electrical issue, lights dim with electrical load

The MOT check failed my 1986 XJ600 because of the tail light flashing when the indicator lights were turned on. It doesn't turn on and off, but when the indicator light is on the intensity of the tail light decrease. It turns out this problem does not only affect the tail light but the entire electrical system on the bike. The intensity of all lights will drop as more load is applied to the battery. The same thing happens with the engine running. I've tried replacing the battery and checked all ground connections I could find. I've tried disconnecting every connector, one by one, to try and isolate the problem to a specific component, to no avail. As I'm writing this I realise it's natural for a lead-acid battery's voltage to be relative to the load, but should turning the indicator lights on really cause a noticeable drop?

Should I expect the battery voltage to be ~14 V when the engine is running? I measured around 12 V yesterday but I've seen higher voltages before. Is the voltage from the rectifier relative to the engine's RPM? What's the best way to test if the alternator or rectifier is bad?

Lots of questions.. I'm thankful for any advice on what I can do to move forward.

 

Pre,

Welcome to the site! Someone will be chiming in soon with some good advice. I do know that if the charging system is only putting out 12V at 3k RPM something is wrong. Had to be able to charge that battery and run the system.

 

The charging system is related to RPM, so you should have 14 to 15 volts at 5000 rpm.

Be sure to check the simple things first. Remove and inspect the main and signal fuse contacts for corrosion - replace the fuses just to be sure.

If you have a DMM check the voltage at the fuses with the flashers working - use the battery negative for the DMM black probe. You should have battery voltage at the main fuse, and slightly less at the input of the signal, head, and ignition fuses. Checking those might reveal where the loss is occurring if it is happening on the high side. You would be looking for a drop in voltage corresponding to the flasher lights illuminating.

The earlier XJ series have a common ground for the main harness located under the tank at the ignition coils with one or two eyelet terminals - not sure about the location on the later model. If the testing on the high side is OK, perhaps try and locate that ground (if you haven't already) and be sure it is clean and secure.

 

I cleaned the ground connection you mentioned @Rooster53, started the engine and had it running at ~3000 rpm while measuring the voltage over the battery to a bit over 14 V, so I guess the alternator and rectifier is working. I turned on the flashers lights and saw that the green neutral light and tail light was still affected. However, I noticed something I haven't seen before (or maybe I just didn't pay attention). The characteristics of the issue differs depending on if the engine is running or not. With a stopped engine the lights will dim for as long as the flasher lights are lit but with the engine running they will only dim for a fraction of a second every time the flashers illuminate before returning to their original intensity, as if current is consumed by a recharging capacitor. Also, the brightness doesn't seem to drop as much, so the flashing effect on the tail light for example is much less obvious. Maybe this is as good as it gets?

Regarding the fuses, I've replaced them all with inline fuses as the original fuse box fell apart. I haven't measured the voltage drop between the main and ignition fuses since, but I remember it was a bit above 0.5 V before that, of which the ignition switch was responsible for 0.3 V. The voltage does indeed drop as the flasher lights illuminate, but I don't know by how much. If I remember correctly I could not get a good reading from my cheap multimeter, and I'm not sure if it's a symptom of a problem or just a natural behaviour of a lead-acid battery. I'd like to measure the voltage drop again though, with the engine running.

 

So it affects the headlight also?

You could do it that way, but I would think it would be easier to isolate in the worst case scenario - engine not running. This would also give a bit info on the battery / battery connections as it shouldn't drop significantly with that small of a load.

I would also suggest using the brake lights as the load since they should be similar to the flasher load but the load will remain constant, and that may help from getting odd meter readings. Ideally, there would be no loss from the battery/wiring/fuses/ignition switch, but that is not the case. However, I suspect to notice a significant difference in the intensity of the tail light you would have to have something in the order of at least a couple of volts. The approximately .5 volts you noted is probably fairly typical.

 

And just curious are you running the 21W flasher bulbs? (no location on huse, but assuming across the pond)

And, maybe try connecting and external battery just to rule out the battery, or a high amp charger.

 

Yes, it does, but it's quite hard to spot the difference in intensity on a bright day like today.

I've measured the voltage drop (with a good battery) between the main fuse and the others and found that it's around 0.5 V with all lights off and the same if I press the brake. If I turn on head and tail lights the drop increases to around 1 V. With lights on, but headlight disconnected, the drop in voltage returns to 0.5 V.

Here's a short summary of my results

ignition switch on, no lights
main | others
12.3 V | 11.9 V

lights on
main | others
11.7 V | 10.7 V

lights on + brake
main | others
11.5 V | 10.3 V

In fact, they are 23 W. I'm in Sweden, so assuming you mean the great Atlantic pond and you're in America then yes, I'm on the other side

While checking the flasher light bulb wattage I noticed some very inventive solutions to fix broken light bulb connectors. One involved a cap from a bike tube nozzle... I'lll replace the flasher housings as I get a feeling they could short any moment.

 

I checked my XJ750 Seca and got similar results, and I also included measurements for the tail connector including the ground pin. The drop on the ground side was .5 volts and would be an important measurement for troubleshooting your issue:

Main - measured at main fuse
Others - measured at input to fuse block - ignition switch output
Tail input - measured at tail light connector input
Tail ground - measured at tail light connector ground pin
DMM ground placed on negative terminal of battery

Ignition switch on, no lights other than tail light
main | others | tail input
11.9 V | 11.6 V | 11.5

Ignition switch on + brake + tail light
main | others | tail input | tail ground
11.7 V | 11.1 V | 11.0 | .5 V

Ignition switch on + headlight + brake light
main | others | tail input
11.3 V | 10.2 V | 10.2 V

I keep re-reading this and thinking there really is a problem, but maybe not. You noted it was much less obvious with the bike running, was the check done with the bike running?

Since you can run without a headlight (not an option on unmodified US imports) was the tail light flashing checked with the headlight illuminated?

Can you legally run LED bulbs in the tail light - they have internal constant current devices so I believe they could better tolerate voltage fluctuations.

 

if you are putting out high voltage suspect the Voltage regulator if low voltage think brushes and copper rings on rotor.

the charging system can be ohmed out

 

Thank you very much for taking the trouble of measuring on your own bike. Yes, it does look very similar.

I didn't take it to the MOT myself but I was told the inspector flashed the indicator lights with the engine running and that the symptoms seemed to get worse with increased rpm (according to the inspector). I don't agree with that though. Maybe the increased intensity of the lights at 14.5 V made the flashes easier to spot. I've been trying to solve this on and off for a few weeks now, cleaning connectors, replacing the fuse box etc. it may have just gradually improved without me noticing.

I can't run without a headlight, I just pulled the lamp connector to rule out that the 1 V drop was caused by anything else than the headlight.

LEDs should be fine in the tail light, but I'll try the MOT again before I go down that road

 

I just had to replace my brushes. Had about 11.4 volts at idle before and 12.5 volts with new brushes. Top voltage went from about 12.9 to 14.

 

The charging voltage seems fine now, but I remember I measured somewhere around 12 V when revving the engine some time before starting this thread and before cleaning the ground connection for the wiring harness. Maybe that connection was bad causing the charging voltage to drop and the system behave more like when running on battery.

I'm still not sure if the flashing issue is normal with the engine off..? If you put in neutral gear on your bike and start a flasher, is the brightness of your neutral light affected at all?