.
The good news is that I was able to program three backup
PROMs for my SIM4-01 board and I ran them to verify that
they where correct.
Thanks for the help from Jonathan Levine who sent me a copy
of the document that made me realize that the firmware and
hardware should program 1702A's even if the specs didn't match.
Dwight
2:25 p.m.
New subject: 1702A programming (was Bugbooks - Mark 80 )
HI Dwight and all,
At 03:54 PM 1/27/99 -0800, Dwight wrote:
Programming the 1702A's
Well in some sence I have a little egg on my face.
The programmer I have was infact designed for
1702A's but doesn't do it following
the specifications.
First 1702 versus 1702A's. As was mentioned, the
1702's couldn't take the same duty cycle by 10X.
It isn't true that you can take a 1702 programmer
and program 1702A but you can take a 1702A programmer
and slow down the pulse rate and program 1702's.
The difference is that the 1702A's require an address
complement phase at the start of the cycle. This phase
doesn't effect the 1702's so the only issue is repeat
rate.
Now to the issue with my MP7-03 board. The first PROM
I tried was either originally blown or I blew it out.
The programming voltage was about 53 volts. I cranked
it down to 47 volts and tried another PROM and it now
works fine.
Cool that you can program them.
I looked at my manual and it was for the MP7-03 too. In my schematic, there
are 7486 gates in the 8 address lines to invert the data. They are inverted
for 60 microseconds by a monostable. The signal goes from (edge connector)
pin #7 to the 7486's connected to pin #39. Does anyonne know for what the
inversion of
the addresses is required?
Have not traced all the timing as you have to see the resultant setup
times, etc. Of course the MP7-03 uses positive voltages, as the voltage
reference point is different than the data sheets. Am nor sure if I figured
out the logic levels. You program a "1" with a negative voltage (data
sheet) = positive in the MP7-03. I guess if the programmed data or
addresses come out complemented, you got those levels inverted, and you do
it again.
It looks also that you can program 1 byte at a time (32 cycles). In the
later 2708, the data sheets says you have to cycle through all locations.
Does anyone know if this is for the thermal duty cycles of the address
lines? This restriction was removed for the later 2716.
-Dave
25 Jan
25 Jan
1:41 p.m.
At 08:03 PM 1/25/99 +0000, you wrote:
It is described very well in the April, May 1976 issues of "Radio
Electronics", including circuit description at the IC/gate level, monitor
source, etc.
It doesn't use a 8228, but bus drivers, 1702A's (1 monitor, 1 spare), and
256x4 RAM's (2 expandable to 4). There was a later one with an expansion
box (more RAM, modem/cassette interface), but I didn't buy one, the price
was too high for me US $125.
-Dave
modem, and
most interesting "Bugbook III", 1975 which
discusses a MARK 80 (8080) Microdesigner of E&L instruments.
I think I found that book (or maybe another one in the same series) in a
second-hand bookshop a couple of years back. It's interesting
historically, but not that useful as (IMHO) it doesn't give enough
details of the MARK-80 cards.
This is somewhat interesting as it consists of a
backplane
with a breadboard area complete with bus connections for
interfacing experiments. Bus is called SK-10 and cards plug
into dual wide DEC edge connectors.
I found a simpler E&L machine at a radio rally last year. This was a
single-board (which formed the top of a slope-top case containing the
PSU) with an 8080, RAM, ROM, keypad, some output ports/LEDs and a
breadboard on it. I grabbed it, of course.
Alas the ROM is missing. Some kind person here sent me the source/hex
dump of it, so one day I'll get round to burning a 1702...
9313
days inactive
9315
days old
11 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
cswiger＠wilma.widomaker.com -
ddameron＠earthlink.net -
elvey＠hal.com