• Please review our updated Terms and Rules here

ASUS m/b destroying serial mice with -12V!

Chr$

Chr$

Experienced Member
Joined
Aug 13, 2021
Messages
155
Location
Saxony, Germany
I wonder if anyone has experienced this before. I have an ASUS P5A-B Super Socket 7 Motherboard. It was happily running Win 98 the other day when all of a sudden the serial mouse stopped working. I restarted and got the 'no mouse detected message'. Tried another mouse, same thing. Tried removing com ports, reinstalling etc and nothing worked (although USB mice do work fine). Since then I've also tried a different PSU.

For sanity I checked both the mice on my IBM XT and neither of them work now, and they were absolutely fine before!

So I got out the multimeter and checked the serial port - with one lead on a good ground and with the PC on I measure -12v on pins 2, 4 and 7. In continuity mode those 3 pins also show 0 ohm shorts whereas pin 3 (which should be gnd) shows about 10 ohms to ground. With the computer off, pin 3 is 0 ohm to gnd, as it should be. I suppose -12v where it shouldn't go to has killed the 2 mice. Serial port 2 is the same, with -12v on pins 2, 4 and 7 when the PC is on.

What on earth is going on? The serial port chips on the m/b are a pair of Holtek HT6751.

I've disassembled one of the mice and it's mostly passives with a Motorola LXC437159P chip. To get that mouse working again is it worth obtaining and fitting a new LXC437159P?
 
Update. No idea what's going on but one of the mice does now seem to work on the XT. The other doesn't. I turned off the m/b serial ports in the BIOS and with an ISA serial card the same mouse also works.
 
RS232 is defined as (up to) ±15V for its signal levels. PC power supplies only provide ±12V, so that is what you get. However, some devices (such as USB serial converters) may cheat; I've seen regular CMOS voltage levels (0V..5V) in production. Serial mice (and some other input devices) get powered from the data lines, which is bending the rules slightly.

As a kid, I've apparently fried both a serial port and the mouse connected to it with software. Was a 386-class machine, and I still don't know how I managed that, but they never let me anywhere close to that machine after that. (Lots of curses, as they had to operate Excel for the rest of the event without a mouse.)
 
The only thing that was a little unusual at the time the mouse died was that I had a molex to usb adapter attached to one of the molex connectors, just to provide 5v to charge my phone. All PSU voltages are still good and win98 wasn't doing anything much at the time, so it's a mystery but certainly I can no longer get a mouse to work on the m/b integrated serial ports.
 
Chr$ said:
The only thing that was a little unusual at the time the mouse died was that I had a molex to usb adapter attached to one of the molex connectors, just to provide 5v to charge my phone. All PSU voltages are still good and win98 wasn't doing anything much at the time, so it's a mystery but certainly I can no longer get a mouse to work on the m/b integrated serial ports.
Click to expand...

I look at those adapters and am depressed by the lack of diodes to prevent backfeed.
 
hornbetw said:
You may have blown one or more of the RS232 driver/receiver chips on the m/b.
Click to expand...

So, one or both of the Holtek HT6751 chips. Is it worth removing one to see if one of the ports works? Then swap them etc and test again. If one is ok then that should do the job. they are right next to the plastic floppy port, so bit of a pain the de solder. I've already checked the pins and reflowed them all.
krebizfan said:
I look at those adapters and am depressed by the lack of diodes to prevent backfeed.
Click to expand...
Just in case it had anything to do with the failure, I certainly won't be using that again.
 
According to the datasheet, Holtek HT6751 chips are motor driver chips intended for camera zoom/focus, etc. functions. They can't be kludged to function as RS232 driver/receivers.
 
Looks like a drop in replacement for the TI SN75185. Their are other manufacturers of compatible chips. It looks like the Holtek chip only comes in SOP-20 package. Digikey sells the TI SN75185DW for $1.44/ea. If you can solder SMT ICs, it should be relatively easy to replace.
 
Svenska said:
RS232 is defined as (up to) ±15V for its signal levels. PC power supplies only provide ±12V, so that is what you get. However, some devices (such as USB serial converters) may cheat; I've seen regular CMOS voltage levels (0V..5V) in production. Serial mice (and some other input devices) get powered from the data lines, which is bending the rules slightly.
Click to expand...
That's fine, the Amiga does that as well on it's serial port. RS232 is very generous about what voltage you actually use.
 
hornbetw said:
Looks like a drop in replacement for the TI SN75185. Their are other manufacturers of compatible chips. It looks like the Holtek chip only comes in SOP-20 package. Digikey sells the TI SN75185DW for $1.44/ea. If you can solder SMT ICs, it should be relatively easy to replace.
Click to expand...
Thanks. I wonder if I have something compatible on a scrap board. A GD75232 also seems to be the same thing. It would be a quick and easy job if it weren't for the plastic floppy port right next to the chips. I'll use an ISA card for now and it can be a project for another day.
 
I found a donor board with a pair of GD75232's, so swapped them over. Here are the empty pads on the ASUS P5A-B. I thought I took a finished pic, but that doesn't seem to have worked! There is very, very minimal floppy port melting (I can live with that).

Anyway, I still have no mouse! Win98 reports COM 2 and COM 3 but no sign of COM 1. Also, booting to DOS and using Checkit, it also claims there is no COM 1. No matter what I seem to play around with in the BIOS. And with the ISA card installed the same serial mouse works fine. It's not the end of the world, but I hate it when things suddenly stop working as they should.
 

Attachments

  • 20231002_142524.jpg
    20231002_142524.jpg
    2 MB · Views: 6
Timo W. said:
That's fine, the Amiga does that as well on it's serial port. RS232 is very generous about what voltage you actually use.
Click to expand...
Works fine for data transfer, may not work fine if the device is powered through the data lines. I've thrown away a decent serial digitizer tablet because of that - it would't do a single straight line unless connected to a "proper" UART. I wouldn't be surprised if serial mice generally fail at TTL/CMOS voltages, either (obviously not an issue on Amigas).

Chr$ said:
Anyway, I still have no mouse! Win98 reports COM 2 and COM 3 but no sign of COM 1. Also, booting to DOS and using Checkit, it also claims there is no COM 1. No matter what I seem to play around with in the BIOS. And with the ISA card installed the same serial mouse works fine. It's not the end of the world, but I hate it when things suddenly stop working as they should.
Click to expand...
Sounds like you have a proper misconfiguration between the ISA card jumpers, the system BIOS settings, and the Windows drivers. You need to sort that out first, otherwise you'll never get anywhere. Keep in mind that you will have interrupt conflicts (COM1 and COM3 share IRQ4; COM2 and COM4 share IRQ3). Mouse drivers require the interrupt, so you must not use the same interrupt while it is loaded.

Anyway. You should start by checking whether your serial ports actually work. Connect a wire between Rx and Tx, then use a terminal program to check for echo. Check the voltage with a multimeter; it should be LOW (negative 12V) when idle, somewhat close to 0V when sending lots of 'U' characters, and quite positive when sending lots of SPACE characters. This should also be conclusive in figuring out which port is where. :)

You can also connect Tx on one port with Rx on another port, and see if characters arrive in the expected direction. Since your ISA card ports work, that should tell you which mainboard line drivers are actually faulty. Or if they all work correctly and you have a broken mouse or a software issue.

If you verified that all ports work correctly, proceed to DOS mode (no Windows!) with a known-good mouse driver and verify the mouse on all ports.

Please report back.
 
Svenska said:
Please report back.
Click to expand...

Ok, I removed the ISA serial card and enabled just the serial port 1 in the BIOS with its IRQ as 4. There is a PCI video card, no other cards are installed. A breakout 9 pin port is connected to the serial port 1 on the m/b (with the stripe on the pin 1 side). I made a loopback cable that connects pin 3 to 4 (and 7 to 8 and 1 to 4 to 6 to 9). I booted from an MS DOS 5 disk I had laying around and for now have just been using Checkit to test.

I answered Y to 'is a loopback cable connected'. It recognised there was a COM 1 but it fails like the image shows. If I answer no to the loopback question then it passes all tests.

If I load ctmouse before checkit, ctmouse does report that it has loaded a mouse to COM 1. And Checkit thinks a mouse is there on COM 1 but it simply doesn't respond at all to the mouse test (no movement, no button presses).

And then when I pop the ISA serial card back in and disable both m/b COM ports via the BIOS, everything works fine with the mouse connected to the ISA card (also IRQ4) and those Checkit tests pass with my loopback contraption connected.

So it must be something hardware related on the m/b, right? It did just stop working one day, so I suppose something randomly failed and as I've changed the GD75232's they can probably be ruled out. I've also de-oxit'd the headers on the m/b and also reflowed the solder on them. Traces are tiny and go under stuff so it's hard to inspect them properly.
 

Attachments

  • 20231003_190425.jpg
    20231003_190425.jpg
    1.1 MB · Views: 9
The GD75232 chips are just line drivers. You need to check the UARTs upstream of them, and any passive components in circuit with them.

On more modern boards, the UARTs are generally in the Super I/O Chip. On older boards, they are sometimes integrated into the South Bridge chip. Even older than that and they're usually discrete chips or TTL logic.

Looking at a picture of your board, it unfortunately looks like the line drivers connect to the South Bridge, so if there are no bad passive components and the traces check out, I would say that the SB chip has dead serial UARTs
 
Oh well, I tried. I'll have to stick with USB mouse or mouse over ISA serial card then and will mark the board as having faulty serial ports.

I wonder how it happened though. Is it possible that my phone pulled too much current from the 5v rail and could that have killed off the UART part, do you think? And whatever happened also seemed to kill one mouse. Inside the non working mouse there are mainly passives and a controller chip... it has an HM8375CP chip. I can't find any data about it, apart from other people that also found that chip in similar mice. If I ever come across a mouse with the same chip, I'll swap them and test it, but it's probably not worth actively seeking such a chip.
 
You must log in or register to reply here.
Back
Top