A strange electrical problem.

[georgesgiralt]

Hello !
I switched for a new battery (flooded, Chinese, cheap) about one year and a half ago.
Since a couple of weeks, I have a problem :
The bike starts fine in the morning because I put the battery on the Optimate at night. Then at noon when I want to get the bike for the lunch, she sometimes fail to start because the starter lacks energy. Starting with the kick starter is fine though.
On the evening, when I want to go home from work, same at at midday.
At first I thought that the alternator was not charging enough. So I checked the rotor, regulator, brushes, and diode board. I ride with the eye on the voltmeter ;-) Everything is fine....
So yesterday I resigned myself to buy a new battery. I ride to the big shop which usually provides me with them at cheap. (but not last time...)
The tech guy looked at the bike with a voltmeter, an ammeter, and a battery testing device.
He said to me that the charging system was fine. That the battery was fine because it can give 215 A peak current, and that I has 0.6A current with the contact OFF... So for him it was normal that the battery dries out... We unplugged the clock but this changed nothing. He refused to sell me a new one...
Yesterday afternoon and this morning, I've checked everything electrical on the bike. I do not find any leak. Everything plugged on and contact off my measuring give between 0.2 and 0.05 mA .. So the clock needs power but not so much.
I removed the contact switch to give it a clean and, of course, this changed nothing... So I'm seeking for your advice here ...Boy, this is frustrating and time consuming !
As always, your collective wisdom is vastly useful !

[AlfromNH]

The bike starts fine in the morning because you leave it on the Optimate overnight. I'm assuming an Optimate is a charger, if you don't put it on the charger overnight, will it start in the morning? If not, then the battery either is being drained or won't hold a charge on its own.

Starting with the battery fully charged, try disconnecting the negative battery terminal overnight and see if it will start in the morning.

[Barry]

I agree. Try disconnecting the battery and see if it starts in the morning.  

That 0.6A drain that the battery guy measured would certainly flatten the battery but it must be an intermittent fault if you are now only measuring 0.2 to 0.05 milliamps.  

[minitoo]

Might check your cables.  They corrode internally and even though they look good they aren't.  This cured my bike with the same issue.

motoelectric.com fixed me up.

[nhmaf]

If you have a cable attached to the battery with a quick connect attachment for plugging in the charger I'd take a careful look at it - you might have a crack or worn spot on the insulation that is intermittently contacting the frame or other metal part somewhere under the seat. 600mA is much too high of a current draw for only a clock, provided that the clock is working normally.  You don't have any other devices that are "on" while the ignition is off, such as an alarm system, do you?

[georgesgiralt]

So, I've followed part of your advices. I unplugged the battery from the negative big wire. Will check tomorrow how things went.
As I let the fused cable from the charger I may get the 600 mA draw... Will see.
I have a "new" main wiring but the headlight part had been removed to fit on another bike so before using this up I have to make a new headlight harness... (and auto parts store in my city do not carry the parts they used to years ago. May have something related to the recent cars being impossible to fix by the average home mechanics... )
I hope I won't put the bike on fire...

[Bob_Roller]

A diode on the diode board on it's way to failing can cause  ' leakage ' and discharge the battery .

I agree with removing the cables from the battery and see what result you get in the morning .

To isolate the diode board for troubleshooting, remove the red output wire from the diode board to take the diodes out of the circuit .

[tvrla]

I'd be looking for the .6A drain. And it appears to be an intermittent fault. :-(

Nothing like chasing down an intermittent electrical fault!

[georgesgiralt]

So week-end is history.
I spent hours verifying and checking things up to no avail. I've cleaned the starter relay contacts, and the ignition switch contacts. All wires look clean and insulation resistance is very very high. Charged the battery yesterday and let it sit on the bench. It has lost a little less than 10 % of it's charge in 24 hours. So maybe something is loose inside the battery producing a leak after a good shake...
Tonight the bike is in shape to run and the battery is on the Optimate tender as I've to be at work at 7 am tomorrow...
Will report if the bike continues to act stupid or if I find something... Thanks for the advices given !

[Bob_Roller]

Have you checked the specific gravity of the electrolyte in each cell  of the battery after fully charging it ??

[Barry]

It has lost a little less than 10 % of it's charge in 24 hours

It might not be as bad as you think. How did you measure the 10% ?  If you measured the voltage immediately after charging then that is not a valid method as you have to wait for the surface charge to dissapate. I would charge it, leave at least 12 hours then measure the voltage, leave another 24 hours and measure again.

It depends on the type of battery but for a wet cell O.2 volts represents 25% discharge so if we start at 12.65 Volts 12.45 Volts represents 75% charged. So 10% discharge would be indicated by a fall of less than 0.1 volts from the fully charged and then rested voltage. This would not be easy to measure accurately in terms of being sure that the surface charge had fully dissapted before the first measurement was taken.

[k_enn]

You may want to take a second and closer look at your starter.  I had similar symptoms several years ago (ok start in am after battery on charger overnight, questionable starting at mid-day or after work).  It turned out that the starter was pretty much shot - turned the engine but not strong enough to turn it enough to start.   I replaced the starter, and the problem went away.  

k_enn

[Bob_Roller]

You brought up a good point k_enn !!!! [smiley=thumbsup.gif]

A starter that is lacking lubricant, will use more power than it should .

You can disassemble the starter clean and relubricate it with a few hours of work .

Most of these starters are 30+ years old and the original lubricants are long gone .

I did this to the starter on my '81 R65, didn't see any trace of oil or grease where it should have been .

[georgesgiralt]

Hello !
The starter has been overhauled a few month back. So IMHO it is OK.
The problem, when not starting, is that the battery is low in voltage.
The alternator seems to charge properly (tension goes above 13.7V) and the battery guy found a leak of 0.6A which I can't find anymore.
So this morning I went to work with my multimeter. The battery had been on the tender all night. Starting was easy. Arriving at work, after 10 KM with the headlight, the battery voltage was at 12.46 V... So as it is a flooded type around 70% charge...
As I suspect there is still a leak, I wonder if I will be able to go to lunch...
This is beginning to become unnerving.
Do you know if the alternator can raise it's output voltage and not charge ?

[georgesgiralt]

You know what ?
She didn't start for lunch.
Voltage was at 12.53V before inserting the key into the ignition.
So I looked at the voltmeter when I turned her on. it said 12V (normally it is above that line) so in my mind I thought it was toast... Starter did not run and the colleagues had to push me ....
I'll check this out tonight.