VOGONS


First post, by abasak

User metadata
Rank Newbie
Rank
Newbie

I just bought a funny little Point of sale PC, a Wincor Nixdorf Beetle/M. It's configured with a PIII 1Ghz CPU, 512Mb RAM and both an ISA slot and a PCI slot. Right now it is working "fine" with a Voodoo3 PCI and a SB16 CT2940 (OPL3 version).

That is all well and good, however when I replaced the CR1/3N lithium battery, I noticed that the system still couldn't keep the BIOS setting when I pulled the power plug/or turned of the PSU. After hours of troubleshooting, BIOS updating and jumper hunting, I took out my voltmeter and noticed that the brand new battery only held a charge of 2.4v. I thought I might've bought a dead or old VARTA battery, so I had it exchanged today at around 4:30PM, now at 10:00PM the new battery is down from 3.1v to 0.6v in a matter of hours. So something is completely wrong.

There is nowhere in the BIOS to change anything related to this, and there are NO jumpers on the boards related to CMOS reset. This is after all a quirky POS (Point Of Sale, not Piece Of S**t) computer.

This is really testing me, and I'm at my wits end. Documentation is sparse, there is a 32k NVRAM chip on the board, however it seems like this is not for storing BIOS info but "This memory chip can be used to store important data - such as sales totals or diagnostic entries - by means of the appropriate software and independently of the power supply." I don't know if it is used for the BIOS as well. I have ordered a new one just to be sure. But is there a way to identify the CMOS memory on the board?

Anyway, do anyone of you guys have any suggestions? How do I troubleshoot this? Should I build some sort of AA battery pack and hook it up to test drain (how would I btw. do that?). How do I measure drain?

Reply 1 of 9, by darry

User metadata
Rank l33t++
Rank
l33t++

Does your meter have an ampere-meter mode ? You could connect any 3.6v-ish 3.0V DC power source EDIT: better check with a low current capable battery or current-limited power source to avoid worsening things and check the current draw with the meter in series with the power source.

Apologies if this was already what you had in mind.

Last edited by darry on 2022-12-28, 23:47. Edited 1 time in total.

Reply 2 of 9, by pentiumspeed

User metadata
Rank l33t
Rank
l33t

One time Asus had a firmware bug that drained too quickly and a newer firmware fixed this on ASUS SP97 series boards.

Cheers,

Last edited by pentiumspeed on 2022-12-28, 22:29. Edited 1 time in total.

Great Northern aka Canada.

Reply 3 of 9, by abasak

User metadata
Rank Newbie
Rank
Newbie

That's a great suggestion. I am not that knowledgeable with voltmeters, so two things: 1. does this require a new battery? 2. where do I measure/check the draw?

Re: BIOS update, it is updated to the newest BIOS, which didn't fix it.

Reply 4 of 9, by abasak

User metadata
Rank Newbie
Rank
Newbie

Update it uses a W83195BR-25 which is a "Clock Synthesizer for Intel 815 Solano chipset", is that where I should probe with my multimeter? And what ampere setting should I put it on?

https://datasheetspdf.com/mobile/370422/Winbo … d/W83195BR-25/1

Reply 5 of 9, by darry

User metadata
Rank l33t++
Rank
l33t++
abasak wrote on 2022-12-28, 22:29:

That's a great suggestion. I am not that knowledgeable with voltmeters, so two things: 1. does this require a new battery? 2. where do I measure/check the draw?

For starters, please share the brand and model of your voltmeter and, if possible, a photo of it .

1) Just to be safe, are you sure you are using the right type of battery ? I presume you identified it as needing a CR1/3N based on the originally installed battery type . CR1/3N are actually 3V nominally (not 3.6V as I initially wrote) and are not rechargeable, so rigging any 3V battery source (i.e. 2 AA cell in series) would work, BUT, such a setup could provide a lot more current than a CR1/3N and if something is wrong on the motherboard that causes it to draw too much current, this might risk causing further damage. Consequently, it would probably be safer to try with another CR1/3N for testing first.

2) you would need to set your multimeter to ampere-meter mode and connect one lead to one of the battery's poles and the other lead to the terminal on the motherboard where that pole is meant to normally connect. For the duration of the test that pole and the battery and the terminal would need to NOT be in contact directly, that way the current will flow through the multimeter . See this for a diagram https://www.allaboutcircuits.com/textbook/exp … /ammeter-usage/

EDIT : Also, did the new batteries have an expiry date ? I am not familiar with how unused primary lithium cells typically age when past they the "best before" date, but I would hasard to guess that their nominal voltage falls below 3V, which would mean 3.1V out of the box would mean the battery was OK to begin with. On the chance I am wrong (and that past due or bad primary lithium cells can actually have a normal nominal voltage out-of-the-box and drop abnormally quickly once in use), it may be better to try sourcing any further test batteries from another vendor (or another brand) and checking expiration date diligently.

Reply 6 of 9, by abasak

User metadata
Rank Newbie
Rank
Newbie

Thank you for your reply.

A: It is a no brand cheap multimeter. Here is a link in Danish to it https://www.xl-byg.dk/shop/millarco-0-visning … timeter-1825564

1) Yes, I am sure it is a CR1/3N, I will buy a bunch more tomorrow.

2) Thanks for that link.

3) No there was no date on the batteries, the clerk at the store wondered the same, and we both looked. I'm going for some at another place tomorrow.

The KM62256CLP-7L NVRAM chip is listed some places as being a BIOS memory chip, so it might be a good idea, that I've ordered a new one, even if the manual says it isn't used for storing BIOS data.

Reply 7 of 9, by dionb

User metadata
Rank l33t++
Rank
l33t++
pentiumspeed wrote on 2022-12-28, 22:28:

One time Asus had a firmware bug that drained too quickly and a newer firmware fixed this on ASUS SP97 series boards.

Cheers,

Second this, it was also an issue on the GVC FR-520. Something related to USB standby power if I recall correctly.

Check for BIOS updates for the board, see if release notes mention battery.

Reply 9 of 9, by abasak

User metadata
Rank Newbie
Rank
Newbie

Okay, I might’ve solved it. I am tentatively optimistic. After having removed the dead/Old NVRAM-chip and having put in yet another new battery it seems like it keeps the settings.

Before the new battery was already depleted, so when I removed the chip it couldn’t keep the settings anyway.