• Please review our updated Terms and Rules here

Problem with IBM PS/2 model 57 486 SLC2 after Cmos battery change

carstenl

Member
Joined
Apr 4, 2021
Messages
23
Location
Denmark
I have got a strange problem, after cmos battery exhange on a model 57.

The machine could not keep the bios settings after the machine was turned off, so I changed the CR2032 battery on the bus adapter ( Riser card )
To get better access to the battery i removed the bus adapter, and changed the battery, with a new one, then I plugged in the bus adapter and put the cover on the machine.
After I have started the machine I keep getting the 00016100,000163000 and 00017100 error !!.

I used the Ref-disk several times, but each time the machine comes with the same errors.

I have also tested the bus adapter with another motherboard, and I get the same error, and just to be sure, I have also tested the battery in my model 77, and the battery works fine.

I have no idea what could be wrong with the bus adapter, but perhaps the connection from the battery to the edge connector is bad
Is there anyway to check the bus adapter, I have not been able to find any info about the pinout from the bus adapter.



Best Regards
Carsten
 

Attachments

  • photo65098.jpg
    photo65098.jpg
    143.7 KB · Views: 10
Ok , theres nothing wrong, you can beathe easy. Its saying that the cmos settings are going, because the battery was removed. Just press enter, then enter setup to change anything needed, and it will save.
 
Hi Twolazy,

I did run the automatic configuration, but every time the machine restarts, the errors codes comes agian :shock:
I have run the automatic configuration about 10 times now, but the problem is the same.
Also tried to press ESC and followed the instructions, but still no luck -

I get the same error, when i use the bus adapter with an other motherboard


Regards
Carsten
 
Hi Twolazy,

I did run the automatic configuration, but every time the machine restarts, the errors codes comes agian :shock:
I have run the automatic configuration about 10 times now, but the problem is the same.
Also tried to press ESC and followed the instructions, but still no luck -

I get the same error, when i use the bus adapter with an other motherboard


Regards
Carsten

I've heard of similar issues on "Bermuda" systems (both 76 and 77), and thought it might be another via issue like the "CuriousMarc" crew found. It sounds like you have isolated it to be a riser issue. Maybe I should try that on my 76 first.
 
I am not familiar with the IBM PS/2 Model 57 486SLC2, but I found a video of it online. The video shows the 'riser' but only one side of it. On that side, I can only see card connectors and the battery holder. Any electronics on the other side?

Does the reference disk behave exactly the same when no battery is fitted ?

I note that one of the errors is 00017100. Described at [here] as "00017100 (Fatal) - on a 90 or 95 means Rolling bit test failure on CMOS shutdown address byte". A "171" PS/2 error is shown elsewhere as the same failure reason. Perhaps: The shutdown address byte may be something stored with other 'CMOS SETUP' configuration information. If the 'riser' contains electronics besides the battery, then perhaps that electronics is where the 'CMOS SETUP' configuration information is stored, and that electronics is faulty.
 
I get the exactly same error, when the battery is not fitted on both SLC2 Motherboards
There are a few small components on the other side of the "riser", i have attached pictures that show the front and back of the " Riser"
Perhaps some of the components are bad, but still strange that it broke down, when i changed the battery.......
 

Attachments

  • IMG_0148.jpg
    IMG_0148.jpg
    210.7 KB · Views: 9
  • IMG_0149.jpg
    IMG_0149.jpg
    227.6 KB · Views: 9
I think i have found out, what is wrong with the risercard.
It seems to be a bad connection from one of the pins on the risercard ( pin 12 on the A side of the risercard )

The connection from the pin to the risercard is disconnected at the top of the pin. The printed circuit is very small there
and it broken, so there is no connection to it.

The connection from the Cmos battery runs through that connector, so that is why I can't measure 3 volts on Vbatt and GND
on the Dallas DS1285Q, when the machine is turned off.

Is there anyone, who can confirm if I´am right about it ? - maybe someone has got a risercard, and could check out if the Cmos battery is connected to pin 12 on the A side of the riser.

If I can´t fix the risercard, could it be an idea to connect the Dallas chip directly with at Cmos battery ?, of course without a battery in the battery holder on the risercard :)

Alternatively I could perhaps use a risercard to a model 56, just to get it to work ?!
I know it is not the right way, but perhaps it could work...


Best Regards
Carsten
 
I have attached pictures..the best I could take with my phone...

Pin 12 is connected to the point with the red circle, and the +side of the battery is also connected to that point.
 

Attachments

  • IMG_0390 (1).jpg
    IMG_0390 (1).jpg
    754.1 KB · Views: 17
  • Ikke-navngivet.jpg
    Ikke-navngivet.jpg
    995.4 KB · Views: 16
Hi, I have the same 00017100 problem with a Model 56 PS/2 and although the 16100 error clearly indicates that the Dallas chip isn't getting the +3V from the battery, the 00017100 as far as I can tell isn't related to it.

After I completely clear the CMOS, the 17100 does not come up even when there is no battery (or even the whole riser card) installed and the machine does boot to BIOS. I suspect that the NVRAM in the Dallas chip is faulty.
 
Last edited:
Hi,
It looks like the Dallas chip does not get the +3v from the battery, it is just like the error I have got.
The strange thing is, that I get the same error on two different model 57 mainboards

I suspect the the risercard is faulty, but i do not have a spare to test with - is it perhaps possible for you to see if
you have +3 volt on pin 12 on your risercard ?.

My risercard is for a model 57, but I think the risercard for a model 56 is the "same standart" it has just 3 sockets instaed of 5 for the model 57.
It is just an idea, how knows maybe it is so. ?

Perhaps it is possible to use external 3 volt battery on the Dallas chip ? ( like on the old 486 mainboards )
I have not tested it yet, because I do not know if it is safe to do that.


Regards
Carsten
 
Dallas is getting the +3V in my case, I have measured it. And in any case, the 17100 error does NOT come from the battery or the lack of it. Only error 16100 (battery missing) and 16300 (setup not configured) come if the Dallas chip is missing the +3V. The computer however will boot to BIOS settings even if the battery is missing.

17100 is an internal error for the CMOS RAM and I suspect the Dallas chip is broken (probably yours too), and this error is fatal in my case (Model 56) and will prevent booting even to the BIOS/reference disk. I can only get around it if I completely empty the CMOS, then the 17100 does not manifest itself. It comes back if I put a healthy battery and then do a cold boot, ie. power cycle the machine.
 
I think you are right, the Dallas chip is proberly broken
I can get it to boot on the refdisk, and do a automatic configuration, but after it restarts I still get the same error codes and the
warning, that tells you that there is an error in battery powered configuration data. Then it tells me to set time and date, and it runs automatic configuration again...and after reboot it is all over again.

Have you been able check if there is +3volt on pin 12 of the risercard ?.
I will try to use an external battry for the Dallas chip, and see what happens


Regards
Carsten
 
Yes, pin 12 on my riser's backside (the side pointing to the PSU's direction) has +3V. You can try to fix the connection on your riser by soldering (should be rather easy) or like you said, just hardwire a battery to the pin on Dallas.
 
Thanks (y), it is the same pin where the connection is broken,
I will get some help from a friend, the has got better soldering equipment then I have got - it looks like an easy fix
If not, we will hardwire the Dallas chip.

I will get back to you, when we have done some testing - it will properly be in the weekend.

I found this post on Vogons .:

He also had problems with the Dallas chip, but is is a DS1585S chip


Regards
Carsten
 
The problem with the DS1585S is a very, very common one (I've fixed/replaced a few) but unfortunately nothing to do with this problem. There's a battery inside the DS1585S and it just goes bad and you have to replace the whole chip or do some dremeling and carve out the contact points for a new coin battery.
 
We have fixed the risercard, and the machine works just fine now :),
And thanks for the Help(y)

I hope, that you will be able to fix your machine - I will send you a PM.


Regards
Carsten
 
Finally managed to solve my problem after I got the new Dallas 1285Q chip and a friend of mine changed it. The dreaded 17100 did still come even with the new chip but after taking it off and soldering back in a few times the PC started to work perfectly. So in the end I'm not sure if the culprit was a bad solder joint or the chip itself.
 
Back
Top