VOGONS


First post, by cuco81

User metadata
Rank Newbie
Rank
Newbie

Hello, I just got my self a PC Partner MVP3bs7-954 socket 7 motherboard for a retro gaming build. Installation has gone smooth, windows98 is set with all drivers and my 233 mmx is running at 290mhz @ 3.3v stable.

Pretty happy so far but I'm hitting a wall with the USB header. I have a standard 2 usb connector with the 10 pin plug connected to the motherboard but windows is not detecting nothing at all after installing the NUSB drivers. The BIOS shows USB onboard chip enabled, the plug is connected to the right pins according to the wire diagram , and the voltmeter measures ~5v on pin #1 when mobo is powered on.

Anybody else experience this issue?

Reply 1 of 12, by kingcake

User metadata
Rank Oldbie
Rank
Oldbie

According to the manual for that board it's a 8 pin usb header with non standard (by today's use) pinout.

Screenshot 2024-03-10 203649.png
Filename
Screenshot 2024-03-10 203649.png
File size
46.66 KiB
Views
442 views
File license
Public domain

Reply 2 of 12, by Repo Man11

User metadata
Rank Oldbie
Rank
Oldbie

OP, consider yourself lucky. I too was once saved by luck (23 years ago) when connecting the USB header to an AT motherboard. I found out my P55T2P4 had a USB header, and I bought a cable from an Ebay vendor. It had two five pin connectors, and no instructions, so I plugged it in and powered it on. Ahead of their time, Asus used the nine pin arrangement, so if I'd received a nine pin USB cable I would have been home free, but as it was I connected one side correctly, but the other had a ground wire connecting to the +5 volt pin! What saved me was that the AT power supply had a ground fault detection, and it wouldn't power on. I then did what I should have done initially, and researched it and connected it properly.

The good news is that you've been provided with the pin out, and you can easily rearrange the pins to match.

Attachments

"I'd rather be rich than stupid" - Jack Handey

Reply 3 of 12, by cuco81

User metadata
Rank Newbie
Rank
Newbie

Thanks guys, I did realize the manual diagram showed a weird layout. Seems each usb connector is opposite to the other. The 10 pin connector on my USB panel is actually split in two plugs so I flipped them in the right order before connecting. I do wonder if the previous owner did something wrong and shorted something. Apart from the usb header, everything else works fine though.

I might have to install a usb expansion card if I can't get that sorted, sigh.

Reply 4 of 12, by Repo Man11

User metadata
Rank Oldbie
Rank
Oldbie

If I were you I would check the +5 pins with a meter. If they don't have the proper voltage, there may be a blown fuse that can be repaired. It could also be a bad solder connection, or a cut trace.

"I'd rather be rich than stupid" - Jack Handey

Reply 5 of 12, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Advise against running mmx at such a high voltage. That’s in reality 1 whole volt over what that processor will probably be stable at 233mhz at. I know they are speced at 2.8v

Anyway,

Try to avoid an add in card. They will slow this system down by as much as a 1/3 in benchmarks. Especially the usb 2.0 ones.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 6 of 12, by kingcake

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2024-03-11, 03:41:

Try to avoid an add in card. They will slow this system down by as much as a 1/3 in benchmarks. Especially the usb 2.0 ones.

The USB controller in the chipset will be internally connected to PCI just like an addon card, won't it?

Reply 7 of 12, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
kingcake wrote on 2024-03-11, 03:55:
Sphere478 wrote on 2024-03-11, 03:41:

Try to avoid an add in card. They will slow this system down by as much as a 1/3 in benchmarks. Especially the usb 2.0 ones.

The USB controller in the chipset will be internally connected to PCI just like an addon card, won't it?

Idk, but the effect of add on cards on socket 7 systems is well documented. Especially usb2.0 cards.

I think the 1.1 cards may not be as bad? Unsure. But it’s a big enough problem that all my systems use internal hubs off of the chipset usb.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 8 of 12, by cuco81

User metadata
Rank Newbie
Rank
Newbie

OK, I think I'm getting closer to the culprit. Measured the pins on the usb header, both vcc pins measure ~4.5-4.8v but they fluctuate. I then ensured I plugged both usb connectors the right way according to the manual diagram and got the following:

1)My 128gb usb stick remains undetected.
2)My 1gb usb stick is detected but unusable, shows as unknown under usb root hub.
3)My Creative pebble speakers get power from the usb port. They work when connected but as soon as I connect one of my usb sticks on the other port, the speaker sounds starts cutting off intermittently, as if there was a voltage drop.

I removed and re-installed the usb controller drivers but saw no change. Next thing to check is the controller itself. There might be a hardware issue. Will have to check the controller chip and connections.

Reply 9 of 12, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

I think some boards turn off power to USB ports when in suspend mode, so they have a FET between the PSU 5V and the USB 5V. Could be something like that causing a bit of resistance on that 5V line, so when anything tries to draw an current the voltage drops. If there is a FET that switches the power then it'll be near the header.

Reply 10 of 12, by the3dfxdude

User metadata
Rank Oldbie
Rank
Oldbie

What is the voltage from the vcc pin after you plug something in? Probably not enough.

I think it's odd so many people are coming in wanting working USB on pre-2000 hardware. I never really used any USB before '99 at all on Win98 other than the odd mouse or scanner or peripheral that just came out by about then. Certainly didn't really care for it in the socket 7 era. I remember USB really only started to get used widely about when WinME came out. Yes I remember these headers on the boards, thinking, it'd be great when stuff finally arrives that I'd want to use it...

Reply 11 of 12, by cuco81

User metadata
Rank Newbie
Rank
Newbie

I never used USB either in those days but at this moment it's extremely important that I have a working USB as it allows me to do what I couldn't do back in the day. It allows me to use my CH flightstick and throttle on all my simulation retro games (ef2000, mechwarriror2, xwing alliance, wing commander, etc.).

I haven't measured voltage after plugging the devices, not sure how to honestly. Will have to get a bit educated on where to start looking. Probably most capacitors need changing and some other things might need some looking into.

Reply 12 of 12, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

The two ports share the voltage. Read voltage on one set of pins, while the other set is plugged in.

Honestly, I don’t think this is your issue though.

Find vcc, and gnd of both.

Put red on vcc, black on vss(gnd) and the two in the middle are data. They will be in the order on the header. Plug it in, enable it in bios, install drivers, and it should work.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)