Voltage mystery - regulator problem?

[03cranec]

I have what I thought was a regulator issue with my 1980 xs400G, but I just replaced the regulator and I'm getting the same behavior, so I'm wondering if it could be something else. Here's the story:
- I've got a topped off battery that reads 12.5V across the terminals
- When I put the key in, the battery drops to 12.3V - this seems normal
- Once I start the engine, the voltmeter is unable to get a reading anymore. The display just freezes on whatever the last number shown was. I tried it on the 20V and 200V ranges and got the same result, so it's not maxing out the range.
- I the resistance readings on the alternator are all good. I also checked the AC current coming over the 3 white wires from the alternator and I'm getting 40-100V+ when I rev the engine, so this seems all good.
- Resistances on the rectifier all check out so I don't that's the issue.
- By process of elimination, this left the regulator, so as I said, I just replaced it - but nothing changed. I got the regulator used off ebay, so there is a chance it is bad as well, but it supposedly came from a working bike.

So here're my questions:
- What might make the voltmeter put out non-sense readings like that only when I start the engine?
- Is there any way to confirm that this is / is not a regulator issue?
- Do I potentially have no issue at all? Or just a bad voltmeter?

A little back story for more context: I recently replaced the whole alternator/stator unit because I was having charging issues. When I got in there to replace it I found the wire frayed almost apart on the old stator, so I know this was causing the original issue. I only discovered this new odd voltmeter behavior when I went to check the installation of the new stator. Right after the installation the voltmeter (a different voltmeter than the one I'm using now) did show 15-16V across the battery terminals with the engine running, so that's when I ordered the new regulator. I'm now unable to reproduce that behavior though. I may not have any issue here, but I'm afraid to ride the bike as is without confirming that my battery is not under or over charging.

Thanks everyone! This forum has been super helpful on a couple easier issues I've had, so I'm hoping someone can help me out with this trickier one.

-Chris

 

[bcware]

A fully charged battery should sit at around 12.8 volts. 12.5 volts is about ~75% charged. 12.3 volts is about 50%.

The multimeter could be broken; I have busted a few cheap ones in my day. Try with a second meter. You can get a real cheap one from wal mart for about $5 I believe.

Also, I'm at a bit of a loss here. What's your issue aside from not getting a voltage reading? Is the bike doing something wrong? Or is your meter just not getting a reading?

 

[03cranec]

- Yeah, the battery was at 12.8-13V, but then I started and stopped the engine a bunch of times trying to figure this out, so I think that's when it dropped to the low 12s.

- I don't think the voltmeter is broken because it's reading fine when the engine is off, but I will try another one. I will need to drive the bike to get to the other one though, which I'm afraid to do, because:

- I was having the issue that the bike died after riding it for an hour. That's why I replaced the stator. All the readings coming off the stator look good now, so I don't think I have a problem there now. But I haven't run the bike more than 5 minutes since, because I haven't been able to get the proper voltage reading on the battery to confirm that it's charging correctly. Except one time, with the other voltmeter, the one I'd have to drive to, which said I was getting like 16V to the battery. I don't want to ride on it now because if that's right, then I think I'll over charge and damage the battery. But now, having ordered a new regulator, I'm having all this trouble getting a new voltage reading... sorry if this is rambling, am I making sense?

 

[03cranec]

I guess I side question would be, how concerned should I be about running the bike if the regulator is actually bad? How quickly will badly regulated / unregulated voltage damage my battery?

 

[bcware]

I can't explain why you can't get a reading when the bike is running; because I can't I have to lean towards the meter being the culprit. The display should never 'freeze.' Before you do anything else you need to make sure that meter isn't the problem. I would not drive the bike until you've tested the voltage at the battery; it's too risky and you could get stranded.

 

[03cranec]

Got it, thanks, makes sense. I'll try to pick up a new voltmeter today and report back. Thanks for the quick answers!

 

[03cranec]

Ok, I picked up a brand new voltmeter, and I am getting readings now, although they are still confusing. The voltmeter is digital and has 3 ranges that I'm using: 20V, 200V, and 500V.

With the engine off the battery reads ~13V on all three ranges. With the engine on, the battery reads different voltages using each range:
-20V range jumps around from 5-15 volts
-200V range holds pretty steady around 18V
-500V range jumps around from 50-100V

I'm not sure what to make of this, or which measurement to believe.. Any idea what could cause this behavior? In any case, I'm inclined to believe I have a bad regulator. Any thoughts?

 

[03cranec]

By the way, I checked I get the same behavior with both regulators, so maybe it's not the regulator? Or both have gone bad in the same way?

 

[bcware]

Are you getting good contact from the terminals? Do you have another vehicle you can test the meter on? A car or another working bike? You could try those just to make sure your readings are good.

You want the 20V setting.

If you're measuring across the battery terminals you should never get a reading *lower* than the potential that exists in the battery. A faulty regulator can make it read *higher* but not lower.

Again, I suspect it is not the regulator that is the problem here.

 

[16VGTIDave]

Double check the rectifier, you may be getting AC through to the battery. This would explain the strange readings.

 

[03cranec]

Dave, you may have nailed it. I went back and checked the rectifier and now I'm getting overload resistance measurements on all 6 diodes in both directions. The multimeter I'm using also has a diode test, which is showing OL on all in one direction and 600V+ voltage drop across each diode in the other direction. The manual says it uses 1mA current for this test, so a little V=IR math seems to indicate that although a little current is getting through, the resistance is massive. I guess I blew out my rectifier diodes, yes? I have no idea how this would explain the strange voltage readings described before though - any thoughts there? In any case, I'm guessing replacing the rectifier should be my next step, yes?

Thanks guys for your help here. Really appreciate it.

 

[bcware]

It sounds like your diodes are testing out fine. OL on one side and 600 on the other means the diode is good, not bad.

 

[03cranec]

Ok, damn, you're totally right. My multimeter's diode test output is in mV not volts, so I was actually seeing a ~.6V drop, which is the correct reading. So back to square one...

To answer your previous questions, yes, I believe I am getting good contact on the terminals - it's a new battery, so they are clean. Perhaps the vibration of the engine is disrupting the connection once I turn the engine on, but that doesn't seem right. I'll try testing the multimeter on another vehicle when I have a chance, but it's brand new, and seems to be working well in all other applications, so I don't think it's faulty.

One thing I'm noticing is that before I even touch the terminals, I am getting some voltage reading. In order to reach the battery, the multimeter wires are passing within a few inches of all the other electronics of the bike, so maybe something is putting off a magnetic field or some other kind of interference that is messing with the measurements? Not sure if that makes any sense, just a thought.

I'm really at a loss now...

 

[bcware]

Don't despair; like I said, it may just be your meter or the way you are taking readings that is causing the problem. Thats why I want you to test the meter on something you know to be working to rule it out.

Do you have probes on the meter or alligator clips? How are you holding the probes to the battery and revving the engine at the same time? The probe ends can actually be quite difficult to hold against a surface for a good reading.

 

[03cranec]

I have probes, not clips. I measured first by holding against the battery terminals. But then I was also able to measure at the rectifier connection, by shoving the probes up the back of the connector in the proper holes. That freed my hands to rev the engine.

 

[bcware]

That may be the problem; those probes can be troublesome and often need pressure applied to keep them in contact. Do you have a second pair of hands that can rev the bike while you hold the probes against the battery terminals? Failing that get some alligator clips. Clip one end to the probe and the other to the battery terminal. If the probes are moving at all the reading will often change.

 

[03cranec]

Ok, thanks, I'll try that too. Shouldn't I be able to get normal readings at engine idle though? I thought I may need to rev to get up to the full ~14.7V for charging, but I should be seeing something between 12.8-14.7V at idle, no?

 

[bcware]

It should be about normal for the battery at idle, yes; like 12.8 or so. By around 4,000 rpms you should be seeing a significant boost in the voltage to 13 or 14 volts. The repair manual will have the exact values, but that's basically how it goes.

 

[03cranec]

Just to close this out... I eventually got the bike to a shop where they had a much higher grade multimeter. All the readings were right where they should be using that DMM, and the bike has been running fine for about a week now, so I think I'm all good. I guess it was just something going wrong in the cheap DMM.

 

[bcware]

Thanks for the follow-up. It's good to know she's running right as rain now!