VOGONS


First post, by Chupperson

User metadata
Rank Newbie
Rank
Newbie

Hi, I recently removed the leaking Varta battery from my 286 and am trying to connect an external 3.6v lithium battery to it with a battery holder wired up to the external header. It looked like there was charging current going to the positive terminal, so I added a 1N4002 diode in line with the positive lead with its negative end facing the board, but I'm still getting the same reading when I test it. The voltage on the positive terminal slowly rises to about 390mV, then I get OL for a second followed by 0.9V, then it starts back at zero and rises again. Video of meter output is at https://www.youtube.com/watch?v=t1AgC6aCiP0
Am I connecting something wrong? Do I need the diode? Is this enough voltage to worry about?
The motherboard is an unknown make similar to the Biostar MB-1212V but missing some components near the ISA slots.

Reply 1 of 3, by Horun

User metadata
Rank l33t++
Rank
l33t++

so I added a 1N4002 diode in line with the positive lead with its negative end facing the board

You connect the diode off the Positive terminal of battery to + of the Ext. batt connection "in line" with Cathode to the batt connection.....
It should not allow any reverse bias more than a few uA. Did you read in mA ? The actual voltage in this case is not as important as the reverse bias current.
A good test would be a 100ohm resistor in place with the diode in place of the Ext battery and read the current. just my opinion...
A good test would be a 100ohm resistor in place of the Ext battery with the diode and read the current. just my opinion...
fixed grammar

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 2 of 3, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie
Chupperson wrote on 2023-02-12, 00:27:

Hi, I recently removed the leaking Varta battery from my 286 and am trying to connect an external 3.6v lithium battery to it with a battery holder wired up to the external header. It looked like there was charging current going to the positive terminal, so I added a 1N4002 diode in line with the positive lead with its negative end facing the board, but I'm still getting the same reading when I test it.

How did you measure that charging current? Because what you are showing on the YT video is measuring voltage, not current. Usually motherboards will not charge the extrnal battery - some can be jumpered like that but it's rare, most just can't due to a diode like what you added, except it's already on the mobo.

Chupperson wrote on 2023-02-12, 00:27:

The voltage on the positive terminal slowly rises to about 390mV, then I get OL for a second followed by 0.9V, then it starts back at zero and rises again. Video of meter output is at https://www.youtube.com/watch?v=t1AgC6aCiP0
Am I connecting something wrong? Do I need the diode? Is this enough voltage to worry about?

What you are seeing, most likely, is a small reverse current through that diode (and possibly another one on the mobo) that charges some parasitic capacitance. Then, when the voltage causes the meter to change ranges the new strap is probaly 10Meg input resistance (vs. gigaohms on the lowest setting, which is most likely direct input to op-amp) and it discharges or removes the stray capacitance, this causes meter to go back to lower range and everything repeats. Always know what you are measuring, and how.

I'd suggest to measure the actual charging current, if any, properly using mA or even uA meter. Anything below 10uA you can ignore. 10-100uA is not great for lithium cells but typical 2x or 3x AAA/AA pack will eat it and not care. In fact it might just last a bit longer with good quality cells. Above 100uA you want an extra diode because, while it won't overheat the pack, cells might start leaking eventually due to excessive corrosion of the outer shell. And small lithium cells will, eventually, get damaged by such current, and start thermal runaway.

Reply 3 of 3, by Chupperson

User metadata
Rank Newbie
Rank
Newbie

You're right, as a layman I wrote "current" when I should have written "voltage." Now that I know to test it, I am reading 0 uA of current. In fact, with a 100 ohm resistor in place of the battery, I'm reading 0 volts on the battery header, so it seems like I'm safe to install the battery and see if my settings and clock get power.