Curvy Leakage and Stator Tests
 

Reg-rec was overcharging and I've replaced it with an aftermarket unit.

I've been carrying out the usual tests and thought I'd do a leakage test. On that front there is a dispute between Messrs Suzuki and Haynes

For the leakage test the workshop manual specifies no more than 1mA whereas the Haynes says the limit is 0.1mA. I'm getting 0.2mA.

I definitely got 0.1mA the last time I did this test over a year ago. I've recently changed the headlight and indicators and fitted a voltmeter (hence the overcharging discovery). Disconnected all that and still getting 0.2mA.

Not a problem according to Suzuki but something I should be investigating according to Haynes. Who should I believe?


I've also checked the continuity on the stator. Both manuals agree on 0.2 to 0.55 ohms. I'm getting 0.7 ohms across all combinations of the three yellow output wires and no reading to ground. They're out of spec but not by much and, as all three are the same, is it worth worrying about?

Re: Curvy Leakage and Stator Tests
 

Speaking in general terms, 1mA is a low and acceptable value for quiescent current ("leakage"). Put into context, it'll take 1000hrs to drop 1Ah from the battery, that's 6weeks give or take.

Natural charge loss from a disconnected battery can be anything up to around 0.5% per day, so possibly in the region of 20% loss or 2Ah from a 10Ah battery in the same time.

If you're measuring way less than 1mA then it's fine. Anything down around 0.1mA is likely to be of the order of the accuracy of the measuring device, the difference between 0.1mA and 0.2mA is negligible.

I wouldn't worry too much about the winding resistance measurements, if they're all about the same and there's no path to earth, then check the open circuit volts between phases with the engine running (i.e. disconnect the alternator lead and measure between each pair of the 3 wires). If they all read about the same, of the order of 70V(ac) then all is well. Open circuit volts can vary quite a lot, 50-90V(ac) is probably a sensible range to expect but they should all be more or less the same between each pair.

Re: Curvy Leakage and Stator Tests
 

I do like the org, especially when I see knowledgeable and considered questions and answers like the above.
A very useful exchange there. Thanks chaps!

Re: Curvy Leakage and Stator Tests
 

You need to be using a pretty good multimeter with a very low ohm range and self calibration feature to make these measurements. Something from Maplin just wont do it, so don't worry about it unless you have the right gear

Re: Curvy Leakage and Stator Tests
 

Thanks embee. I unplugged the output wires on the new reg rec and the reading dropped to 0.1mA so it looks like that's the reason for the change. It's definitely the reg rec itself that's leading to the slightly higher reading, not the bikes wiring.

As you say, accuracy of the meter is probably relevant - my multimeter goes to 1 decimal point in that range so the actual difference could be really tiny.

I'll do the load tests on the stator and reg rec output next time I get a chance and see how that goes.

Run-of-the-mill DIY item so it sounds like close is good enough. :)