Electronics > Repair
ADVANTEST R9211B
<< < (12/39) > >>
stfjohn:
Be patient: to do all these measurements, I have to reassemble the motherboard on the chassis each time and can therefore only measure from underneath the printed circuit board, soldering pin strips onto the components to connect the logic analyzer! All this soldering and unsoldering will soon lead to me having to throw everything away! However, I can keep the analyzer (an old H.P. 1651A) for a while longer. Let's see what I can do....
m k:
Dang!
But props for the commitment.

I'm using UTP cat5 Ethernet wall cable wires for many things, it's a single core thing.

Good for the analyzer, and not a simple one.
So 32 channels, I think you can have all necessary info from a single connection set.

I'm putting stuff up so that you don't need to roll your thumbs.
Or to do too many wrong things after wrong tips, so no hurry.

So 32 channels,
first 16 pins from #53.

Next 82C79 _CS and IRQ pins 22 and 4, to check that access actually goes there.
Next 68450 _CS, _IRQ and _BEC0-2 pins 17, 21 and 29-27.
Next _CS pin 31 of all 3 timer chips.

So 2 + 5 + 3 more, six to go.

Next 32801 _SPC, ST0 and ST1, pins 21-23, for a sanity check, maybe it's sending its own.

Next CPU CLK, _HALT and _RESET, pins 15, 17 and 18.
Halt and reset can be dropped out when it's clear that they are not obstructing anything.
You can also use a POD clock of course.

Timing.
CPU clock first, you can add more triggers when you know what is happening.
For example if interrupt happens and it is cleared away you can skip everything up to that point.
Page 8-12 of the Reference Manual has an example.

I'm not familiar with the analyzer, must read some.
Maybe somebody around can give some insights.
stfjohn:
Here are the scans in the exact sequence you requested in your last post. (Trigger A22)
Photo 1: pod zero pin 53; pod 1 pin 52 etc. to follow.
Photo 2: pod zero pin 22 (82C79) etc. to follow according to the exact sequence of your last post.
m k:
Sorry, my bad, #53 was reply #53.
But no panic, plenty of info from 53 and down.
It can be even better than the intended one.

Second picture POD 2 00 and 01 are ok.
CPU _HALT and _RESET, POD 2 10 and 11 are also ok.

But I'm not sure what are POD 2 02-08.
There should be 5 pins from DMAC, 3 pins from time chips and 3 pins from FPU.
So CPU CLK should be POD 2 13.

Those 3 pins above clock are not very timer like, so my guess is 32081 FPU.
But no panic here either, we can put that aside.

POD 1 00, is constantly down and 04 is constantly up, so GND and Vcc.
Means that 01 is CPU A23 and flipping, so address is at least 8x xx xx.

POD 1 binary addresses from the left.
0000 0000 1001 00 = 0090
0100 0000 1001 00 = 4090 short
0100 0000 1001 00 = 4090 short
1100 1100 1101 11 = CCD3
1100 1100 1101 11 = CCD3
and repeating

Short pulses can be something else.
So what option is that CC?

Then my second bad, those FPU pins are not accurate, I mixed ST0 and ST1.
Did you use name or number order?

Correct is _SPC 21, ST1 22 and ST0 23.

But anyway, if CPU is accessing that FPU and I/O map is correct something is not right.
If the FPU is doing its own the earlier CC address can be a result from that and so nothing real.
Can you easily remove that FPU chip in one piece?

If you have a real solder sucker in hand you may also try disconnecting the _SPC pin only.
It happens so that you suck the through hole empty and position the pin in the middle of the empty hole.
You can of course use some insulation aids, but that disconnecting used to be pretty easy back then.
Just push the pin with something towards its body and if its hole is clean enough the pin is disconnected in no time.

You seem to have POD 2 12-15 free.
Maybe FPU has something, so connect its D0, D2, D6 and D7, pins 11, 9, 5 and 4 to free POD pins.
If status D0 is set there is an error.

You can also trace _SPC signal backwards.
It goes through something before it reaches the CPU.
stfjohn:
Last scan:
Pod 0  on p. 22 (82C79)
Pod 1  on p. 4 (82C79)
Pod 2  on p. 17 (68450)
Pod 3  on p. 21 (68450)
Pod 4  on p. 29 (68450)
Pod 5  on p. 27 (68450)
Pod 6  on p. 31 (68450)
Pod 7  on p. 19 (32081)
Pod 8  on p. 20 (32081)
Pod 9  on p. 21 (32081)
Pod 10  on p. 17 (CPU)
Pod 11  on p. 18 (CPU)
Pod 12  on p. 11 (32081)
Pod 13  on p. 9 (32081)
Pod 14  on p. 5 (32081)
Pod 15  on p. 4 (32081)
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod