Quick question

[the108penguins]

thanks steve, and everyone. looks like i'll be busy for a while. I'll let you know how things progress.

[crazy steve]

I should have noted that my 'procedure' is dependent on the wiring harness being unmolested as far as the charging/ignition circuits go. If a PO has gotten in there and tapped/moved loads around or otherwise butchered the harness, then it can turn into a nightmare...

The other thing to keep in mind is a marginal or poor connection can change resistance. If it heats up, resistance (and power consumed) will go up too, and while this will generally reduce power to the load, sometimes it just adds load. It's possible you won't find a single cause for this, but this is from a bunch of little problems that add up to a big problem.

Like I said, this takes patience....

[natemoen]

That's a good and helpful elec test write up. Helpful for all!

[crazy steve]

First, your meter. It should work, the only problem I see is it's rather high .1 ohm resolution according to it's specs when reading low ohm values, particularly if it's calibration is off any. So getting an accurate reading on the stator windings will be tough. The correct reading should be .4 ohm, with a range of between .36 and .44 being acceptable. This is beyond the resolution of this meter; so if you get a reading of other than .4 ohm, you may still have a problem with the stator..

I want to expand on this a bit so everyone knows what I'm saying.

'Resolution' is the smallest value the meter can accurately read. This will vary by 'range', but by and large, when checking ohms on 'low range' most meters will only have a .1 ohm resolution. You do need to check your meter though; some cheaper meters won't achieve this (having a .5 or even 1 ohm resolution) and this can give you inaccurate readings.

In the case of the stator winding above, if you're at either end of the acceptable range, the meter may 'round' to the closest .1 ohm as that's its resolution. If calibration is off a bit, it could round the wrong way and give you a reading of .3 or .5 ohms, which according the FSM means a bad stator. To get a 'benchmark' to check your meter, pick up a 5 ohm resistor with a 1% mil-spec tolerance.

I'll note that this is a problem primarily with digital meters. The analog meters with a 'R x 1' log scale where you could zero the meter before measuring will give you a more accurate reading if the meter is of decent quality. Newer isn't always better....

[natemoen]

Good to know.

[cywelchjr]

You still haven't found the problem, so don't throw that battery just yet...

First, your meter. It should work, the only problem I see is it's rather high .1 ohm resolution according to it's specs when reading low ohm values, particularly if it's calibration is off any. So getting an accurate reading on the stator windings will be tough. The correct reading should be .4 ohm, with a range of between .36 and .44 being acceptable. This is beyond the resolution of this meter; so if you get a reading of other than .4 ohm, you may still have a problem with the stator.

<--Snip--> Removed because I agree

I have to take exception to one thing here. Many of these stator coils are out of spec now what with their age, and still just fine. I have not actually seen one that wasn't a good bit high on spec, and every single one of them works perfectly, providing very close to spec output current. Field coil is indeed a different animal, and those MUST be in spec or you will have trouble, but the stator coil can be as must as 250% high and still work at close to 90% of what it originally put out. That resistance doesn't really come much into play until your pulling the max the unit can put out for an extended period, which is another good reason to do everything you can to reduce current draw (LED bulbs, HID headlight etc). In addition, I would start as part of the troubleshooting process by removing ALL the light bulbs, headlight, tail etc and see what happens. Something is pulling too much, and it's gotta be found and fixed. BTW, the first thing I did on mine was pull the taillight bulbs and replaced them with LED units from Auto Zone. This is the one super easy thing you can change that draws ALL the time, and it's a significant reduction. Also, when the headlight is on, reving the engine should cause the brightness of the headlight to change, getting brighter with higher RPM's, if it doesn't something is either drawing too much, or the system is not really charging anyways.

As Steve said (based on what I said earlier), I drove 40 miles a day for a week while waiting for a replacement reg, and charged at night. The lights were getting a bit dim, and the last day it died a block and a half from home (we cheated, SWMBO came over with the car and we hooked up the jumper cables and charged it for 5 minutes and then I rode it the rest of the way home). It was the last day because the replacement reg had arrived in the mail that day, so I replaced the reg, put just enough charge on it so it would run and took it for a test ride. Nice bright lights and everything worked. Came home and put it on the charger overnight and that was pretty much the last time other than the couple of times it sat for a week or so that I put a charger on it (ok, one other, when I put the new battery in, and that was done out of the bike).

[DAVINCI]

I want to expand on this a bit so everyone knows what I'm saying.

'Resolution' is the smallest value the meter can accurately read. This will vary by 'range', but by and large, when checking ohms on 'low range' most meters will only have a .1 ohm resolution. You do need to check your meter though; some cheaper meters won't achieve this (having a .5 or even 1 ohm resolution) and this can give you inaccurate readings.

In the case of the stator winding above, if you're at either end of the acceptable range, the meter may 'round' to the closest .1 ohm as that's its resolution. If calibration is off a bit, it could round the wrong way and give you a reading of .3 or .5 ohms, which according the FSM means a bad stator. To get a 'benchmark' to check your meter, pick up a 5 ohm resistor with a 1% mil-spec tolerance.

I'll note that this is a problem primarily with digital meters. The analog meters with a 'R x 1' log scale where you could zero the meter before measuring will give you a more accurate reading if the meter is of decent quality. Newer isn't always better....

+1 on that, Steve, digital meters have no place on these bikes.

[cywelchjr]

+1 on that, Steve, digital meters have no place on these bikes.

I'll disagree again. For some things yes, but for most things a GOOD digital meter will work as well or better than an analog meter (being a tech I have a good one, medium priced etc, not one of those cheap radio shack ones, I think mine was like $50 or so, no point skimping on something important). I can even check for pulses with it, I just set it to ac and if I get pulses it gives an ac reading

[DAVINCI]

Well, then we agree to disagree. The digitals are just too damn slow to react. My best and most reliable meter is a VTVM, with 100,000 ohm per volt input.

[cywelchjr]

Hey, I guess you just haven't used one of the good ones, like a Fluke or similar. The good ones the pro's use, react very fast, not like the cheap ones that need 15 seconds to tell you what range they are on (mine is NOT auto range, those are crap, I have to select a range, so it doesn't have to figure out what range to use, it uses the one I tell it too).

I understand what your saying, even with it there are things that an analog meter would be better for, but for the limited uses, I'll buy a scope first (I won't get the cheap junk even in analog) as it will do everything the analog meter will and a whole lot it won't (used to have one before I got flooded out in 95, never replaced it though), including being able to do ignition system analysis if you know how (things like primary coil and wire resistance etc). If I really ran into something that I needed it for, I would find an inexpensive scope before the meter, probably a surplus military unit like I used to have. I'd still have it if I had stored it instead of having it open for use (was a sealed Navy unit, totally immersible in water if stored properly. Lost a really nice spectrum analyzer at the same time too, that was gone no matter what, it wasn't milspec.

[3Phase]

I have to ask: Where is the 14.<mumble> volts being measured?

If the voltage isn't being measured at the battery, start checking from wherever you measured 14 volts then work toward the battery until it's not 14 volts any more and you have located the problem.

Make sure the battery connections are good and the posts aren't broken. You can get a slightly higher than normal voltage reading at or near idle if the engine is running but a battery connection is bad/intermittent because the charging system isn't charging the battery, it's just running the engine and lights.

When the RPMs dropped the alternator output would eventually go below TCI voltage and without the battery to supply the current to maintain the system voltage the engine would die.

[cywelchjr]

I have to ask: Where is the 14.<mumble> volts being measured?

If the voltage isn't being measured at the battery, start checking from wherever you measured 14 volts then work toward the battery until it's not 14 volts any more and you have located the problem.

Make sure the battery connections are good and the posts aren't broken. You can get a slightly higher than normal voltage reading at or near idle if the engine is running but a battery connection is bad/intermittent because the charging system isn't charging the battery, it's just running the engine and lights.

When the RPMs dropped the alternator output would eventually go below TCI voltage and without the battery to supply the current to maintain the system voltage the engine would die.

Good point. I was assuming that it was being measured across the battery terminals like it's supposed to be, but I should not have done so.

[the108penguins]

yes it is being measured across the battery terminals

[3Phase]

The battery is good? 14 volts at the battery when the engine is idling but the battery runs down and the engine dies. I have a voltmeter on my bike and last summer when my XS did that the positive cable was loose at the battery.

The voltmeter would show over 14 volts even at idle but the engine would die at unexpected times, the battery would eventually go dead and require charging. When the bike unceremoniously dumped me on my keester in the parking lot when the engine quit as I was pulling out of the usual oil-slicked parking space in front of the 7-11 I got motivated and tracked it down.