• Please review our updated Terms and Rules here

Acer V10 486 Motherboard Quirks

J

jmetal88

Veteran Member
Joined
Jul 23, 2010
Messages
878
Location
Derby, KS
Since my 'Wanted' thread about a Socket 3 motherboard kind of turned into a support thread for a motherboard I already had that I thought was broken, I thought I'd finally start a new thread and continue posting about it here. There doesn't seem to be any full documentation for this board out on the Internet, although the most helpful resource so far has been this page:

http://museum.ttrk.ee/th99/m/A-B/32472.htm

First things first: Now that it's working and I've made some effort to hook things up to it, it looks like this thing is loaded with nonstandard pinouts!

First, here's what I discovered about the hard drive LED connector. The pinout ended up looking like this:

Acer_V10_HD-LED.png

Seems standard enough, to me, just double the pins it needs, is all. Hooking an LED up straight to the positive and negative pins makes it light up when there is hard drive activity. The positive lines are connected to +5V via 150 Ohms, and the negative lines are switched to ground during hard drive activity.

The real problem was the front panel LED connector. This is what I arrived at after a lot of measurement and experimentation:

Acer_V10_FP-LED.png

This connector required me to do a lot of wire splicing on the front panel connectors of my case. The +5V pin has no current limiting resistor on it. It's wired straight up to the power supply. You can use it to run the Power LED, but you need to put your own current-limiting resistor in place. I accidentally burned out the power LED in my case because I connected it without checking this first. Luckily, indicator LEDs are cheap, so it doesn't really matter. The TURBO pin switches between 5V and GND depending on whether the computer is running in high speed or low speed mode. In low speed mode, the pin presents a HIGH logic signal, and in high speed mode the pin presents a LOW logic signal. I hooked my turbo LED up between the +5V pin and this pin so it will light up when the computer is in high speed mode. The GND pin is exactly what it says it is, a connection to ground. The keylock pins do exactly what they say they do as well. If you short them together, the computer boots normally, but if you leave them disconnected the computer will refuse to boot. There is a two-pin DIS-LOCK jumper elsewhere on the board that is directly connected to these two pins. If it is shorted together, the computer will always be unlocked regardless of the state of the keylock pins on the 8-pin header. The RESET pin has approximately 3.3V on it and resets the computer when it is shorted to ground. There is no pin at position 6 (I assume this was the KEY pin for the original connector) and I have absolutely no idea what pin 5 does.

The IDE and FDD headers on this board are standard pinouts, and I can only assume the 50-pin SCSI header is a standard pinout as well (I don't have any working drives I can test it with). I haven't been able to test anything regarding the COM and LPT headers yet. I suspect the COM 1 header is a standard pinout. It's a 10-pin header with 9 pins available. If it's anything like other 9-pin serial headers I've seen, the 5-pin row would be connected pin-for-pin to the top 5 pins of the serial port and the 4-pin row would be connected pin-for-pin to the bottom 4 pins of the serial port. COM 2, I'm not so sure about. It's a 16-pin header with 15 pins available. I suspect it's meant for a 25-pin serial port, but am unsure how it would be wired up. I'm also not sure about the LPT header. Most computers use a 26-pin header with one pin missing, but this one uses a 20-pin header. If anyone has any ideas for relatively painless ways to get the proper pinouts for the COM 2 and LPT headers, I'm all ears! I'll post all info here once I've figured it out, as this board could really use some better documentation posted online.
 
Okay, I've also just discovered that this board only supports 5V CPUs. I must be pretty lucky with the AMD 486DX4-100 I have, because it's clearly labeled as a 3V chip but has no problems running on the approximately 4.85 volts that this board is supplying. I'd probably better put the original DX2-66 back in the board for the time being to be safe. According to the AMD chip's datasheet, it should have fried on this supply, as its maximum supply voltage rating is 4.6V, and its operating voltage is 3.3V +/- 0.3V. I guess I need to try and find a 486 voltage regulator board if I want to keep using the DX4. Someone in my other thread was under the impression that the AMD DX4 had a voltage regulator built in, but I can't find any evidence of that in the datasheet.

EDIT: Whoa, looks like those regulator boards are hard to come by. Somebody's got five of 'em listed on Amazon, but I'd have to pay nearly $45 for it. But then again, by any indication, it looks like a motherboard that would support the DX4 natively would cost me nearly as much anyway, if eBay prices are any indication. I wish I knew what happened to the original motherboard my parents had in this case. It's the one the DX4 came from, so it'd support it no problem. I would have to take up an extra slot with a multi I/O card on that one, though, since it didn't have anything integrated like this board does.
 
Last edited:
Looks like the SEEBITS program on this page would help me figure out the parallel port pinout:

http://www.terryking.us/parport/

I'll try it out later. Looks like the only thing I wouldn't be able to figure out from this would be which pin goes to Pin 1 on the 25-pin port, but, you know, process of elimination and if it's not one of the data pins and not one of the ground pins...
 
Chuck(G) said:
It's also pretty straightforward if you can simply determine (with a continuity tester), which pins on the header are grounded.
Click to expand...

Well, I'll be checking for grounds first, but since this is a 20-pin header with 19 pins accessible, I should only find two, and that won't help much with the positioning of the other pins. But I can use the program I linked to to set the outputs 'high' or 'low' and figure out which pins those are with my multi-meter, then I can jumper them back around to all the pins that *aren't* grounded and continue flipping bits until I've figured out what the other 8 pins are. Then I should just have the one left over that goes to pin 1 on the 25-pin connector.
 
Alright, this program helped me get almost all of the pins down. It didn't help me get Pin 1, which isn't supported by this program anyway, but it also didn't help me get Pin 14 or Pin 17.

EDIT: I might be alright to make an educated guess at this point. The pins are for the most part in order. Data bits 1-7 are consecutive on the header, with only bit 0 being off by itself for some reason. Two pins I can't determine for sure are pins 4 and 10, which are on either side of pins 6 and 8. Pin 6 should be connected to pin 15 on the D-SUB connector and pin 8 should be connected to pin 16 on the D-SUB connector, so I feel like it's probably safe to assume pin 4 should go to pin 14 and pin 10 should go to pin 17. That leaves the actual pin 1 of the header going to the actual pin 1 of the D-SUB connector.
 
Last edited:
Here's a simple text table pinout of the LPT header, from what my testing shows. Question marks by the pins I'm not sure about.

Code: 
|Header Pin	D-Sub Pin|
--------------------------
|        1	1?       |
|        2	2        |
|        3	3        |
|        4	14?      |
|        5	4        |
|        6	15       |
|        7	5        |
|        8	16       |
|        9	6        |
|        10	17?      |
|        11	7        |
|        12	18-21    |
|        13	8        |
|        14	22-25    |
|        15	9        |
|        16	10       |
|        17	11       |
|        18	12       |
|        19	13       |
|        20	NC       |
 
I haven't figured out a good way to test the serial port pinouts yet. I'm tempted (for the 9-pin header anyway) to just plug a port into it using the standard wiring scheme and try it on something. For the 15-pin header, I have no idea. I actually just shot an e-mail off to ALi to see if they knew where to get a copy of the datasheet for the M5105-A4E chip that's on the motherboard. If I knew the pinout for that, I bet I could figure out where the traces go using my multi-meter and figure out the serial port pinouts that way.

EDIT: Nope. Even the 9-pin is a nonstandard pinout. From where the key pin is, I'd expect the ground pin to be on pin 2, using the standard serial header pinout (would be pin 9 using standard pin numbering as well). Since I figure getting the chipset info from ALi is probably a long shot, I need another way to check which pins are which.
 
Last edited:
I *may* have figured out the 15-pin serial, although the 9-pin (which I thought would be the straightforward one) is still a mystery for now.

A post on the Vogons forum indicated this board was originally used in the Lucent Technologies Intuity MAP/5 server. I found some pictures of one on eBay and noticed that, indeed, only 15 pins out of the full 25 are present on the serial port on the rear of the computer. Now which 15 pins they were gave me a big clue as to what the motherboard pinout might be.

In fact, the top row had 8 pins present and the bottom row had 7 pins present (much like the odd row on my header having 8 pins present and the even row having 7 pins present). And in the 8 pin rows, the positions of the two ground pins line up perfectly (pin 1 and pin 7 on the D-SUB, pin 1 and pin 13 on the header, and pin 13 is the 7th pin down in the odd row of the header). So I'm thinking it might be pretty likely for this serial port to be wired up for the first 8 pins connected directly down the odd row, and pins 15, 17, 18, 20, 22, 23, and 24 to be connected in that order down the even row.
 
Tried to verify the pinouts as best I could with a multi-meter. Basically I just traced each pin to its respective buffer and noted whether it was a driver or receiver. The 15-pin header had the drivers and receivers EXACTLY where I expected, so I'm pretty confident in the pinout I mentioned above. The drivers and receivers on the 9-pin header seem to defy all kinds of number-order pin configurations, so I'm still really, really unsure what the pinout is there.
 
Haven't been able to do any serial port testing yet, but I did discover something else about this motherboard that it took me a while to notice because I hadn't done any real work in Windows 3.1 until yesterday. The computer was hanging when I tried to open text files in Notepad, specifically when I was browsing around the Open File dialog box. Since I was planning on replacing the hard drive with a smaller, more suitable one, I went ahead and did so and got to reinstalling everything. I noticed at that point that, with the generic VGA driver, I had no such issue with the Open File dialog box. When I reinstalled the video card driver, the problems started up again. So I started digging around in my graphics card driver documentation and pulled this out of the README:

Code: 
*  If you are encountering some lock-ups during heavy DMA transfers or
   your motherboard supports bus time-outs then you may try setting
   the following option in the [STLTH64.DRV] section of your SYSTEM.INI

     BUSYCHECK=ON  --  This is set to OFF by default.

I added that to my SYSTEM.INI and no more problem! So I guess this motherboard supports bus time-outs, or otherwise has some other quirk that requires this setting on my graphics card.
 
Okay, I dragged out my oscilloscope and my multimeter to finally try and figure out the serial port pinouts. I was able to 100% verify the pins of the header for the 9-pin port, and verify all but 5 of the pins on the header for the 25-pin port (but since I know what pins are present on the 25-pin port thanks to the picture I found, I'm pretty confident in the pins I wasn't able to detect).

Pinouts are as follows:

Code: 
     COM 1 (9-pin)
----------------------
HEADER PIN | D-SUB PIN
     1     |     NC
     2     |     9
     3     |     1
     4     |     4
     5     |     5
     6     |     6
     7     |     8
     8     |     7
     9     |     2
    10     |     3

Code: 
    COM 2 (25-pin)
-----------------------
HEADER PIN | D-SUB PIN
     1     |     1
     2     |     15?
     3     |     2
     4     |     17?
     5     |     3
     6     |     18?
     7     |     4
     8     |     20
     9     |     5
    10     |     22
    11     |     6
    12     |     23?
    13     |     7
    14     |     24?
    15     |     8
    16     |     NC

Question marks on the 25-pin table are by the pins I couldn't find a way to verify, but the pin numbers are the ones present on the picture of the rear panel I found for a machine that uses this board and if they are in any other order it would completely break the otherwise completely consistent pattern I observed in verifying the other pins on this header.
 
Alright, I finally got a serial/parallel port bracket and modified it for this motherboard. So far, I've only tested the parallel port, but using the pinout I provided earlier in this thread, it works great with my ZIP drive! I may never get around to fully testing the 25-pin serial port (I don't really have anything I can think of to use it for and I'd have to purchase additional connectors and ribbon cable to use it anyway), but I should be able to test the 9-pin serial port shortly.

EDIT: 9-pin serial is working pretty well, too, although I've only been able to test it with my old EPROM programmer (which communicates at 1200 baud).
 
Last edited:
You must log in or register to reply here.
Back
Top