Image instability with STE booster and system memory over 512k

[DementED]

Maybe it's this problem again viewtopic.php?p=80133#p80133

I think it could definitely be a variation of that problem. @exxos do you have any more of those buffer boards in stock?

[rubber_jonnie]

Personally I think your clock wire is too long and it's also going past the main power input, the large capacitor and numerous other components.

I've installed two boosters now, and run 4MB of RAM in both machines, and I don't have any issues, but my clock wire is the original in both cases and runs the shortest route possible.

I'll try and find a picture, or take a new one.

Thank you. Looking forward to seeing it. Just to confirm, I'm using the clock wire that came with the Exxos kit. I tried running it exactly as shown in the post with installation instructions here on the forum (i.e. A bit closer to the SIMMs) but this made the interference worse.

It's got to be something to do with noise on the clock line. All I can suggest is perhaps source some micro-coax which has shielding, solder one end of the shielding to ground leave the other end unsoldered. Then run it the same as you have before, or try even more directly since the shielding should absorb any electrical noise.

I'll try to find some micro_coax cable. @exxos is there any that would be approved for use here? You mentioned before that different wires create a different kit and would be an experiment.

Thanks again.

I have to say that looks nothing like the clock wire I have. I'll try to get a pic and post it.

[Steve]

@rubber_jonnie are you looking at the tubing he put the wire inside I guess?

[rubber_jonnie]

@rubber_jonnie are you looking at the tubing he put the wire inside I guess?

Yes possibly, but it's still very long and goes via a lot of components, would be better shorter.

[exxos]

Maybe it's this problem again viewtopic.php?p=80133#p80133

I think it could definitely be a variation of that problem. @exxos do you have any more of those buffer boards in stock?

I think I saw them the other day. They never been sold officially as it was just a beta test fix. It's hard to say how widespread problems are as not many do tests and try fixes out.

Don't mess with the 32mhz wire. You need to use the one which came with the kit. Using any other wire will change its capacitance and impeedence and will cause all sorts of new issues.

Keeping the wire away from the ram is the main thing. But it's still unknown why that even causes problems in the first place. I know parity Simms cause all sorts of problems. I can only assume that coupling on the 32mhz wire causes random GND bounce somewhere which skews timings just enough to cause huge malfunctions. That's known as even a X10 probe is enough to cause chaos.

It's possible why changing the RAM amount is have such changes as it's changing the current though the grounds. A few pF in the wrong place and a change of just 1ns is enough to break timings. The STE seems to be gaining traction for being the worst ST these days !

[DementED]

I think it could definitely be a variation of that problem. @exxos do you have any more of those buffer boards in stock?

I think I saw them the other day. They never been sold officially as it was just a beta test fix. It's hard to say how widespread problems are as not many do tests and try fixes out.

Don't mess with the 32mhz wire. You need to use the one which came with the kit. Using any other wire will change its capacitance and impeedence and will cause all sorts of new issues.

Keeping the wire away from the ram is the main thing. But it's still unknown why that even causes problems in the first place. I know parity Simms cause all sorts of problems. I can only assume that coupling on the 32mhz wire causes random GND bounce somewhere which skews timings just enough to cause huge malfunctions. That's known as even a X10 probe is enough to cause chaos.

It's possible why changing the RAM amount is have such changes as it's changing the current though the grounds. A few pF in the wrong place and a change of just 1ns is enough to break timings. The STE seems to be gaining traction for being the worst ST these days !

Thank you @exxos . Please let me know if you do find them and I'll happily pay your costs for one. I'm using non-parity SIMMs.

It's annoying because I have 2 later model STEs that work perfectly (but they don't have the socketed CPUs).

@rubber_jonnie the purple tubing is the shroud. I don't want the wire to go any closer to the RAM as that causes more problems.

[rubber_jonnie]

I think I saw them the other day. They never been sold officially as it was just a beta test fix. It's hard to say how widespread problems are as not many do tests and try fixes out.

Don't mess with the 32mhz wire. You need to use the one which came with the kit. Using any other wire will change its capacitance and impeedence and will cause all sorts of new issues.

Keeping the wire away from the ram is the main thing. But it's still unknown why that even causes problems in the first place. I know parity Simms cause all sorts of problems. I can only assume that coupling on the 32mhz wire causes random GND bounce somewhere which skews timings just enough to cause huge malfunctions. That's known as even a X10 probe is enough to cause chaos.

It's possible why changing the RAM amount is have such changes as it's changing the current though the grounds. A few pF in the wrong place and a change of just 1ns is enough to break timings. The STE seems to be gaining traction for being the worst ST these days !

Thank you @exxos . Please let me know if you do find them and I'll happily pay your costs for one. I'm using non-parity SIMMs.

It's annoying because I have 2 later model STEs that work perfectly (but they don't have the socketed CPUs).

@rubber_jonnie the purple tubing is the shroud. I don't want the wire to go any closer to the RAM as that causes more problems.

Ok, understand the shroud, but you're still running it by the power connector and the large cap etc and I'm wondering if the extra power load caused by additional RAM (Although I know it isn't a lot) is what the problem is and your 32Mhz cable is picking it up. The shroud will probably not help with EMI if it's just tubing.

I'll hopefully dig out mine today and add some pics, but both work just fine.

[exxos]

Thank you @exxos . Please let me know if you do find them and I'll happily pay your costs for one.

Can you PM me your address.

I've also documented similar problems here viewtopic.php?p=89049#p89049 though I don't know the cause of what happened then.

@rubber_jonnie it would be useful to know what motherboard and shifter number you have in your machine.

[rubber_jonnie]

@rubber_jonnie it would be useful to know what motherboard and shifter number you have in your machine.

Yep. I will check shortly.

[rubber_jonnie]

Ok, for reference here are my two 32Mhz STE boards:

First is the soldered in booster and it's the same CA4003290 board and Shifter as @DementED :

PXL_20231026_104537291.jpg (427.44 KiB) Viewed 2371 times

Second is the push fit booster, also a CA4003290 but a C300588-001 Shifter:

PXL_20231026_105307738.jpg (412.76 KiB) Viewed 2371 times

Both run fine at 32Mhz, and my 32Mhz pickup wire is taking a more direct route.

Now it's difficult to tell from your picture, but it looks like your feed from U405 is going directly to the TOS switcher. Can you confirm it's going to the booster board first please and then from IIRC the CE tap on the booster? If it goes direct to the TOS switcher, that's incorrect it should go to the booster, then that feeds the TOS switcher, but as I said, it's hard to tell from your image.

I did zoom all the way in but I couldn't tell properly, as it got a bit blurry.