Battery light on, alternator OK

oldskewel · Jun 8, 2019

UPDATE in AUG 2022 - following extensive debugging and no repairs at all, this problem eventually fixed itself and has been fine for over a year now. Still possibly useful info in here for others, but that was the "answer" in my situation. Still on the original alternator at 136k miles, having replaced the alternator pulley (OAD) at around 90k.

Also to be clear for people finding this with a common alternator failure ... this is not that. In a true alternator failure, you will have electrical problems where the alternator does not generate current and maintain voltage as it should, and you would probably get a warning light and message. In my case, I got the light, sometimes the message, but NEVER had any electrical problems I could measure.

=======

2011 Odyssey LX, about 110k miles, no problems at all until ...

the red battery light (that is supposed to indicate a problem with charging) has been coming on intermittently starting a week or so ago, along with the CHANGE bATT message. When this first happened, I checked voltages = all good, the battery is less than 2 years old = good, I pulled the 2P connector on the battery negative terminal to reset the battery management, all connections tight. So I just told my wife (who drives this car) to pay attention and keep me posted.

Now that she's on a trip a few hundred miles away, the one-time problem has become intermittent and is now happening more frequently. Following instructions over the phone with me just now, she confirmed 12.2V battery voltage this morning before driving, then the red light came on while driving, then at the end of her brief trip, without shutting the engine off, she went and confirmed 13.9+V at the battery, while idling. This done using a voltmeter on the battery posts.

So it seems to me that the alternator is good and the battery is good.

Any ideas on what could possibly cause the red charging indicator light to come on like this?

HondaUpNorth · Jun 8, 2019

I would replace the battery. You may have a bad battery - rare but can happen. But for some reason I thought the alternator voltage was supposed to be around 14.5V.

Best to take it in and get checked out.

Sent from my iPhone using Tapatalk

bobodaclown · Jun 8, 2019

Had a similar situation. It was a bad alternator, although it tested good. If you decide to replace it from an auto parts store ensure that they bench test the replacement. I got two that were bad. Maybe it was just bad luck. Anyways replaced the alternator all is well.

oldskewel · Jun 8, 2019

Thanks for the ideas, but I mostly need help on the last part of my question there,
"Any ideas on what could possibly cause the red charging indicator light to come on like this?"

I know batteries and alternators can fail, but mine appear to be fine. I know that taking it to someone else might get me an answer which may be helpful or not.

What I don't know is what could cause the red charging light to come on while the alternator is clearly generating sufficient voltage.

I think it is controlled by the PCM, and found another report of a guy with his Odyssey under warranty where the dealer replaced the battery, alternator, fuse box (I'll guess they wanted to replace the ELD and figured they'd overkill it while billing to warranty), and PCM, with a few days of diagnostics by the best the dealership had to offer ... with no improvement. <== And BTW, I don't want to be THAT guy.

Unfortunately that thread got derailed by people offering general alternator advice, and the 3 people ('07, '08, '08) with similar issues never posted conclusions.

https://www.odyclub.com/threads/che...threads/check-charging-system-light-alternator-battery-diagnostics-good.234226/

Ideally someone else out there has had this problem, solved it, and stuck around long enough to tell the solution. I promise I'll keep this updated with whatever answer I find.

I found this interesting article describing the Honda charging system

https://opensiuc.lib.siu.edu/cgi/viewcontent.cgi?article=1011&context=auto_pres

The article is from 2009, so I'll guess that it is relevant to my '11 and the '07-'08 issues I mentioned.

And most of it makes sense, but it appears to have the diodes connecting to ground drawn backwards in the circuit diagram, which throws a little doubt on things. One interesting quote from that:

"The charging system utilizes the L circuit to inform the driver of any charging system faults. Over the years Honda has used two methods for illuminating the charge warning indicator lamp. On older models, the L circuit directly provided ground for the warning lamp if a problem was present. If everything was in spec, the voltage regulator removed the ground by providing positive source voltage on the L circuit. However, on late model vehicles, the ECM sends source voltage to the L circuit. If a problem occurs, the voltage regulator will pull the voltage on the L circuit to ground. If this occurs, the ECM will sense that the signal voltage has been pulled down and it will then send a “charge warning lamp on” signal through the CAN bus network to the gauge control module. In this case, the gauge control module will directly switch the indicator lamp on."

oldskewel · Jun 9, 2019

Update - car made it 400 miles back home now, light did not come on at all during the trip.

I'll inspect things again, but am leaning towards considering the light to be a bogus result of a PCM fault, and will learn to ignore it as long as things run fine.

May install a digital voltage gauge (like I did in my '99) so system voltage is always readily visible.

And maybe I'll swap batteries with one of my other cars to rule that out.

egads · Jun 9, 2019

You cannot test a battery by just checking it's voltage. You have to have a load test.

John Clark · Jun 9, 2019

12.2v is low. Battery voltage should be around 12.6v. When the alternator is charging it should be 13.5-14.5v. I've seen the intermittent battery light when high loads are placed on the alternator like both AC units, radio, etc. Then when loads are reduced the battery light goes out. Checking the voltage with the battery light on is the key here.

If the battery light is on when the voltage is above 13v then something is still wrong with the alternator as the switch for the light is in the alternator. I had this with a Denso alternator on a 2006 Ody. Installed new alternator and it worked great except after about 20 seconds the battery light came on even though charging voltage was 14.5v. Confirmed the alternator was bad, exchanged it with another Denso, and all was well. The owner had brought me that alternator and had confessed that the girl behind the parts counter dropped it on the counter (from about 2 ft) when she was pulling it out of the box.

oldskewel · Jun 12, 2019

Updates:
Swapped in a similarly new (2017), identical battery from another car. No problems in that car / no changes in this one before or since.

Installed a digital voltage gauge so system voltage is always displayed, for debugging and real-time reassurance that things are OK.

Codes read = none. No CEL, other warning lights, etc. The CHANGE bATT came on early in this, and I told my wife to ignore it, so I have not heard more on that. I figure it is fallout from this issue, and concern with charging is more important than the battery, even if it's bad, which it isn't.

No changes in the problem. Displayed system voltage, and everything else points to there not being any problems at all with anything. But the red light still comes on intermittently.

Still hoping for some details on what controls the red light in this car (2011 LX). The thing I cited a few posts above still seems the most detailed I can find:

"The charging system utilizes the L circuit to inform the driver of any charging system faults. Over the years Honda has used two methods for illuminating the charge warning indicator lamp. On older models, the L circuit directly provided ground for the warning lamp if a problem was present. If everything was in spec, the voltage regulator removed the ground by providing positive source voltage on the L circuit. However, on late model vehicles, the ECM sends source voltage to the L circuit. If a problem occurs, the voltage regulator will pull the voltage on the L circuit to ground. If this occurs, the ECM will sense that the signal voltage has been pulled down and it will then send a “charge warning lamp on” signal through the CAN bus network to the gauge control module. In this case, the gauge control module will directly switch the indicator lamp on."

Basically, that seems to simplify to: The PCM sets it, and the VR could be the cause of it. And my interpretation of that vis a vis my situation is that the PCM is probably screwing this up, but as long as nothing else is wrong, I'm on high alert but can't see what else to investigate. Definitely no plans to replace any parts based on a spurious red light.

No other problems with the PCM that I'm aware of.

The alternator is original. I had it out at around 90k miles when figuring out a problem with the OAD (alternator pulley). I inspected the brushes at that point, and figured they'd last to 170k miles based on their length at that time. Everything else looked good with the alternator, and as shown, charging is good. Voltage is ~13.9-14 Volts, regardless of headlights / AC / etc. Being an LX, the electrical load is not as high as some other Odysseys.

Even if the VR is erroneously telling the PCM to set the red light, as long as things are charging as they should I'm inclined to just roll like this.

I have a couple of Foxwell NT520Pro's for other cars, but neither with the special Honda diagnostics software. Never had a need for it since my Hondas have been so good. I could buy the software for $60. Does anyone who has this tool know whether it might be able to shed more light on what/where the problem is?

John Clark · Jun 12, 2019

I can assure you that an internal failure of the alternator can/will turn on the battery light even when the alternator appears to be working and charging normally. I've seen it with my own eyes on a 2013 Odyssey.

If you unplug the alternator connector you'll see the light go out so the alternator definitely sets the light. It sends the signal to the PCM and the PCM then sends it to the gauge module but the alternator controls that light. The service manual actually has this as step 5 in the diagnostic tree. If unplugging the alternator 4 pin connector, then turning the key on, turns out the light then the alternator should be repaired/replaced. If the light stays on when the connector is unplugged then the PCM, or wiring to/from the PCM, is the problem.

I would recommend investing the $60 on the Foxwell software. The Foxwell tool does a pretty good job with Honda. There could be DTC's stored that are not accessed on the generic OBDII side of the computer.

kselfri · Jun 12, 2019

I had gotten my tested at the autoparts store, along with the battery, and they tested good. The alternator was actually bad (diodes). It would also show 14 volts w/ my voltmeter. Testing the alternators in these vehicles is not straight forward. You must have the high-end testing equipment the reputable shops use, not the local auto parts stores. Nor can you rely on a voltmeter reading as with older cars. If your charging system warning light comes on, it is probably your alternator. I wouldn't replace the alternator with a factory Honda one in this case, since they're not that good. If you replace it, use Denso ONLY. I have a Corolla w/ a Denso alternator which has lasted 350K miles.

bobodaclown · Jun 12, 2019

If it throwing the light there is something wrong with the alternator. Get it bench tested. I think the cable is m30 to beech test it. I went through a crazy process when I replaced mine 2012 with 120k miles on it.

oldskewel · Jun 13, 2019

John Clark said:
I can assure you that an internal failure of the alternator can/will turn on the battery light even when the alternator appears to be working and charging normally. I've seen it with my own eyes on a 2013 Odyssey.

If you unplug the alternator connector you'll see the light go out so the alternator definitely sets the light. It sends the signal to the PCM and the PCM then sends it to the gauge module but the alternator controls that light. The service manual actually has this as step 5 in the diagnostic tree. If unplugging the alternator 4 pin connector, then turning the key on, turns out the light then the alternator should be repaired/replaced. If the light stays on when the connector is unplugged then the PCM, or wiring to/from the PCM, is the problem.

I would recommend investing the $60 on the Foxwell software. The Foxwell tool does a pretty good job with Honda. There could be DTC's stored that are not accessed on the generic OBDII side of the computer.

Thanks for all the info.

Behavior is still the same. No apparent problems at all other than that the charging light comes and goes, even though the voltage gauge is displaying 13-14V when running.

I went through many of the debugging steps in the manual this evening. Followed through to the test you mentioned (in bold above), and ...

one time I got the light staying on, which points to further testing of wiring and the PCM, and ...
another time I got the light going out, which says to repair or replace the alternator.

I guess things will get more conclusive once the problem stops being intermittent. But thinking it through, trying to figure out how things are supposed to work based on the diagnostic process ...

We only get to this point in the diagnostic flow chart if the light stays on even when the engine is running, so the assumption is there's a fault somewhere, either in the alt or in the PCM / wiring / beyond. This test intends to choose one or the other.

With the alternator disconnected, engine stopped,
- if the light stays on, the manual points to the PCM / wiring / beyond
- if the light goes out, the manual points to the alternator.

Suggests that the alternator, when connected, is supposed to turn on / off that light (via the PCM, which matches the info cited ia posts above). When things are working properly, the light should not come on when the alternator is not connected.

If it was the PCM turning it on erroneously, and it was doing it all the time, not intermittently like I have, then it would still be turning it on now, when the alternator is disconnected. So that's why the diagnostic points to the PCM / wiring if the light stays on when the alternator is disconnected. The alternator can't be doing it (since it's disconnected) and if something is turning it on, it must be the PCM/wiring.

Conversely, if the light goes out when the alt is disconnected, that means the PCM/wiring is working correctly. So it must be the alternator turning it on when connected and running, which indicates a problem with the alt.

An intermittent alt failure would
- intermittently turn the red light on when running - got this
- when disconnected for this test, would ALWAYS turn the light off, whether or not the alt fault is active - I do NOT have this, instead it is intermittent

An intermittent PCM/wiring failure would
- intermittently turn the red light on when running
- when alt disconnected for this test, would INTERMITTENTLY turn the light off, whether or not the PCM/wiring fault is active - I do have this

Since I'm seeing intermittent results for this test, I'm sadly thinking it is not the alternator, but rather the PCM / wiring

And when I get to that point, I'm reminded of the other posts, mentioned above, where the dealer, probably following the same diagnostics / logic, and with unlimited parts and $$, replaced PCM, fuse box, alternator, etc. with no improvement.

Comments on the logic here? Would love to be wrong.

Still wondering how bad it could be considering it is putting out 13-14V at all times. Will probably keep trucking along until the problem becomes less intermittent. At the end of the day, if I have a bogus charging light, and no other problems, I can live with that, given the digital voltage gauge.

nolanelana · Jun 13, 2019

Check the VTEC spool valve for leaking oil. It is directly above the alternator. When the spool valve leaks, engine oil drips into the alternator, causing intermittent internal shorts and failures. This causes the “Check Charging System” indicator to come and go. Eventually, it destroys the alternator. If the spool valve is leaking and replaced soon enough, you might save the alternator.

oldskewel · Jun 13, 2019

nolanelana said:
Check the VTEC spool valve for leaking oil. It is directly above the alternator. When the spool valve leaks, engine oil drips into the alternator, causing intermittent internal shorts and failures. This causes the “Check Charging System” indicator to come and go. Eventually, it destroys the alternator. If the spool valve is leaking and replaced soon enough, you might save the alternator.

No problems at all with that. Dry. Good.

But considering external factors that may be relevant ... the only thing I can think of is that about 5k miles ago I did a timing belt change and valve adjustment, all of which went smoothly, with no problems at all. In doing that, I disconnected the wiring to the alternator to peel back the wiring that goes over the front cylinder head, giving better access so I could do the valve adjustment carefully. If there is a problem anywhere that is not in the alternator, I'd guess there is a flaky wire in there somewhere.

Maybe a good next test will be to try wiggling all the wires to see if that can make the light come and go. As of now, it comes and goes pretty randomly. E.g., 400 mile drive home a few days ago = no light the whole way. Driving to work Monday, after swapping battery and installing voltage gauge = no light. Driving home = light. Starting the car on the street yesterday = no light; by the time I had it in the driveway to work on it = light. No apparent correlation with alternator load or temperature.

escape · Jun 19, 2019

I had the same problem on a 2013 back in 2019. It turned out to be the thin copper plate connecting the positive wiring harness to the battery. Read on if you care for more detail. I had the alternator and battery checked twice by the local Advanced Auto for free, both times their tester said both battery and alternator were good. Then I took it to an auto shop I've used for years and trust. The owner brought out his $4k tester and checked it for free. It said all is well. The battery was three years old so I decided to replace it for good measure. I bought one on sale from KMart which meant I had to install. When I removed the positive terminal I noticed how thin the copper plate that connected the heavy red wire to the metal round piece that goes over the terminal. When I went to clean off what little oxidation there was I bent the flat copper ever so slightly and it snapped in two. I could not believe how thin that copper was. And upon further inspection I could tell where the copper was deteriorating, hum - copper to metal with electricity flowing - not good especially when there is less copper than metal. Ask a old telecom guy when negative voltage is used to an old phone circuit. Great, now a little preventive maintenance replacing the battery turned into a major repair. The condo association fines for any auto repair work they catch you doing so I had to have the car towed to the shop mentioned above. Cost to replace the harness was around $125. On the positive side, I have not seen the battery indicator flash ever since. I strongly believe that the problem was the harness at the battery end. The after market harness's copper was more than twice as thick as the original.

alphonse · Jun 19, 2019

Would a slipping serpentine belt or failing tensioner trigger the battery light?

oldskewel · Jun 19, 2019

escape said:
I had the same problem on a 2013 back in 2019. It turned out to be the thin copper plate connecting the positive wiring harness to the battery. Read on if you care for more detail. I had the alternator and battery checked twice by the local Advanced Auto for free, both times their tester said both battery and alternator were good. Then I took it to an auto shop I've used for years and trust. The owner brought out his $4k tester and checked it for free. It said all is well. The battery was three years old so I decided to replace it for good measure. I bought one on sale from KMart which meant I had to install. When I removed the positive terminal I noticed how thin the copper plate that connected the heavy red wire to the metal round piece that goes over the terminal. When I went to clean off what little oxidation there was I bent the flat copper ever so slightly and it snapped in two. I could not believe how thin that copper was. And upon further inspection I could tell where the copper was deteriorating, hum - copper to metal with electricity flowing - not good especially when there is less copper than metal. Ask a old telecom guy when negative voltage is used to an old phone circuit. Great, now a little preventive maintenance replacing the battery turned into a major repair. The condo association fines for any auto repair work they catch you doing so I had to have the car towed to the shop mentioned above. Cost to replace the harness was around $125. On the positive side, I have not seen the battery indicator flash ever since. I strongly believe that the problem was the harness at the battery end. The after market harness's copper was more than twice as thick as the original.

Thanks for the report. By "same problem" do you mean you had the red charging light on but the voltage was reading around 14V? That's a brief summary of my problem. More completely, I do not seem to have any problems anywhere that I can find (voltage, current, battery state, etc.) other than that red light coming on.

And to be clear - the only thing that makes this a tough problem for me is that the voltage is good while the light is on. If the light were on while voltage was low, that is a completely different problem to solve - and much more common and normal - failed brushes, slip rings, diode(s), VR, belt/tensioner, etc. could all cause that.

I'll take another look at the positive terminal wiring as you suggest, but I don't expect to find anything - things I can see really are all looking good.

Since my last post, I did go ahead and see if wiggling wires could recreate the false warning light. Specifically, with the key OFF, I disconnected the alternator 4P connector, then turned the key to ON (without starting). This is the service manual test. The light correctly went out (and before doing this, the intermittent problem had gone away). So then with my daughter carefully watching for that light to come on even for a flicker, I wiggled all the wires I could find, starting at the alternator and working back toward the battery + elsewhere. No change. Not conclusive, I know.

And the car is still running just like it was has since this problem started - perfectly except that the red charging light comes on intermittently. And the digital voltage gauge always indicates 13-14V, even when this light comes on.

Mr.Tugtug · Jun 19, 2019

In this case just find a way two locate the signal that triggering the battery indicator, if schematic on hand, test point from the source. I would suggest isolating the switch signal would pin point the problem. just make sure to know what signal is being switched

oldskewel · Jun 27, 2019

Updating ...

since around the time of the wiggling-the-wires test a week or two ago, the intermittent problem has gone into hiding. Neither the charging light nor the CHANGE bATT message have been seen since then.

To me, this suggests a flaky wiring issue somewhere, but I'll continue to be cautious about drawing a conclusion. Will of course post any significant updates or solutions.

It would be much easier to debug the wiring if the problem were not intermittent. At this point, with the permanently installed voltage gauge, we can believe that even if we know the charging light may be wrong.

I ordered a Foxwell NT520Pro (as I also have for some other cars) with the Honda software, and will see if that can tell me anything interesting once it arrives.

oldskewel · Aug 11, 2019

Updating again, about 3 months and 4k+ miles after this started ...

Same story. Problem has become more rare, appearing only once in the last month. Charging light comes on while the voltage gauge indicates no problem at all.

I did receive the Foxwell, and think I plugged it in and searched for any relevant tests, but still did not find any problems. Still waiting for the problem to either fix or reveal itself.

On the above suggestions about a slipping belt, or similar issues ... yes, a slipping belt could cause the alternator to not charge, which would turn the light on, but it would also cause a voltage drop. So I'm not expecting something like that to be the root cause - it must be a problem that causes the light to come on while the charging is actually OK. And if that problem ends up being something whose only effect is a false alarm, we can live with that, having the voltage gauge in place so we can confidently ignore the warning light.

Battery light on, alternator OK

oldskewel

HondaUpNorth

bobodaclown

oldskewel

oldskewel

egads

John Clark

oldskewel

John Clark

kselfri

bobodaclown

oldskewel

nolanelana

oldskewel

escape

alphonse

oldskewel

Mr.Tugtug

oldskewel

oldskewel

Top Contributors this Month

Recommended Communities