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

[3lectr1cPPC]

I tried to rebuild two PowerBook 500 Series batteries. What could go wrong?
Well, I didn't light anything on fire which is a good start, but I don't have working batteries either.
I used pre-built 8-cell AA NiMH packs that @jmacz already used successfully on his own rebuilds.

Both batteries register proper voltage on the + and - terminals now, but neither of the EMM boards (Energy Monitoring Module, early complex "smart" BMS) in the batteries want to come online and talk to the computer. The computer will run off of the batteries if I unplug it, but it does not detect that a battery is in the computer. This means that Mac OS will put the computer to sleep as soon as it is able to. I cannot cold-start off of the batteries. If I lock up the system so Mac OS cannot put the computer into sleep mode, it will stay powered off the batteries, but I cannot recharge them in their current state.

Background on the batteries:
Both are the earlier of the two main revisions of these batteries (large internal ribbon cable). They came out of a PowerBook 540 that I got around August 2022. When I got the batteries, they did not show external corrosion at all, and were detected by the computer. I ran the three programs mentioned earlier and they also saw the batteries and ran tests. I remember both Lind and EMMpathy found errors that were able to be corrected in both batteries. At that point, both EMM boards tested 100% fine in all three tools, but the original cells were too tired to run the computer. A couple months after this, I cracked the cases open on both batteries in order to remove the original cells.
One of the EMM boards was perfect, and the second was was nearly perfect, with the only damage being some corrosion around the + battery connection on the internal ribbon, and a partial tear in the same area. It still has continuity.
After cell removal, the EMM boards sat around for over a year before I attempted the rebuild.

Diagnostic tests:
Lind BU500 - Fails to detect the battery.
EMMpathy - Fails to detect the battery.
Apple Intelligent Battery Update - Detects the battery, immediately throws an error that it cannot be updated.

Tests to the EMM boards:
Battery 1:
Battery Negative and the pin two to the right of it registers 5V, suggests the EMM board is receiving power and is exhibiting normal behavior.

Battery 2:
Same pin registers nothing, indicating the EMM is seemingly dead. Path from external contacts to the EMM board itself have continuity (in both batteries as well).

@jmacz had this same non-detected issue that turned out to be an EMM fault, apparantly with the Apple chip. Pin 28 (bottom right) was shorted to battery negative/ground. Neither of mine have pin 28 shorted to ground.

Other variables:
I have not replaced the one cap between the battery bays that I believe might have something to do with the batteries. Could this cap being bad cause non-detection? I'll bet said cap is on its way out in both my laptops.
My power supply situation is weird. I have my custom-built adapter, which only powers the computer. VBATT for battery charge is not connected as it requires a regulated 2A supply. I do have another OEM charger that's been recapped, but it's still faulty. On that one, VBATT works perfectly, but VMAIN will drop from 16V down to under 5V after around a minute of operation, and a quiet buzzing noise will begin coming from the supply. I figured that I could probably charge the batteries with the OEM supply until I get a better solution, but I can't do that without getting the computer to recognize the batteries. In the short time where VMAIN is working on the OEM charger, I still get the same results with non-detection.

Anyone know where to even start looking for faults? Finding working EMM boards isn't exactly trivial nowadays with battery corrosion at play...

 

[3lectr1cPPC]

Nothing obviously visually off on either.

 

[jmacz]

Tagging @alexGS as he's gotten batteries to work as well.

So just to be clear, you have two chargers:

• Custom Built - vmain works but vbatt is not hooked up so it's driving 16V on the vmain pin only.
• Faulty OEM - vbatt works but vmain conks out after about a minute.

I assume you've been using the custom one with vbatt disconnected to test the batteries?

When I was rebuilding my batteries, I was using a fully working OEM one. I would expect all of the systems within the laptop (specifically the battery related ones) to be powered by vmain but I'm curious whether the lack of power on vbatt is causing the system to not communicate with the batteries? I never tested with a non-working vbatt side.

I've got some stuff I have to take care of this evening but let me see if I can try hooking up my spare connector to my bench power supply and only give it power on the vmain side and then see what happens with my laptop and batteries.

That said, the battery that has 0V inner connection pad, that doesn't seem normal. Like I mentioned, the only case where I saw that was on my dead battery that had lots of issues, the faulty/shorted Apple EMM chip, a dead transistor, and a dead resistor.

I guess the current guesses are...

• Battery with 5V on inner connector
  • Maybe the capacitor on the board
  • Maybe the fact that vbatt doesn't work
• Battery with 0V on inner connector
  • None right now

 

[3lectr1cPPC]

I did boot off the OEM charger for my first test a few days ago with the first battery and it had the issue then as well so I doubt it's that, but I don't honestly know. The VMAIN rail seems to run for a shorted amount of time each time I power the supply, even after leaving it unplugged overnight so by now it probably wouldn't even be able to start up all the way before VMAIN dies. What I've done because of this is to instead boot off my VBATT-less adapter charger, then swap out quickly to the OEM one, and then I'll run a test. Now it's at the point where I can barely start a test before VMAIN dies on the OEM supply though, it's bad. Wish I could get that resolved. My 2nd supply that I canabalized for the adapter was even worse, it would have both rails die after a few minutes. Both recapped. Ugh.
The problem is that this is all tempting me to buy your PowerBook 520 that comes with a working OEM supply, when I really shouldn't. I don't need a 520. Yet I'm still tempted...

The latest interesting discovery which I mentioned in the first post is that Apple's Intelligent Battery Update utility actually DOES detect BOTH batteries, but immediately fails anyway. I assumed the instant error it gave meant it couldn't detect it until I thought harder about it and just ran it with no battery. Instead, it then told me to insert a battery, which I did, then about 30 seconds later it threw the error. So it sees something. That could mean it's able to communicate with the EMM on some level, or it might just actually be checking for voltage at the battery unlike the other utilities and the OS which clearly don't.

You know, I'm starting to think that these "intelligent" batteries may be a bit less smart then they want me to think they are...

Worth noting that I tested all the ceramic caps on at least one of the EMMs and found nothing shorted. Maybe it's just time to finally make a mouser order and get those two caps changed.

 

[3lectr1cPPC]

So just to be clear, you have two chargers:

• Custom Built - vmain works but vbatt is not hooked up so it's driving 16V on the vmain pin only.
• Faulty OEM - vbatt works but vmain conks out after about a minute.

Correct. The OEM one is recapped, and had the VMAIN issue both before and after. Old caps were leaking though.

 

[jmacz]

The problem is that this is all tempting me to buy your PowerBook 520 that comes with a working OEM supply, when I really shouldn't. I don't need a 520. Yet I'm still tempted...

Well, you're saved (kinda) because the PowerBook 520 sold a few weeks ago.

 

[3lectr1cPPC]

Ah, phew. Was still on your list when I checked a few days ago.
On purchasing a 2A regulated supply - can that be any supply that's RATED at 2A, or will the battery circuit try to draw more than that? Do I need some special kind of supply that will limit the current, or do they have that built-in by default usually?

 

[jmacz]

let me see if I can try hooking up my spare connector to my bench power supply and only give it power on the vmain side and then see what happens with my laptop and batteries.

I tried powering my laptop with a cord that only had the vmain side powered, and I didn't notice anything special. But on second thought, this wasn't a valid test. The battery systems work even without any power cord plugged in (powered by the battery). Where it might matter is if the battery wasn't charged enough and can't supply the EMM with enough power, but that's not true in your case. So scratch this.

On purchasing a 2A regulated supply - can that be any supply that's RATED at 2A, or will the battery circuit try to draw more than that? Do I need some special kind of supply that will limit the current, or do they have that built-in by default usually?

I don't know enough to say. I purchased a 16V 2A rated power supply and when I checked the current drawn through it by the battery, I saw it pulling about 1.9A. The one I got had over current protection (along with all the other protections).

Another option if you want to quickly get your packs charged outside of the charger, is to pick up a NiMH RC charger. They usually come with alligator clips to make it easier to charge your pack. I think they are cheap too.. usually like $15?

 

[3lectr1cPPC]

I figured the charger probably wasn't an issue here. Might cause problems with charging though. Thanks for confirming!

What supply specifically did you buy? I can just get that one.
RC charger probably wouldn't work since I've soldered the battery wires together to the ribbon cable off the EMM. Could probably make it work but it would be inconvenient to take the battery apart for charging every time. I'd rather just get my adapter set up for it, or better yet, actually manage to get an OEM charger that actually works right.

 

[GRudolf94]

Like Egret/Cuda and friends, EMM is presumably a 68HC05E1. I have the board somewhat disassembled with values noted down, but haven't actually started on reverse engineering it. You can check for the basic things as in "it's getting power to the right pins" (Vddsyn being power to the on-chip PLL clock synthesizer wired to the crystal on pins 3 and 4), and "there is a clock on 3/4", even maybe assert reset by shorting the line to ground. The most likely explanation is that boards left around somehow got ESD'd?

 

[3lectr1cPPC]

Chip looks like this under the label, sure looks like the EGRET. Photo from jmacz. I also notice some component differences between his later revision board and mine. The colorful ones (diodes?) on the right are arranged much more normally compared to mine.

No idea what damaged mine, but it's frustrating.

 

[jmacz]

What supply specifically did you buy? I can just get that one.

I picked up this one:

https://www.amazon.com/gp/product/B097C67KLD

Cracked open the case, tossed the cable, and wired it with the 16V 4A I'm using for the vmain.

 

[jmacz]

@3lectr1cPPC any progress or any new insights?

 

[3lectr1cPPC]

Nope. I haven't touched it in a few days because I have no new ideas.

 

[jmacz]

If that chip is a 68HC05E1 as @GRudolf94 mentioned, I guess it wouldn't hurt to check whether it's getting any input power? As mentioned previously, pin 14 (Vss) is ground, pin 13 (Vdd) should be power. You might try checking that to see what voltage you see if any. You might also check pin 1 and pin 2 as well.

I had mentioned on my faulty battery, pin 28 was shorted to ground. I'm not certain that's an issue anymore. Looking at the datasheet, pin 28 is used to connect an external filter capacitor and if unused, should be tied to Vdd or Vss. Both my working batteries had pin 28 not tied to ground and the faulty one did. But it's no longer clear to me that that's certain to be a fault. It's possible just the chip's sram/rom was corrupted.

What's interesting is the chip has a self-test mode. The datasheet describes the values needed to be provided to /IRQ (Pin 1), /RESET (Pin 2), and PB1 (Pin 11) to trigger a self test, with the results showing up on PA0-PA3. The results can provide info on whether it's bad I/O, bad RAM, bad timer, bad ROM, bad interupts, bad device, or everything's good.

 

[3lectr1cPPC]

Pin 13 and 14 both show no power whatsoever, on both batteries. Soooo, what is responsible for downconverting the battery's voltage to whatever voltage the EMM chip needs? (I'm assuming 5V?)

 

[jmacz]

Pin 14 should have no power, it's ground (Vss) but pin 13 (Vdd) I would think should have power as long as it's connected to the battery cells as that's what is powering the chip and I think the EMM is on all the time as long as the battery has charge? I will see if I can take one of my working batteries apart later today to confirm what I'm seeing on 14.

 

[3lectr1cPPC]

I meant that I had one probe on 13 and the other on 14, as that should give me the voltage reading off 13 as 14 is ground. Is that correct?

 

[jmacz]

Yeah, I think that's right. I'll check my battery.

 

[3lectr1cPPC]

No, it couldn’t be that… that’s too easy…


D’oh!
That’s one down, the one with no power to the EMM. Now I’ve gotta figure out if it’s gonna charge with my weird charger situation, and the other battery.