VOGONS


First post, by VivienM

User metadata
Rank Oldbie
Rank
Oldbie

After my previous attempts at a DOS machine ended with somewhat of a conclusion that there was some subtle fault in the motherboard, I picked up this Gateway E3200 system.

When I got it, I noticed that it would power on as soon as I plugged the power cable in, the power button appeared not to work, and it complained about the CMOS battery. Oh, and there were reasons to suspect the primary IDE channel didn't work, but I didn't get to any kind of conclusion on that.

I naively figured, well, the CMOS battery is a CR2032, that's easy to replace. Except that, well, to reach the CR2032 on this more or less requires taking the motherboard out (actually, I now realize there was another way to reach it, but it's a month too late for that). It's a sliding mechanism that goes into the riser board that has the power and lots of other things. Okay, take motherboard out, replace CR2032, put it back in... and now it doesn't want to turn on. Okay, maybe now the setting for what to do after power loss has changed. And... power switch still seemed dead. It's also possible I haven't slid the motherboard fully in, it just feels like it isn't despite my many attempts.

Fast forward a few weeks, buy a replacement power switch/front LED/front fan assembly. Swap it in. And... well... after much fiddling with it, here is where things are at - if I plug in the power, the fan starts running, the power LED starts blinking, the power switch does nothing, no beeps, nothing.

These NLX systems are weird because all the power goes through the riser board, as do the front panel switches (and the IDE cables for that matter) and I can't find any good way to tell if the motherboard and the riser board have a good connection.

Thoughts? Am I just cursed when it comes to finding a late DOS system?

Reply 1 of 3, by DaveDDS

User metadata
Rank Oldbie
Rank
Oldbie

I've got a couple of gateways which both to the same at "plug in".

They start running for about a second (lights fan etc.) and then power-off.

I've never had to dig deeper, but I suspect the ATX power-on line defaults to ON and then gets turned
off by the BIOS.

Perhaps yours isn't running enough for BIOS to run at all -- I'd check the ATX power-ON signal and see
if it comes uo ON ... also try switching it OFF and confirm that power control actually works.

Dave ::: https://dunfield.themindfactory.com ::: "Daves Old Computers"->Personal

Reply 2 of 3, by VivienM

User metadata
Rank Oldbie
Rank
Oldbie
DaveDDS wrote on 2025-07-03, 03:02:

Perhaps yours isn't running enough for BIOS to run at all -- I'd check the ATX power-ON signal and see
if it comes uo ON ... also try switching it OFF and confirm that power control actually works.

I don't think switching it off works, at least not using the power button (which I just replaced)...

Honestly, I suspect the motherboard isn't properly connected, but I don't know how to fix (or even confirm) that at this point. It's plugged into the riser card as far as I can shove it...

Reply 3 of 3, by DaveDDS

User metadata
Rank Oldbie
Rank
Oldbie
VivienM wrote on 2025-07-03, 03:22:

I don't think switching it off works, at least not using the power button (which I just replaced)...

Honestly, I suspect the motherboard isn't properly connected, but I don't know how to fix (or even confirm) that at this point. It's plugged into the riser card as far as I can shove it...

I'm not families with this exact machine ... I'm assuming it has an ATX (or equivalent) supply where the power is controlled by a digital input into the
supply instead of the older design where power was physically switched between mains<>PSU.

These designs typically have a low-power "always on" rail which feeds the mainboard power control circuitry.

Assuming you can figure out "whats what", I'd check the control signal and see is the PSU is actually getting told
to turn on/off, and guessing NOT - check the low-power "always on" rail.

Dave ::: https://dunfield.themindfactory.com ::: "Daves Old Computers"->Personal