exxos's Atari, Amiga & retro forum

Thanks for the update. Seems like it has a few problems going on causing various problems.

So I decided to post a few pictures... I was debugging a lot to get to this point...

My modified NetUSBee at this point, plus "proof" that it works now. My previous "manual reset' hack is removed already. Is there a NetUSBee problems topic? Maybe I should have post there instead, since my STE's issue is basically solved by now.

Chain-Q wrote: 14 Jun 2022 17:47 Is there a NetUSBee problems topic? Maybe I should have post there instead, since my STE's issue is basically solved by now.

Nothing in particular no. But maybe you could do a dedicated guide page for this as it could be included in our future wiki.

I guess we need more people to try your fixes next.

I'd be interested in doing your 'proper' fix once you've documented it. I wonder if Lotharek would be interested in your findings, I mean, if anything it will reduce the amount of warranty exchanges he probably has to do (even I warranty changed one of my units with him once when it just randomly stopped working)

The changes make no difference on my TT.

Then again none of the "write using read" devices work on my TT. I think it's because the address lines ring like bells on the cartridge port. I think it needs small capacitors adding to them to dampen the wild swings. (The transition of 0V to 5V bounces up to ~3V, down to 1V, up to 4V, down to 2V etc. for 10s of nanoseconds before stabilising, just in time for it to go low, usually.)

stephen_usher wrote: 14 Jun 2022 19:50 The changes make no difference on my TT.

But according to what you write, the USB part of your cart also doesn't work, right? I should make it explicit, these changes most likely only cause a change, if you have non-working networking, but otherwise the USB part of the card works. At least that's a sure sign that this reset problem my changes aim to fix exist on your hardware.

If the USB also doesn't work, then most likely these changes won't fix things, and you shouldn't do them, because the main problem lies elsewhere.

The USB part isn't fitted. It's the "lite" version.

Steve wrote: 14 Jun 2022 19:04 I'd be interested in doing your 'proper' fix once you've documented it. I wonder if Lotharek would be interested in your findings, I mean, if anything it will reduce the amount of warranty exchanges he probably has to do (even I warranty changed one of my units with him once when it just randomly stopped working)

Just to wrap this up here too, I posted my "proper fix" based on a 3 pin MCP-101 power monitoring chip over at atari-forums.com, with links to pictures documenting the thing on Twitter. And it indeed works, even with the previous config pin bodges removed. In my opinion, this confirms the previous theory of the power-on reset hold time was insufficient.

I also finally sent an e-mail to Lotharek about my findings. Lets see if he answers.

From the look of it the surface mount version would be ideal. Glue it upside-down on top of the RTL chip and then take bodge wires to the correct locations. That would be neat and low profile and fit in the normal metal shield.

Yes the surface mount version is an option. You maybe also be able to fit it somehow in place of the original reset cap/resistor, and of course that's also the best candidate for the redesign. Just to note, the through-hole version also fits in the metal case if laid down next to the RTL chip, there's plenty of space on that side between the metal case and the PCB.