Hi, all
I got an external battery holder today to plug into my 386DX40. After about half an hour of use I saw small smoke coming out of the case. Turned out, one of minus connectors started melting through the plastic.
Am I doing something wrong, or is this case just not suitable for this kind of motherboard?
did You prevent them from getting charged?
wrote:did You prevent them from getting charged?
How do I do that?
the thing is the following, normally the onboard battery should get charged, so i You connected these to the same traces where the old battery was connected You might need a diode or something similar i think; if You attached these to a dedicated header though You'd have to check with the manual whether charging is prevented already;
I'm a bit surprised a motherboard would need ~6V on the CMOS battery connector, if those 4 AAs are in series.
wrote:I'm a bit surprised a motherboard would need ~6V on the CMOS battery connector, if those 4 AAs are in series.
I wouldn't think so either, 4 fresh alkalines is close to 6.5v which may be enough to damage the CMOS.
I have a couple of 486 motherboards that require 6VDC when using an external battery (it's specified in the manual), and they definitely do not keep the time with 3 or less cells. I don't think it's unusual, especially considering those 4AA external battery packs for PCs were quite common back in the day.
Besides, I remember reading a thread in the Vintage Computer Forums that there usually is a diode connected to the external battery input so there you have a voltage drop, and the input voltage for the circuits powered by the battery is well within operating values (don't have the link to that thread at hand though).
In any case, it would be good to check that the board isn't trying to recharge the batteries through the header you're using. Can you boot up without the battery plugged in and measure the voltage between the pins of the ext. bat header you're using?
^ yes, I can confirm this 6v business, here is a pic from my 286 mobo manual
May be advisable to actually use external NiCad batteries (2x 3v). The problems with corrosion (which do not manifest in these batteries until 10-25 years later) can be "solved" by sealing the battery holder with something like tape, and keeping it away from motherboard components. I think this would be the best solution as the batteries would never need replacing until they fail in that aforementioned 10-25 year lifespan (probably more/less depending on the brand)
The issue with AAs is that they won't be recharged (unless you put a stupid amount of effort into using NiMH charging, or you use NiCads). So at some point a lot sooner, you will have to replace it (and loose your CMOS settings of course).
wrote:The issue with AAs is that they won't be recharged (unless you put a stupid amount of effort into using NiMH charging, or you use NiCads). So at some point a lot sooner, you will have to replace it (and loose your CMOS settings of course).
Alkaline AAs have a tremendous amount of capacity and even with their relatively high self discharge rate compared to lithium batteries , I would think they could last at least a good decade before self discharge kills them. I would really recommend sticking to lithium manganese dioxide batteries for the CMOS if possible since they seem to be the be proven to work well in the application and don't typically develop leaks over time like alkaline, carbon zinc, or even Nickle metal batteries.
