Have a little problem... Two batteries have died in my TL over the last few months. In both cases it looked the same: suddenly no power at all, 6V across the battery.

The first one was almost 3 years old, so I didn't think much of it dying, just replaced under warranty and moved on.

But the new one died this weekend too, and died spectacularly: it was swollen, apparently leaked acid from the top, was standing in a small puddle of it, the acid managed to go through the tray down to the battery bracket and the frame, ate away the paint, and I had hell of a time cleaning that mess up.

When I went to the store for a replacement I was pretty sure that I had to get an alternator too soon because signs of overcharging were hard to ignore. But once I came back with a replacement and measured voltage across the battery at idle, 2000 and 3000 rpm, I didn't see voltage going higher than 14.5V, which I think is quite normal. So right now I'm at a loss for what to do. I don't want to replace the alternator just for the sake of it, but I lack a test case to prove that it's wrong.

Is there something that can be checked in alternators/regulators other than the charging voltage? Is that the only thing that is being regulated, or are there also some current sensors, or perhaps the alternator is supposed to shut down completely when the battery is full?

 

Most of the shops offer testing. Be prepared to pay something around $20. You should get printout similar to this one:


It tests for voltage, charging current, it should be able to determine if any of the diodes are short or open, etc.

Maybe it was defective battery? Sometimes they sit on shelves for years, and shop just slaps new sticker on it.

Test for parasitic draw too, as our cars are known for HFL modules going bad, and discharging battery when car is off.

 

More likely than not it's a parasitic draw that's killing your battery.

 

Nah, if there was a parasitic draw, the battery would just discharge rather than swell and leak acid.

I'm reading that there is a control signal from ECM to the regulator that selects high vs low charging rates. 14.5V on battery would be high rate. So now I need to check that signal and see if it ever commands low rates, and if it does, whether the alternator complies.

 

 

No, it's still the original one. I am ready to replace it, I just need a justification. After all, it does put out proper 14.5V.

There are a couple of things there apparently that can go wrong besides alternator... There is an ELD in the fuse box that detects electrical load and tells ECM. Then ECM depending on the load and a few other things can issue a high/low signal to the regulator on ALTC wire, and finally regulator should either output high voltage (14.5V) to sustain the load and charge the battery or low (12.5V) to just keep battery from draining. It might be that the regulator never goes into the low mode, and hence the overcharge. I'll need to trace this link from alternator back to ELD and see if the low is ever commanded or the load is properly reported. Hopefully it won't be the ECM who's at fault here.

 

So I moved forward with this a little. Did some reading, hooked up a scanner, and this is what I saw...

- Alternator appears to correctly report its load to ECM. The percentage varies depending on the load.
- ECM is always requesting the alternator to generate high voltage, i.e. 14.5V, it never drops down to the low 12.5V mode.
- I'm quite surprised by the ELD (load detector) signal. According to the scanner, ELD tells ECM that when everything is turned off (climate, lights, audio, navigation), the electrical load is about 22A; and once I turn on headlights and rear defogger, the load jumps up to 60A.

The ELD readings look strange to me. If someone knows what nominal electrical load in an idling 3G TL should be, could you please compare it to my readings above and let me know? If ELD is overestimating the load, that could explain why ECM is continuously running the alternator in high voltage mode.

 

Well, ELD's signal is legit. When it says 20A, I measure 40A on alternator's output and 20A on the battery. When it says 60A, I have 70A on the alternator and 10A on the battery, thus the numbers add up. So, is it ECM then that's cooking my battery by forcing the alternator to always run high??

 

Did that second new battery swollen up too? Maybe first battery was just bad, like had a shorted cell or something, and you just got unlucky?

I'm pretty sure that if you leave 14V on that battery all the time nothing will happen (isn't that how battery tenders work anyway?) (what is gassing voltage for lead-acids? something around 15V if I remember correctly).

I tried to look up some info in the service manual, but they don't go too far with diagnostic.

 

No, the latest battery's still OK. It may not even fail until summer time, too cold and dark right now, not much power and opportunity for boiling.

I have faint theory about the loads. I do have an aftermarket alarm/remote starter with its own battery. I need to check both to see if the small battery's still OK and if they together are not drawing too much power. If they do, that might cause ECM to run alternator in 14.5V mode all the time. Plus a have a friend with 7G Accord, it's about the same car as 3G TL. I'll hook up to it and check what normal ELD numbers should look like.

 

Checked Accord's readings... Same 22A at idle, and it's also running alternator in 14.5V mode all the time. I've seen some articles saying that one of the conditions for charging low would be intake temperature above 68F, hence the reason for 14.5V now.

So all in all, charging looks normal and I have no idea why the new battery leaked and died so fast.