• Hello MLAers! We've re-enabled auto-approval for accounts. If you are still waiting on account approval, please check this thread for more information.

PowerBook 500 Series Battery Rebuild Failure (Oops, all dead EMMs!)

One other thing... the factory default capacity is 1752 mAH. For all my batteries, it seems like after repair the measured total capacity is stuck at 1752 mAH. I don't see it deviating at all from that value. That seems odd. Not sure what it means by measuring capacity but it doesn't seem right that it's stuck at 1752 always regardless of the battery pack I've used for new cells (which are at 2000 and 2100 mAH).

When I get some time, I'm going to take a look at the System 7.1 source code as well as disassemble EMMpathy to see if I can get some more insight into where this value comes from and/or whether there's something that may be keeping our rebuilt packs from fully charging.
 
One other thing... the factory default capacity is 1752 mAH. For all my batteries, it seems like after repair the measured total capacity is stuck at 1752 mAH. I don't see it deviating at all from that value. That seems odd. Not sure what it means by measuring capacity but it doesn't seem right that it's stuck at 1752 always regardless of the battery pack I've used for new cells (which are at 2000 and 2100 mAH).

When I get some time, I'm going to take a look at the System 7.1 source code as well as disassemble EMMpathy to see if I can get some more insight into where this value comes from and/or whether there's something that may be keeping our rebuilt packs from fully charging.
I think to measure actual capacity, you have to run it right down using Amnesia, then plug in the power. It must then charge right up, after which you can restart and see the reported capacity in EMMPathy.
 
Does it have to Amnesia? I thought that tool just allows you to fully run the battery down. But why does that cause EMMpathy to show a new setting as opposed to normally draining the battery (not fully) and charging it back up again?
 
@alexGS hmm, I forgot about this but were the original stock battery cells AA or A? I'm guessing they might have been "A" since there's still a lot of room left over inside the battery case.
 
After draining with Battery Amnesia, I was watching EMMpathy while charging the battery. We might possibly not be getting to full charge --OR-- we may not be using all of the capacity before the PB thinks the power is too low.

As charging begins, Measured Total Capacity remains at 1752mAH. The remaining capacity starts building up from 0 (basically how much mAH the software feels has been pushed back into the pack) and the % of charged increases accordingly. I noticed the Current Temperature also starts rising.

At around 79% I noticed the Current Temperature is sitting around 54C and rising. This is above what EMMpathy shows as the Max Charge Temperature which is 50C. The % keeps toggling between 79%, 79.5%, 80% as the temperature continues rising all the way to 58C. At this point the Remaining Capacity has gotten up to 1400mAH (which again is what the charging has put back into the pack -- not necessarily how much the pack has -- although if Battery Amnesia drained the battery, the capacity in the pack should be fairly low). Then as the temperature hits 59C, the pack shows Calibrate as the Charge State... and then shortly after the Remaining Capacity jumps to 1752mAH, the Measured Capacity is still 1752mAH, and it stops charging. The Charge State then keeps flipping between Maint Charge (maintenance/trickle charge) and Wait EMM. Right now, it's still doing this and the temperature has gotten up to 60.5C and stopped there.

I'm curious if it's stopping the charge due to the temperature? This is a 2100mAH pack. It's put in 1400mAH.

Most of the determination on pack charge I believe is done via the voltage. I did read somewhere (mentioned this on my PB160 battery thread) that outside of capacity differences between A, AA, and AAA, there's another difference which is the potential for voltage drop during load. This is why I was trying to remember if the stock cells were A size cells or AA. I think they may have been A. If the voltage is less consistent on load with the AAs than the As, I'm curious if it's possible that my pack has about 1/3rd of a charge remaining (+/- a few hundred mAH) but due to voltage fluctuations, it's treated as empty?

The temperature is finally coming down ... now at 59.5C as I'm writing this (and the charge state is still flipping between the maintenance charge and waiting for emm states).

I also read somewhere that trickle charging (ie. maintenance charge?) aren't good with modern NiMH packs after they are fully charged. I think they said any cell with a listed capacity over 2000mAH doesn't have enough headroom to properly handle trickle charging or top off charging continuously after a full charge?

EDIT: additional thing that's interesting. On a full pack, when I fire up Battery Amnesia, it reports that there's only 47-50% left to drain. Even on a full pack. Hmm.

EDIT2: Battery Amnesia is supposed to almost fully discharge a battery by disabling all the system warnings and continuing to put load on the battery to drain it as fast as it can. The last packed I drained was at around 90% charge (based on what System 7 is telling me) and it took 74 minutes for Battery Amnesia to drain it.
 
Last edited:
Answered my own question - duh, just needed to look at the cells I removed last week. The stock cells are indeed size "A".

I guess I could try building a pack with NiMH A sized sells. I see Sanyo's available online. Which means I probably need to buy a spot welder.
 
I'm curious if it's stopping the charge due to the temperature?
This is how you're supposed to detect that a NiMH battery is charged. Early hybrid cars did exactly this and it had the added benefit of being a proper feedback loop so they could then throw an error when the packs degraded.
 
This is how you're supposed to detect that a NiMH battery is charged. Early hybrid cars did exactly this and it had the added benefit of being a proper feedback loop so they could then throw an error when the packs degraded.

I was only familiar with the delta peak method (via voltage). I see the delta temperature method is the preferred one.

With this then I'm curious how to maintain the calibration (if required) of the battery temperature sensor? I've been reusing the ones that were already present within the stock battery casing thinking they were used only to monitor the temperature, not that they were critical to the charging process. I get that it's not necessarily the absolute value but the rate of change, but nonetheless.
 
With this realization, I may have to revisit using 2A for VBATT. The cells I'm using can support it for fast charge, but it's likely it's causing increased temperature and probably premature cutoff then? I'm recharging a second battery right now and it's been fully drained. 880 mAH has been pumped back in so far (it's reporting 50.5% capacity) and the temperature has already gone to 51C.

Probably should have used the stock 1A instead of getting greedy and wanting to charge faster with the 2A.
 
Well hmm, with the parts 540c I received a month back, it included what was supposedly a dead power brick. The intent was to save the color matched power cord and the actual plug (that goes into the PowerBook) for future use. I just tested it and it's showing 16V on both VMAIN/VBATT, and the ground pins seem ok too. I just tried it on my 540c and it looks like it's working?? Maybe it has some other faults? But after my batteries cool down, going to try charging using this stock charger and monitor whether I see any difference.
 
It may be intermittent, both of mine were. Pay close attention to see if one or both power rails fail during use.

Ok.

I've also ordered a set of 8 new Sanyo 2100mAH "A" size NiMH cells (flat top). And a spot welder with some nickel battery tabs. I've got one more full enclosure with what I am hoping is a working EMM board (albeit a Panasonic EMM-P one). Not sure if it really matters that I'm using Sanyo cells since these are the wrong capacity anyway. I guess I can also try with my existing/working three EMM-S Sanyo boards. Mouser has Panasonic branded ones which is another possibility.

I mean, I'm getting like 1.5 hours with a single one of these batteries right now, 3 hours combined with two, so that's honestly good enough. So this is really just to satisfy my curiosity.
 
It may be intermittent, both of mine were. Pay close attention to see if one or both power rails fail during use.

LOL, nevermind... with it plugged in for about 15 minutes, I can smell some fishiness (the usual capacitor leak smell). I'm going to shut it down while it's still semi-working so that I can replace the caps later and still have a working unit. Bummer.
 
That's all of them at this point unfortunately. Give it a recap and maybe it will stay working.

Yeah, I have another one that's completely dead. Maybe I'll recap both at the same time and see if the other one comes to life. Project for sometime in the future though.

I'll have to dig around for a 16V 1A power supply otherwise I will try this test with my bench power supply, with a rigged connector.
 
I looked but I can't find a 16V 1A supply online with the same barrel width as the old ThinkPad chargers which I've used for my adapters. I guess I could wire it up directly though.
 
Alright I bought a 16V 1A adapter, plan is to chop the connector off and directly solder the wires into my hacky adapter. If that works reliably then I'll go ahead and design a 3D-printed cage for it at some point.
 
Yeah, finding a 16V 1A adapter seems difficult. I found one on eBay earlier for $7 so I decided to buy one to see if it will work.
 
Back
Top