520st with piggy back ram extension troubleshooting

[sideburn]

Hello all,

 

i just pulled out and fired up my 520ST I had back in the early 90s and to my surprise it booted up!

 

I cracked it open and saw that it’s got the piggy back extra 512K expansion bringing it to 1M. 
 

the problem is the upper expanded bank is not active. It was ONCE and I saw it was all there in sysinfo but it was flakey and the screen was going crazy if touched the board. 
 

so I pulled the Roms and sprayed with deoxit and reseated them and that problem went away but now the machine only sees 512K (238K available). Ram test is passing with no errors. 
 

so I put the oscilloscope on the RAS1 and CAS1L and CAS1H lines and I do see activity on RAS1 but both of the CAS1 lines are stuck high. 
 

on power up I see both CAS1 lines go low real quick and then back to high. 
 

any ideas what could be holding them high?

 

could it be the 74LS latches? Or is all control of those pins handled by the MMU alone making it look like the MMU is bad?

[TGB1718]

I don't see any 74LS latches on those lines, the CAS and RAS come directly from the MMU through 68 ohm resistors, obviously

in this part of the circuit the CAS1 lines are not connected, but when I  did the same piggyback to mine bitd I think I remember

having to put the missing resistors in

 

[sideburn]

10 minutes ago, TGB1718 said:

I don't see any 74LS latches on those lines, the CAS and RAS come directly from the MMU through 68 ohm resistors, obviously

in this part of the circuit the CAS1 lines are not connected, but when I  did the same piggyback to mine bitd I think I remember

having to put the missing resistors in

 

Correct. The resistors are there. RAS1 is oscillating but CAS is stuck high. Trying to figure out why. Who’s in charge of driving the CAS lines low?
 

Also I just noticed the video is good on the monochrome monitor but it’s out of sync on the color monitor. Not sure if that’s related. 

[TGB1718]

I looks like it comes straight from the MMU, so if it's not "wiggling" it likely to be the MMU.

 

Are you checking CAS1 at the MMU itself, just to eliminate things like a broken track etc.

 

[sideburn]

Ah ok that’s what I was suspecting. That’s going to be hard to find. 

[sideburn]

On 1/19/2023 at 8:27 AM, TGB1718 said:

I looks like it comes straight from the MMU, so if it's not "wiggling" it likely to be the MMU.

 

Are you checking CAS1 at the MMU itself, just to eliminate things like a broken track etc.

 

Yes I checked direct of the mmu. 

[sideburn]

On 1/19/2023 at 8:27 AM, TGB1718 said:

I looks like it comes straight from the MMU, so if it's not "wiggling" it likely to be the MMU.

 

Are you checking CAS1 at the MMU itself, just to eliminate things like a broken track etc.

 

I got a new done 520 ST board and swapped all of the socked chips to the problem board and no change.

So The MMU is good. Any ideas what else might be holding CAS1H and CAS1L high?

[snarkdluG]

5 hours ago, sideburn said:

I got a new done 520 ST board and swapped all of the socked chips to the problem board and no change.

So The MMU is good. Any ideas what else might be holding CAS1H and CAS1L high?

The ram itself? 

[sideburn]

21 minutes ago, snarkdluG said:

The ram itself? 

Would that cause the CAS lines to stay high? Any way to figure out which ones are bad? And it’s a full 512 that’s not showing up. Both CAS low and high. 

Edited January 27, 2023 by sideburn

[snarkdluG]

Any of the ram chip getting hotter then normal on the upper bank? Otherwise a diagnostic cartridge might help? 

[ijor]

12 hours ago, sideburn said:

I got a new done 520 ST board and swapped all of the socked chips to the problem board and no change.

So The MMU is good. Any ideas what else might be holding CAS1H and CAS1L high?

 

CAS1 signals are for the second RAM bank, which is not present and not used in the 520ST. That's why are shown as unconnected at the schematics. MMU will not assert those signals if it is configured as a single bank. The few pulses you see at startup is when TOS is trying to sense RAM presence on either bank. If TOS doesn't detect any RAM on the second bank, if will configure MMU for a single bank RAM and CAS1x will never be asserted.

 

Please post a clear picture of your ram expansion.

[sideburn]

4 hours ago, ijor said:

 

CAS1 signals are for the second RAM bank, which is not present and not used in the 520ST. That's why are shown as unconnected at the schematics. MMU will not assert those signals if it is configured as a single bank. The few pulses you see at startup is when TOS is trying to sense RAM presence on either bank. If TOS doesn't detect any RAM on the second bank, if will configure MMU for a single bank RAM and CAS1x will never be asserted.

 

Please post a clear picture of your ram expansion.

Ok so my full expansion showed up once in sysinfo. But my screen was going wonky so i re seated my chips. After reseating the TOS roms, the screen was all good but I lost the ram expansion. 
 

so TOS is not seeing the ram now. 
 

I will post a photo today. 

Edited January 27, 2023 by sideburn

[sideburn]

Here’s some photos of the board and ram expansion

[ijor]

18 hours ago, sideburn said:

Here’s some photos of the board and ram expansion

 

That's not my area of expertise, but sounds like something is not so good with the soldering. If you are good with soldering you might want to redo, or at least to check carefully the wiring and the actual piggy back.

Edited January 28, 2023 by ijor

[sideburn]

50 minutes ago, ijor said:

 

That's not my area of expertise, but sounds like something is not so good with the soldering. If you are good with soldering you might want to redo, or at least to check carefully the wiring and the actual piggy back.

Me neither. Well the soldering all loos good. It’s odd that both the upper and lower bank are not working. I need to figure out what tos does to “see”  the ram. Not sure what CAS / RAS do exactly. I thought CAS were chip selects. 
 

it’s also strange that the full 1mb showed up once. And the. After reseating chips it went away. 
 

[TGB1718]

On 1/28/2023 at 2:52 PM, sideburn said:

It’s odd that both the upper and lower bank are not working.

It's a 16 bit data bus, both have to be working, I would assume if either top or bottom half is not working

the other half will be disabled as @ijor says, if it can't see the RAM (all 16 bits) then it will stop asserting the upper half signals.

[sideburn]

2 hours ago, TGB1718 said:

It's a 16 bit data bus, both have to be working, I would assume if either top or bottom half is not working

the other half will be disabled as @ijor says, if it can't see the RAM (all 16 bits) then it will stop asserting the upper half signals.

I see. Ok son when I’m power up cat e the CAS goes low once and I’m not noticing. Is there any way to find out which RAM chip might be the problem?

[TGB1718]

unless one/any are getting hot to the touch, I'm afraid there's no easy way, my 520 ST had a similar issue

after I piggy backed mine in the same way as yours, I just had to go and re-touch every solder joint, luckily it cured

the problem, maybe that's all that's wrong with yours, but obviously even replacing the chips one at a time

is going to be a real pain and may even damage the currently working ones.