MB-1433UIV won't hold CMOS settings \ VOGONS

MB-1433UIV won't hold CMOS settings

Topic actions

Post a reply

First post, by dionb

Posted on 2019-09-29, 21:51

dionb Offline

Rank l33t++

Rank: l33t++
Posts: 5761
Joined: 2017-12-23, 15:35
Location: Amsterdam

Recently got myself a little unexpected gem as part of the - a Biostar (?) MB-1433UIV. That's common enough (sold one myself a few weeks back) but this is revision 7, which means 3.3V and 4V supported, as is the Cx5x86. That makes it pretty much the ultimate VLB motherboard, which was exactly what I wanted. Better yet, apart from awful HDD support (chips for XTIDE on NIC already ordered) it works perfectly with my 5x86 and other stuff.

Just one issue: CMOS settings.

The old barrel battery has been removed without visible damage. The board has a 4-pin header (JP8), which needs a jumper over 2-3 if 'internal' battery is used, otherwise an external battery can be connected to pins 1 and 4, with 1 being +. So that's what I'm trying to do:

3x AAA rechargeable batteries, fully charged (measured 3.78V on the connector pins), red positive connected to JP8 pin 1, negative connected to white (black was on wrong pin...) JP8 pin 4.

Should work? Well, it doesn't. After a few minutes without power, all settings gone. Tried inverting polarity. Same again.

Any suggestions how to get this board to remember what I tell it?

Reply 1 of 7, by Baoran

Posted on 2019-09-29, 22:31

Baoran Offline

Rank l33t

Rank: l33t
Posts: 2200
Joined: 2017-04-01, 08:39
Location: Finland

I thought there should be more than 3.78V in those batteries (1.5V each). I have a motherboard that has similar connector and there is 5V in the battery connector so I always assumed I would need at least 3x1.5V which would mean 4.5V there. With 1.26V each it seems like the batteries are empty or close to it.

Reply 2 of 7, by dionb

Posted on 2019-09-29, 23:06

dionb Offline

Rank l33t++

Rank: l33t++
Posts: 5761
Joined: 2017-12-23, 15:35
Location: Amsterdam

Baoran wrote:
I thought there should be more than 3.78V in those batteries (1.5V each). I have a motherboard that has similar connector and there is 5V in the battery connector so I always assumed I would need at least 3x1.5V which would mean 4.5V there. With 1.26V each it seems like the batteries are empty or close to it.

NiMH rechargeable AAA batteries have a nominal voltage of 1.2V, not 1.5V (no-rechargeable alkalines) so actually at 1.26V they're over spec.

I assumed the original barrel was a 3.6V barrel, but if the CMOS needs 5V, that might explain something. Tomorrow I'll try with some alkalines for 4.5V - if they do seem to hold settings, I'll add a diode and stick with that.

Reply 3 of 7, by Baoran

Posted on 2019-09-29, 23:18

Baoran Offline

Rank l33t

Rank: l33t
Posts: 2200
Joined: 2017-04-01, 08:39
Location: Finland

dionb wrote:
Baoran wrote:
I thought there should be more than 3.78V in those batteries (1.5V each). I have a motherboard that has similar connector and there is 5V in the battery connector so I always assumed I would need at least 3x1.5V which would mean 4.5V there. With 1.26V each it seems like the batteries are empty or close to it.

NiMH rechargeable AAA batteries have a nominal voltage of 1.2V, not 1.5V (no-rechargeable alkalines) so actually at 1.26V they're over spec.

I assumed the original barrel was a 3.6V barrel, but if the CMOS needs 5V, that might explain something. Tomorrow I'll try with some alkalines for 4.5V - if they do seem to hold settings, I'll add a diode and stick with that.

At least my motherboard that had similar connector had 5V between pins 1 and 4 when checked using multimeter even if originally it had lower voltage barrel battery. Normal AAA batteries might work as long as it doesn't try to recharge them like my motherboard does.

Reply 4 of 7, by TheMobRules

Posted on 2019-09-30, 00:54

TheMobRules Offline

Rank Oldbie

Rank: Oldbie
Posts: 1441
Joined: 2016-01-19, 23:29

I have that motherboard, I first tried a 2xAA battery holder (3V) but CMOS would not keep time and settings when turning off. Using 3xAA alkaline batteries (4.5V) did the trick.

I think there is a bigger voltage drop when using the external battery connector vs. the barrel battery terminals due to some additional diodes, so the CMOS/clock chip may not be getting enough voltage with the NiMH...

Reply 5 of 7, by dionb

Posted on 2019-09-30, 06:23

dionb Offline

Rank l33t++

Rank: l33t++
Posts: 5761
Joined: 2017-12-23, 15:35
Location: Amsterdam

Alternative might be to solder in a few pins where the original battery was connected...

You say the bigger voltage drop is due to additional diodes. That suggests it should be possible to use alkalines (or other non-rechargeable batteries) without additional diodes. I'll check that out this evening.

Reply 6 of 7, by Doornkaat

Posted on 2019-09-30, 06:57

Doornkaat Offline

Rank l33t

Rank: l33t
Posts: 2328
Joined: 2017-04-02, 06:50

The external battery pack often had another voltage than the onboard rechargeable battery. If 3.6V doesn't do the trick the PCB is likely designed for a 4.6V external battery.

Reply 7 of 7, by dionb

Posted on 2019-09-30, 17:33

dionb Offline

Rank l33t++

Rank: l33t++
Posts: 5761
Joined: 2017-12-23, 15:35
Location: Amsterdam

Looks like that was it. Replaced the 3x 1.2V rechargeables with 3x 1.5V alkalines and it now holds its settings!

Also looks like there are diodes in place to prevent charging of the external battery, so I can probably just leave it like this 😀

Go to top of page Go to top of page

Back to General Old Hardware

Main menu

Common searches