Help resolving STE Booster instability?

[IngoQ]

Hiya,

I'm in the process of fitting my STE Booster and it seems like the machine is quite instable, when switching to 32MHz. I am trying to figure out a scheme, but I have the gut feeling that it has to do with the 32MHz wire to U405. When I change the position of the wire, the stability improves slightly.

I tried using a shielded cable, but it made the instability worse, for some reason...

In addition, when swapping out the SIMM modules, stability again improved by a margin. But the machine still runs for a couple of minutes, and when starting GB6 it randomly crashes after some tests

I double, tripple and quadruple checked every connection and solder joint, and am a little lost here... Did anyone encounter something like this, and has a hint?

[IngoQ]

To further harden my theory about the 32MHz cable:

I temporarily "routed" the cable over the modulator and then downwards right hand of the SIMs. The STE just completed 30 full GB6 cycles in 32MHz without crashing...

So what do I do now? Using a shielded cable? What type of cable? Any ideas?

[exxos]

U405 what is the exact number ? You need the F version, not S.

Also you need 8 chip simms, not parity 9 chips.

as for cable, just use the thin black cable i supply in the kit, and route it from the ic downwards, then across the front of the simms to the booster.

Only time I saw issues with moving the wire was when using parity simms. Other wire locations just caused video interference.

[IngoQ]

U405 what is the exact number ? You need the F version, not S.

Yep, checked that before I began, it's the 74F257.

as for cable, just use the thin black cable i supply in the kit, and route it from the ic downwards, then across the front of the simms to the booster.

Nope, no cable in my kit, but I have plenty of thin black wire here, so no issue. I used that first, and then (out of desparation) tried a thicker one, which made it worse. Thicker antenna, I guess

Also you need 8 chip simms, not parity 9 chips.
Only time I saw issues with moving the wire was when using parity simms. Other wire locations just caused video interference.

Yep, that's it .
I remember the discussion about parity simms, and whethere they have an effect. Turns out that both my sets are partity (4MB with 9 modules, 2MB with 3 modules). I guess this settles the debate then

I only have some EDO-Rams lying around, so no quick fix then. Luckily I know an excellent store, that has newly made SIMMs on stock....that I just ordered

While waiting on these, is there a temporary way to convert parity to none parity modules? Like cutting traces on Pins 26, 28 and 29 or removing one module?

[exxos]

Yep, that's it .
I remember the discussion about parity simms, and whethere they have an effect. Turns out that both my sets are partity (4MB with 9 modules, 2MB with 3 modules). I guess this settles the debate then

IIRC "rubber_jonnie" also had similar issues.. Yes, I did tell people "elsewhere" about such issues..

While waiting on these, is there a temporary way to convert parity to none parity modules? Like cutting traces on Pins 26, 28 and 29 or removing one module?

You can remove the 9th chip, think its pin 30 end, just heat gun it out...

[exxos]

Actually..

[IngoQ]

You can remove the 9th chip, think its pin 30 end, just heat gun it out...

Tried that, but not much change. It still gets instable to the point of not booting, when the cable comes too close to the board on pretty much any spot. Only when I have it at least 5 cm away, it works.

I'll wait for the SIMMs to be sure. If the problem still persists, I will have to think about some sort of shielding... maybe a coax cable? I am wondering, if I HAVE to connect the shield to ground...

[exxos]

Tried that, but not much change. It still gets instable to the point of not booting, when the cable comes too close to the board on pretty much any spot. Only when I have it at least 5 cm away, it works.

I only saw that with parity simms. Basically something else must be floating somewhere.

I'll wait for the SIMMs to be sure. If the problem still persists, I will have to think about some sort of shielding...

Yes probably the best idea for the moment.

maybe a coax cable? I am wondering, if I HAVE to connect the shield to ground...

No, dont screen the cable. a few pF on a long cable at 32mhz will drop the 32mhz down to 1 volt and cause massive ringing. You shouldn't need to go to those lengths anyway. something is at fault somewhere which is only showing when the 32mhz line is close to the fault. The 32mhz line isn't at fault (unless you can scope it and see massive ringing)

I have tried all routes with the 32mhz line, all over the board, and never seen any stability issues, only with parity simms. I tried 40 of them and the fault was replicated and solved on every simm.

If your PSU recapped ? One possibility. Or maybe you have some simms which still do not like the 32mhz wire near them for some reason. What speed are they ?

Also assume you have my dual tos board and connected the CE cable correctly to the ROM board ? Have you tried 8mhz mode to see if its stable ?

I would suggest if the new simms don't solve the issue, you send me the motherboard to diagnose.

[rubber_jonnie]

I did have some strange memory issues, but luckily I had a big bag o' ram from back in the old days, and found some suitable 8 chip SIMMs that worked fine.

In terms of the 32Mhz wire, I came pretty much direct from U405, across the top right of the SIMM sockets (You should be able to see in the picture), and direct to the Booster, using the wire supplied with the kit, but keeping it as short as possible.

It was only the SIMMs that caused trouble in my case though, as once swapped out for a different set, it was all stable.

IMG_20160802_223606.jpg (4.55 MiB) Viewed 6363 times

[IngoQ]

No, dont screen the cable. a few pF on a long cable at 32mhz will drop the 32mhz down to 1 volt and cause massive ringing. You shouldn't need to go to those lengths anyway. something is at fault somewhere which is only showing when the 32mhz line is close to the fault. The 32mhz line isn't at fault (unless you can scope it and see massive ringing)

I tried using a shielded network cable, results were better, but the issue remained. So I switched back to thin black wire Still have no scope, so no idea, how the signal looks like...

If your PSU recapped ? One possibility.

It's a DVE DSP-508A, modified to your specs.

Or maybe you have some simms which still do not like the 32mhz wire near them for some reason. What speed are they ?

Yeah, most likely. I guess the remaining traces still act like an antenna. This would be conclusive with the fact, that stability improves, when I only use two of the modified SIMMs, and again when I use 3 module SIMMs (with parity), which is the most stable. I guess, as soon as I have the non parity SIMMs, we'll know.

Also assume you have my dual tos board and connected the CE cable correctly to the ROM board ? Have you tried 8mhz mode to see if its stable ?

Yes, and yes. 8 MHz is perfectly fine.

I would suggest if the new simms don't solve the issue, you send me the motherboard to diagnose.

Thanks for the offer

There is another possibility, that my CPU-socket construction, which exposes CPU lines vertically for about 5-8mm will be of fault here. When I can rule out the SIMMs I will have to investigate this. It turned out way to high anyways, so I will have to start over with that.