1998 1999 2000 2001 Accord Coupe Indicator Not Working

[Jon]

Quite out of the blue my near side rear indicator packed up today. Front and repeater continued working at increased rate. Bought new bulb from local Honda main dealer. ( Have known Halfrauds bulbs to throw up canbus errors ) Installed new bulb....no joy ! Meter confirmed earth was good and on live test showed a swinging needle on the positive terminal. Carefully teased copper contacts in bulb holder and replaced bulb.....still no good !  Bench tested bulb all fine. More careful copper teasing and replaced bulb.....still no good ! Scratched head and drank tea then removed bulb holder to bench complete with bulb.....worked fine !  Reinstalled bulb holder in car ......no joy ! Have to admit I am now scratching my head with this one ! Ran out of time today so will pick up again tomorrow. All ideas most welcome.

 

[hughezee]

Very strange indeed, I've had similar issues when I've swapped light units and bulbs always seems to be a poor connection issue that trips these up especially on the brake lights, so I would look at replacing both bulbs and clean all the contacts in the holders? 

[PTR200S]

If your earth is good and the unit works on the bench ok then you need to check the loom feed supply to the holder with a meter.

Should be getting an intermittent 12v ish reading.

[Jon]

Thank you for diving in once again with this one ! Many have looked and I don't blame them for staying away !

You call it "very strange".....I have run out out descriptions. In between showers I have been at it again today. You almost stop believing in what's logical including yourself !

The only new thing I did today was pull the 7.5 amp fuse and test it, it was good. No immediate success when I refitted it, but then as if it was possessed it started working !  I just hope It continues.

It has to be something to do with damp ? ......boy has it been wet down here !

[Jon]

Hi PTR thank you too. My description was poor that's what I meant by reference to the swinging needle on my analog meter.

Till the next time !

Cheers Jon

[PTR200S]

As Stu says, strange indeed !!

Got to be a contact problem of some description, does the bulb holder itself have a separate earth on it or just earthed through the lamp holder itself ? 

[Jon]

Hi PTR

It's all plastic, there are only two wires, positive feed and earth.

I drove 12 miles this evening and all was well. This is the first time I have had an electrical snag of any kind with this car. I suspect damp somewhere so ran the ac and will do while the monsoon lasts.

I also made a note when we get a dry day to pull all the fuses and clean them.

I can only think there was enough contact to give meter readings which broke down under load ?

 

[PTR200S]

Jon, you got gremlins !   Try not to feed them after midnight........ !!!

Does seem to be *wet related*  then, do you not use the car much ? Good move running aircon and a good long run with heater wound up would probably help too... An excuse for a roadtrip I reckon !!!

[Laird_Scooby]

If memory serves correctly there should be a multiplug on the back of the light cluster - take this off and clean all the contacts in the plug and on the cluster. Also check to see if wiggling the wiring alters things any - it could easily be a broken wire that is passing enough current t operate a multimeter but not enough to operate the bulb - the difference being the multimeter (even an analog one) will only take a milliAmp or two, the bulb takes nearly 2A so a thousand times more!

This would also tie in with the damp related improvements - dampness traveling down inside the wire by capilliary action will "bridge" the break causing cluster headaches in terms of finding the fault. A test lamp is helpful here as it actually draws a reasonable amount of current and the more useful ones have an insulation piercing contact pin so the break in the wire can be traced by probing the wire at various points to find where the power is lost.

https://www.ebay.co.uk/itm/CAR-AUTO-ELECTRICAL-PROBE-6V-12V-24V-CIRCUIT-TESTER-LIGHT-LAMP-TEST-Motorcycle/153195472083

They use an 18V, 3W bulb which draws about 166mA @ 18V. In turn this means the bulbs resistance is 108 Ohms. This means @12V, it will pass 111mA but, more importantly, let's assume for a moment the "break" (assuming for a moment that's what we're dealing with) in the wire has a dry resistance of 54 Ohms.

Using Ohms Law, the break of 54 Ohms and the bulb of 108 Ohms will give a total resistance of 162mA which @ 12V will flow a total of 74mA. Again using Ohms Law, this will produce 8V across the test lamp so it will light very dimly.

Let's compare that to a 12V, 21W indicator bulb. They draw 1.75A giving a resistance of 6.9 Ohms (rounded up for convenience) which, when transposed into the equation above, gives a total resistance of 54 + 6.9 = 60.9 Ohms. That means the total amount of current in the circuit will be about 200mA (12/60) which will produce 1.4V on the indicator bulb - nowhere near enough to even think about lighting it and the other 10.6V will be dropped across the break.

Now, changing the bulb for a multimeter, let's say a cheap analogue one with a sensitivity of 10k/V - that will take just 0.0001A or 0.1mA which means that across the break of the assumed 54 Ohm resistance will only drop 5.4mV (0.0054V) - that's why multimeters (even cheap analogue ones, the digital ones are worse with sensitivities typically 10MoHms/V or higher) are less useful on a fault like this than the humble test lamp.

Good luck with it, let us know how it goes!

[Jon]

Wow Dave comprehensive as ever ! Thank you

I will have to read you thoughts again and try and digest in full. Your detailed explanations seem to support my suspicions.

What makes tracing these "gremlins",  as PTR calls them,  so maddening is that everything still looks at first glance to be in factory fresh condition ? 

[Laird_Scooby]

That can very often be the case Jon, everything still looks new or new ish, broken wires aren't alwys visually identifiable by a nasty kink in them.

My first port of call would be to clean the multiplug and cluster where they join.

Also worth looking at, does the indicator work when using the hazards? It's not unknown for hazard warning switches to develop high resistance due to lack of use, then they might be used once while dropping someone off for example and that use comes back to haunt you in the form of a lost indicator.

The one thing you will find when you work out what the fault happens to be is all the evidence was there from the start.

[Jon]

It was frosty here this morning and my indicator Gremlin returned !

Rigged up a 21 watt test bulb with micropins on the tails. Disconnected the near side rear multi plug and inserted test bulb tails direct...no dice. Substituted analog meter and got 12volts flashing as before. Traced blue/green positive feed back to the indicator stalk. Disconnected multi plug from indicator stalk and applied a 12 volt hot supply from the cigarette lighter into the blue/green wire microsocket. Panel repeater green arrow illuminated. Went to the near side rear multi plug and test bulb which had not illuminated. Substituted meter for test bulb. Steady 12volts indicated.

Suggests to me that the Gremlin is located between the indicator stalk and the rear multi plug ?

Will now drink beer and plan my next move.

[PTR200S]

12 minutes ago, Jon said:

Suggests to me that the Gremlin is located between the indicator stalk and the rear multi plug ?

Will now drink beer and plan my next move.

Have you tried a continuity test, off your meter, on the each of the supply wires from stalk to multi plug on lamp too?

Obviously , of course, you will see any "over" resistance as well....

on the other-hand drinking beer sounds a better option 😎

[Laird_Scooby]

20 minutes ago, Jon said:

Suggests to me that the Gremlin is located between the indicator stalk and the rear multi plug ?

Will now drink beer and plan my next move. 

Where are you earthing your meter on this test?

Odd though it will sound, i'm wondering if you're looking at this from the wrong direction. It could be an earth fault as well, even if the earth connection looks good.

Something you can try as a test, get a length of wire (multistrand, not single conductor) and bare about 1/2-3/4" on each end. Temporarily remove another bulb, slide one of the bared ends of the test wire in between the housing and where the bulb makes contact on the earth and then do similar on the indicator bulb.

This is to test for a broken earth track on the PCB within the cluster.

Meanwhile have a beer or three for me!

[Jon]

Thanks guys will try again tomorrow if there's a break in the monsoon.

[Jon]

No rain here this morning so got on with it. Ran a continuity test on blue/green near side rear multi plug to indicator stalk multi plug. New Hong Kong meter chirrupped and read 582 ohms. Thought I would see what the corresponding off side reading showed up...... blue/yellow over there,  ohms were negligible. Then my confidence was shattered when I got similar reading off the adjacent terminal. Wished I had a wiring diagram and was not confident that I really knew what was between the two multiplugs !

(However I did find a non factory brown wire earthed very optimistically with a ring terminal under one of the four nuts retaining the plastic o/s light cluster) More on this later.

Moved back to the n/s and the earth situation. Used a separate earthing location for the meter. Both earth terminals in the n/s rear multi plug checked out. They arrive along with the two o/s earthing wires at a brass 4 way ring terminal and a 10 mm bolt which I removed to reveal some of the best paint on the whole car !

Carved away paint, extended the non factory brown earth from the o/s and replaced bolt. Tested indicators.....solid as a rock. Gremlin evicted ! and who knows the parking censors might work now as well ?

Thanks again both ! Will have to drink beer now to celebrate.

 

 

[Laird_Scooby]

Bad earth then Jon? Obviously someone has had problems previously (non-factory brown wire) but it sounds like you've actually sorted it properly this time.

[Jon]

I hope so Dave it was not really the weather to do what had to be done. Different ballgame to summer tinkering ! I think I was lucky.

I have no idea what function the brown wire satisfies. I rather hoped it was the earth for the parking censors but they are still unserviceable.

What gets me is the painted earth worked for 20 years. My Gremlin was born in the factory !

I would love to know if accurate wiring diagrams are still available for 1998 RH drive Coupes ? 

Thanks for you help

Jon

[Laird_Scooby]

9 minutes ago, Jon said:

My Gremlin was born in the factory !

Certainly seems that way! Look what happened when it got wet though!

Very often they have a "dog-tooth" washer behind the screw or some sort of dimpling on the terminal to aid long term relibability but this can get squashed over time or even just dirty.  Then when it was damp it worked again.

Many moons ago there used to be an old dodge when buying a car. It involved blotting paper. Remove one of the plug leads from the dizzy cap, get the blotting paper wet, wrap it round the brass terminal on the HT lead and drive the car back to the seller. By the time it got there, the heat would have dried the blotting paper so would be misfiring. "Well it was ok until a minute ago, somethings obviously wrong but i like the car, i'll give you <about 1/3 what the seller is asking> and i'll sort the misfire - probably the head gasket as it came on so quick".

Deal done at a knock-down price, buyer drives off, misfiring worse than a blind sniper, gets a mile up the road, stops and removes the blotting paper.

Same idea in reverse as your earth problem.

[Jon]

Gremlin reappeared today seems my celebrations were premature ! 

Ruled out the earth since everything now soldered.....no crimps.

Blue/green definitely the culprit where it T's off to front and rear of anyone knows where that is ? Might try and locate it when it gets warmer. In the meantime have run a new wire bypassing it and tucked it under the plastic/carpet junction.

Bet it's a crimp, I hate them, how much would a bit of solder add to the cost of a loom ?

[Laird_Scooby]

27 minutes ago, Jon said:

how much would a bit of solder add to the cost of a loom ?

Pennies in reality but when you add those pennies up across the cost of building 1000 cars, quite a chunk. It's an "unseen" cost so has no benefit to the original purchaser, as such can't be charged for. At the other end of the scale, the argument now exists that it's lasted 20 years, the car was only expected to last 10-15 years, there's nothing wrong with the original method.

The bit i hate is when you get a "compound fault" as i call it. In other words, you have two things that are performing below expectations to create one fault - you fix one part of it and it all looks good then the other part comes back and bites you in the bum 2 weeks later.