VOGONS


First post, by Keatah

User metadata
Rank Member
Rank
Member

A couple years back I removed a corroded & leaky Varta NiCd battery from my 486 and replaced it with a Lithium Ion coin cell. I used two simple wires to move it off-board to guard against any future damage.

I'm not too pleased with the performance, 2 years into it the battery seems to not charge very well and has a short life of just a few months. Even at the low power drain of a CMOS clock. Internal resistance.. Cheap china crap, you know..

Looking for suggestions for replacement. I want it to stay rechargeable (trickle charge). I heard good and bad things about SuperCapacitors. I was also wondering about using a 3.6v NiMH batt. But I hear they self-discharge. And are still expensive at $8.00 a pop.

Suggestions and discuss please?

Reply 1 of 7, by candle_86

User metadata
Rank l33t
Rank
l33t

don't use a coincell unless you've disconnected the charging circuit for starters, i use a triple battery pack on my 486, it uses 3 AA batteries, can't speak to lifetime as I've only had it hooked up for a year, but it works fine on the external header, you likely damaged your coincell by having it connected fully including to the charging circuit, if your deadset on using a coincell hook it up to the external battery header, that won't try to charge it.

Reply 2 of 7, by Keatah

User metadata
Rank Member
Rank
Member

I want the circuit to charge the coin cell. I completely failed to mention the coin cell is rechargeable. Sorry about that. And it's gone through a few cycles already. It just isn't aging well and seems to self-discharge either in or out of circuit over a few months. So I'm looking for a battery technology/chemistry that won't self-discharge as rapidly.

The 3-cell AA alkaline pack I'm using started out about 4.72v. It's down to 4.43v after exactly 3 years to date. That's with a 7uA - 15uA current draw. Typical for the "Chips & Technologies" 82C206 CMOS settings & clock part. And likely similar parts of the same era. A good performance at 0.1v drop per year. It's likely safe to say it'll last 6-10 years before dropping below 2.8v.

Reply 3 of 7, by candle_86

User metadata
Rank l33t
Rank
l33t

the issue is the CiCd's don't charge the same as Lithium and it can cause damage to the battery, its well discussed in these forums not to do that

Reply 4 of 7, by Horun

User metadata
Rank l33t++
Rank
l33t++

Agree with candle_86 !
The charge circuit for NiCd is different than Lithium and those old boards were designed for NiCd. Is why the Lith did not last and lucky you did not blow up the Lithium batt from wrong charge circuit.
Many of us use Alkaline AA/AAA or Lithium (non rechargeable) AA for old boards, they last 5 years or more typically, about same as a Lith coin cell on a new motherboard.

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 5 of 7, by Keatah

User metadata
Rank Member
Rank
Member

I did a side-by-side test 3-year test. The 2 rechargeable lithium cells not installed, but kept as a reference & spares outside the computer, exhibited the same increase in self-discharge. What would explain that?

Don't see how light trickle charging a battery with just a handful of uA could cause anything to blow up. At worst, small leakage due to chemistry change - which hasn't happened.

Reply 6 of 7, by digistorm

User metadata
Rank Member
Rank
Member

Lithium rechargeable batteries will self-degrade over time, even if they are not in use. It depends on the chemistry and the charge at storage time. They will not last as long as traditional NiCd batteries.

Reply 7 of 7, by Keatah

User metadata
Rank Member
Rank
Member

That's what I discovered. How about NiMH ? Ebay has these 80 mAH NiMH cells in a 3 x 1.2v config for 3.6v. And I heard about SuperCapacitors. Seems 1 Farad @ 5.5v is what's been used.

Want something that won't degrade over just a few years. Like to see a 10-year life (or more).