Execut1ve

hello all, I am having a strange electrical problem on my 2000 ex (d16y8)

I notice it mainly at night with my headlights on, but I recently verified that it occurs with the headlights off as well, it's just easier to see with them on in the dark

the problem is this: from time to time I will notice a periodic dimming in the intensity of the headlights. They will get dim for maybe 1/2 second then return to normal brightness. This might persist for 5-10 minutes. While the problem persists the dim spells sometimes get dimmer and dimmer, and occasionally the SRS light illuminates. All the while the engine seems to keep running just fine with no loss of power. When the lights return to normal intensity the SRS light goes out

I have had the alternator and battery tested and both tested good. This leaves (as far as I can tell) a corroded ground somewhere as the likely cause. does that seem reasonable? can anyone recommend any good grounds to check? maybe some that are subject to the effects of some good old ohio road salt?

I welcome any other advice or suggestions also! thanks!

ezone

Just because the alt passed a test, does not necessarily mean it is good. All it proves is that it was working at the time the test was done.
Similarly, just because it stops working does not necessarily make it bad. There are many reasons it may quit charging.





When the lights dim and the SRS light comes on, the alternator is not charging---for some reason.

How many wires go to the alternator? A problem with any of those wires (or whatever is connected to the other end of those wires) could stop it from charging properly, in addition to any possible ground issues.

Execut1ve

the alternator is connected to 1 wire that is secured by a bolt and one wiring connector that seems to have 4 wires in it.

maybe I wasn't clear in my first post... it's not like the lights dim and the SRS light comes on and stays for 5 min then goes back to normal; the intensity of the lights oscillates between normal and dim with a frequency of about 1Hz, so about 1 normal/dim cycle per second. during that time the SRS light may illuminate but it does not always; when the oscillation ceases the SRS light turns off, if it was on

I'm not super familiar with the electrical system on these vehicles but it seems to me that either a bad ground or some malfunction with how the vehicle maintains the voltage at 12V

and just to be clear, by testing the alternator/battery I mean I removed them from the car and took them to a parts store to be bench tested, instead of using a multimeter to do the tests in the repair manual

ezone

No, you weren't. New info is new.

I was sticking with the "5 minutes of no charge" at a time, that meant to me the system quits charging for up to 5 minutes at a time, then works again.
I read this a dozen times, apparently I'm not awake yet....
Can you tell me more about the 1hz cycles? At what time does it do this?
All the time, only when the voltage dropped, or only when it is charging normally, or what?

AND do the lights flash brighter or dimmer in those one second cycles?

I can't really picture anything on the car that would draw so much power except the starter or a short in a big positive cable...





Different approach now. Well not really much different.
I'm still thinking the charging system shuts off and lets the battery run down to the point the SRS doesn't like it.

IS the alternator on the car already an aftermarket reman, like from DangerZone or the like?
What code is in SRS now? Any related to low voltage?(Just as a confirmation, no big deal because that would be a normal reaction to low system voltage)

Here's a C&P from a Honda Service News:
On all Honda models with SRS, a battery that%u2019s not
fully charged can cause an SRS DTC; and this can
be any DTC from 5-1 through 8-6. In Honda
service manuals, these DTCs are listed as an
%u201CInternal failure of SRS unit,%u201D but low battery
voltage can also cause them to set. So, before you
troubleshoot the SRS for any of these DTCs,
follow these guidelines:
1. Check the state of charge and the condition of
the battery. Recharge or replace the battery as
needed.
2. Check the charging system, and make any
needed repairs
There is a little more, but that's the general idea.


Have you put a voltmeter on it to see just what the voltages are doing while this happens? (system voltage)
I'm gonna GUESS that the 1hz oscillation will probably turn out to be caused by a flaky (or cheapo) voltage regulator inside the alternator. I have seen it from a local (low price) rebuilder that uses the absolute cheapest parts to rebuild his rebuilt alts. And of course the cheapo parts stores.

You need to be doing the testing IN the car, and WHILE the problem is acting up (all checks that have to do with monitoring voltages while in operation). If the problem is not acting up when you test it, everything you test will probably look ok.

The alternator generates power, the voltage regulator controls the voltage from inside it. The PCM turns the charging system on high or low as needed to conserve fuel (ELD system).

I check grounds using the "voltage drop" method first.
If you think a bad ground is the culprit, a fast check is to clamp a jumper cable on to the aluminum alternator housing and to the negative battery cable (eliminates all grounds between battery/frame/engine block/alternator) see if it works right again. (I did find some info about painted brackets causing a no-charge condition)

Then I might be figuring out what each of the 4 small alternator wires are supposed to do, then monitor the voltages on the important ones to see if there is something OTHER than the alt itself--an outside influence-- that is making the system turn charging on and off that rapidly. I bet it will be kinda hard for you to tell much from that though, all voltages will fluctuate as the system shuts down or oscillates.




Something else that happens on these cars is when a timing belt is replaced, the engine mount is removed from the engine and engine is allowed to drop. The wire harness between the drivers strut tower and the engine gets a lot of weight hanging from it, it can damage wires in that area!


Any other symptoms elsewhere? Engine codes for O2 sensors, VSS, speedo acting wrong, or anything else?


Did any of that help?

Execut1ve

when I was talking about the 1Hz cycles I was referring to how quickly the lights seem to oscillate between normal brightness and dim. let me try to explain what is going on again...

at times, for maybe 5-10 minutes total at a time, the lights will dim in intensity (but still remain lit) for maybe 1/2 second. This includes headlights, dashboard illumination, check engine light, everything. Then the lights return to normal intensity for maybe 1/2 second. This cycle of dimness and normal repeats for 5-10 minutes total. Total period of one dim/normal cycle is 1 second, so frequency is 1Hz.

I have not put a voltmeter on the 12V and driven around until the problem appeared yet, it's on my list though.

I'm not sure off the top of my head if the alt is original or reman

other than this the car drives fine; I do have a CEL due to an O2 sensor, but that's a recent problem and this light dimming issue has been going on for much longer (see my other thread regarding the O2 sensor). I have pulled the codes, the only ones are for the O2 sensor, nothing related to the SRS.

your suggestion of a flaky voltage regulator in the alternator makes the most sense to me, given the symptoms

hope that clarifies what is going on a bit

ezone

Ok that explains the cycling a little better.

Now I'm wondering---when the lights dim, just how low the voltage drops.
If the battery voltage at rest is (example) 12.4 volts and it's charging at 13.8 when it works right,......

you are driving and the lights dim, does the system voltage drop far below the battery resting voltage?

Have you tried wiggling wire harnesses while it is running with all the lights are turned on, to see if you can make anything wrong happen?

If you get it to act up and stop the car, leave it running, Will it continue to act up? Wiggle wires at this time?

Just a thought.

My hunch is: I'm thinking that if you cannot cause the problem to act up by wiggling wires, and you cannot get the symptom to stop by wiggling the wires, that the cause of the problem is going to be in the alternator itself.
Oh I didn't know that was you.
No codes huh?


Another angle:
The SRS light will come on when the system voltage goes TOO HIGH. About 16 volts will turn the light on and leave no codes.
Maybe the alt is OVERCHARGING as you drive, and the dimming you see is when it goes back to normal or proper charging voltage??

Execut1ve

I don't have an actual voltmeter, just a small multimeter so I'll have to jimmy rig some semi-permanent connections... any ideas for where I should connect this to measure the 12V?

ezone

If it has a basic voltmeter function, that's all you need.

Any place that has system voltage (B+) and ground, and they don't even have to be near each other either.

Battery posts would be ideal, but the lighter socket might work just as well since it is part of the 12v system when the key is on, and can be far simpler and non-permanent if you have a 12v accessory plug you can chop the wires off of.

Execut1ve

got the voltmeter all rigged up to the 12V accessory socket. After driving around for a couple days with it installed, the problem has not reappeared. However I have noticed that it is not unusual for the voltage to be as high as 15V. Also when I turn on the turn signal the voltage drops by as much as .5-1V in time with the turn signal light. will post anything else I discover

ezone

Seems a little high, but not far off. 13.8-14.6v is typical specs.... I wonder if it really is that high, or if it is a meter discrepancy. I know I have had flaky meters, and still do.
That's normal.

Execut1ve

I will be driving to Boston and back next week, so I anticipate the problem will arise again. Hopefully I'll be able to get some voltages at that time. will keep you posted

Execut1ve

I threw the meter into the truck and measured the 12V... came out at almost 15V as well, so that's a point in favor of a flaky meter

ezone

LOL

Execut1ve

should still be able to use the meter to get useful information though, as I'm more interested in a relative voltage drop than the absolute voltage