SBC FERGUSON BIGBOARD 1 - Help Request
Enrico Lazzerini
i�m writing from italy. Please excuse me for my english form.
Since 2007 it�s grow for me a passion for my SBC in object. During this time
I collected many info I reported on my webpage
http://elazzerini.interfree.it
I�d need support to repair my bigboard 1 so the question is: Are there
people that had or have this board or could be help me just for their
electronic knowledges ?
Below follow my actual situation.:
1st - i give my situation:
I have cognitions on electronic left 30 years ago on TTL and CMOS locic ICs.
I have a good power supply, a multimeter and a dual trace scope. I have a
solder. The board slepts for almost 20 years in a box far away water or
humidity. I lost ist original schematic but i downloaded similar fron Dave
Dunfield webste. I already pulled away all chips and checked every wire from
the schematics to the PCB board with my multimeter. Every wire is correct. I
not tested any resistor and capacitor. The board originally not had
Z80-SIO/0 neither 1488/1489 chips and it worked with a parallel keyboard.
NOW I not have a paralled keyboard and, after insert a z80DART (it'd be the
same of z80 SIO/0) and the 1488/1489 chips, i anyway tryed to connect a PC
with hyperterminal program to the board verifing that pressing a key on the
PC the charather gone to the pins 28/29 of the z80 dart. The manual
describes that the board have to do almost a clear screen after its reset,
then waiting for a CR ascii code from PIO or SIO, but the video not change
neither have a clear screen. I have not any IC Tester to check TTL or any
2114 SRAM tested chips .
2nd - what happends at power on:
On the screen appears a matrix of 24 x 80 charaters that stay fixed without
any blink.
The video clock and the cpu clock are exactly as described in the
documentation.
I already changed the set of Z80 chips (with others) and the entire 4x8=32
4116 chip (those ones tested from a friend on his apple board). The reset
button NOT DO nothing (while i see with the scope that something change at
the RST pin of the Z80 CPU!).
Thanks for any answer and help.
Regards
Enrico � Pisa (ITALY)
Mensanator
> Hello,
> i’m writing from italy. Please excuse me for my english form.
> Since 2007 it’s grow for me a passion for my SBC in object. During this time
> I’d need support to repair my bigboard 1 so the question is: Are there
> people that had or have this board or could be help me just for their
> electronic knowledges ?
Kind of hard to do by correspondance in a newsgroup.
>
> Below follow my actual situation.:
>
> 1st - i give my situation:
> I have cognitions on electronic left 30 years ago on TTL and CMOS locic ICs.
> I have a good power supply, a multimeter and a dual trace scope. I have a
> solder. The board slepts for almost 20 years in a box far away water or
> humidity.
But you almost certainly did not store it away from air.
The pictures show that board is socketed. Some sockets
are high precision (I can see from the round holes),
but I can't tell if all are. Sockets have a tendency to
corrode from air polution. I have a couple Apple ][ in
boxes and I doubt either would work. It may be just that
each chip needs reseating.
> I lost ist original schematic but i downloaded similar fron Dave
> Dunfield webste. I already pulled away all chips and checked every wire from
> the schematics to the PCB board with my multimeter. Every wire is correct.
Unlikely to be the source of the problem, although it
might be useful to know if the board ever worked in the
past. If so, and no evidence of water corrosion, the wiring
should be ok. Hopefully, you didn't damage any sockets with
the probe tips.
> I
> not tested any resistor and capacitor.
You shouldn't need to.
> The board originally not had
> Z80-SIO/0 neither 1488/1489 chips and it worked with a parallel keyboard.
> NOW I not have a paralled keyboard and, after insert a z80DART (it'd be the
> same of z80 SIO/0) and the 1488/1489 chips, i anyway tryed to connect a PC
> with hyperterminal program to the board verifing that pressing a key on the
> PC the charather gone to the pins 28/29 of the z80 dart.
The Z80 must be fully functional before you can talk to it
via an external device.
> The manual
> describes that the board have to do almost a clear screen after its reset,
That is surely a function of the monitor program that
kicks in at reset. The fact that the screen does not
clear is evidence that the monitor is not running.
> then waiting for a CR ascii code from PIO or SIO,
If it's not clearing the screen, it's not waiting for a CR.
> but the video not change
> neither have a clear screen.
That means the video generation circuit is functional,
but it's just a pinball machine, it just blindly dumps
the contents of the video RAM to the screen. RAM chip
contents are usually random at powerup, that's probably
what you're seeing.
> I have not any IC Tester to check TTL or any
> 2114 SRAM tested chips .
You'll have to infer whether the individual chips are
working by checking them with you scope. If a NAND gate
has both inputs high, the output should be low. And you
generally always want touch the probe to the actual pin
of the chip as this can reveal a bad socket contact (you
may see a floating value whereas the same point from the
origin looks ok. Sometimes merely touching the pin makes
it start working again.
>
> 2nd - what happends at power on:
> On the screen appears a matrix of 24 x 80 charaters that stay fixed without
> any blink.
> The video clock and the cpu clock are exactly as described in the
> documentation.
Sounds like this portion of the circuitry is functional,
but it's all mechanical, no cpu required for this.
> I already changed the set of Z80 chips (with others) and the entire 4x8=32
> 4116 chip (those ones tested from a friend on his apple board). The reset
> button NOT DO nothing (while i see with the scope that something change at
> the RST pin of the Z80 CPU!).
I think a Z80 jumps to location 0 on reset, then starts
exectuting the PROM. What you're seeing is the start of
this process. My guess is that it works up until the point
where it executes it's first RTS. When a CPU pushes
something onto the stack (part of the main RAM), it has
no way of knowing whether the RAM actually recorded it.
It won't know until it does an RTS and pulls the same kind
of random garbage you're seeing on the screen. And since
that is supposed to be a return address, the program jumps
to some random place in memory.
Now, it's quite possible that neither the CPU nor the RAM
chips are at fault. There are myriad clock signals involved
that use a lot of gates. I had this exact situation on
an Apple ][. Here, the write strobe went through a NAND
gate (gated by some other signal) before actually hitting
the appropriate RAM chip pin. The gate turned out to have
failed when being transported. No data was ever clocked
into RAM, so the monitor ran up to the point whwere it did
its first RTS.
How do you find this with just a dual channel scope? You
take a small breadboard and rig up a 555 timer with a
potentiometer controlled period that can range from 1 to
a hundred or so clock cycles. You attach this to the reset
pin and slowly adjust the period. Triggering the scope on
reset, you monitor the address lines until you see them go
blurry. That will probably correspond to an RTS. Of course,
you'll also want to check the data bus and verify that
what comes out of the PROM is the actual code that the Z80
is supposed to be executing matches your source code.
Another trick is to make an PROM that contains nothing but
noop instrucions except at the very end where it does a
jump to 0. This should loop continuously with no RAM usage,
allowing you to check all the clock signals. And you have
to check every gate involved at both te signal's origin
and destination verifying logic and watching for floating
inputs. Of course, such a noop loop won't write anything,
but you can always add a few writes, just don't depend on
the contents being what you wrote.
>
> Thanks for any answer and help.
Good luck tracking this down.
>
> Regards
> Enrico – Pisa (ITALY)