Boxerworks.com

BTW Doug, if you ever do rebuild one of these starters, I forgot to mention, generators, DC motors and the like, you under-cut the mica on the commutator after running them through a lathe. On starters, while it won't "break" them, it is not needed and not recommended for it only makes them twice as loud and leads to faster brush and commutator wear as well as additional arcing.

Ken in Oklahoma wrote:There's a lot of good information in this topic, and I've saved some of it to my computer for future reference.

This is a bit off topic to this topic, but when dealing with starter issues commonly the battery condition comes into question. For my money the best way to resolve questions about the battery condition is to load test it. For that Harbor Freight sells two resistive load testers. I bought the smaller one on sale. And you can find 20% off coupons in many popular magazines like popular mechanics and sometimes in the local Sunday paper.

The tester is brutally simple (or so I believe) with a volt/amp meter and a press-to-test switch to apply the load. You apply the positive and negative tester leads to the battery and note the voltmeter. The voltmeter also has color scales indicating cold cranking amps and color scales indicating acceptable voltage drop. You press the button for 10 seconds and see what you've got.

I'm not sure I believe the readings I'm getting. For example I question whether a new charged lead acid motorcycle battery will really deliver up to 600CCA. But even if the reading is off, it works well on a comparative basis. This simple test allows me to say that this battery is pretty damn good or not worth a dam.

Just a thought.


Ken

'

Hey Ken, I think you can find those same cheapies with ammeters built in for running the starter current draw test as well... Obviously, the load shunt is removed for that.

dwire wrote:Tricky, tricky... Good to use lots of places, but help me out...

I like it and use it elsewhere but, how does it not tell him that the contacts are worn in the solenoid, the ground to the same is no good??? Not sure I get it. The only way that meter reads any DC is with a complete circuit, the only way it gets that is through the solenoid - unless of course I have not had enough coffee yet....

To check the solinoid you do the same thing. Put the + lead on the battery cable terminal of the solinoid and the negative lead on the starter terminal of the solinoid and hit the button.

How it works:
The meter reads 0v when there is no load beacuse there is no current flowing through the multimeter. When load is applied, and there is some resistance between the master side of the circuit and the slave, the slave will have a lower voltage. The multimeter sees this redution of voltage as a ground and reads a voltage value, which is the differance between the voltage of the master and slave, or voltage drop.

BTW, I just checked and a cable should have 0 drop, and every connection should be .1 of a volt. So in an ideal world there should be only .2v drop between the battery and starter, and .3v across the solinoid.

dwire wrote:Tricky, tricky... Good to use lots of places, but help me out...

I like it and use it elsewhere but, how does it not tell him that the contacts are worn in the solenoid, the ground to the same is no good??? Not sure I get it. The only way that meter reads any DC is with a complete circuit, the only way it gets that is through the solenoid - unless of course I have not had enough coffee yet...

I am quite interested, but at a loss why when you could just measure the cables' resistance and as well when you disconnect it/them, you can not only work the ends around doing so, but also be ready to move on to taking a current measurement of the starter draw if you need to move forward...

Please let me treat two posts in this one post. I don't know how to put quotes from two posts into one post. But, I can copy and paste.

Doug, your suggestion of checking the DC resistance is fraught with possible error. The crappy crimp and corrosion sometimes will chemically make a crude diode. The result is that the cable is no longer a linear conductor. The resistance will change according to the level of current. The low level of voltage in the mulitmeter may show a different number of Ohms compared to the result of using a higher voltage, which means a higher current. Basically, you run a risk of thinking that you have measured something, but didn't get a result that can be trusted.

Your suggestion of measuring the actual current flow is a very good one and assures a more accurate reading. That will allow a better diagnosis. To use a voltmeter to measure the voltage drop over the length of the conductor is just as accurate and easier than disconnecting to get the ammeter in line.

Keep in mind that an ammeter is just a volt meter with a shunt across it. By attempting to measure current, you are only doing the "measure the voltage drop" of the cable. Both conditions, of course, must be done under the exact load of operating conditions. I think that we tend to forget how an ammeter works. Does this make sense?

Depending on the chemistry, the non linear conductor can amount to an old fashioned spark gap transmitter, or a circuit that can absorb different radio signals and mix them in a way to produce a new signal on some frequency. This new signal can be very strong and transmit for many miles. It is known as spurious radiation, or a spur.

You ask, "So what?" in our older bikes we don't care about a few micro watts of stray RF energy. However, as we have now gone to microprocessors that live on very low level signals, we stand a chance to interfere with the computer on board. They may all be protected from stray RF by this date, but I don't know that.

Garnet said, "You stil haven't determined that you actally need new cables. Other than crappy crimp connections, battery cables don't wear."

It is true that battery cables don't wear, but a crappy crimp is not the only devil to find. As air creeps up the insulation of the cable, the moisture does corrode. This chemical process will seriously reduce the cross section of copper. Basically, over time the wire becomes smaller and smaller. That section within the last inch or so becomes the point of higher resistance.

Isn't it fun to pick even deeper and deeper into things? My Aspberger's is showing.

I am aware of these things... Much will educate others and as well remind me of a thing or two too.

You still have yet to answer how the volt meter measures the (potential) difference that you are speaking of that would be indicated without already taking for granted that the solenoid is making proper contact to ground. To date I have yet to posses a (simple) DC meter that can read DC from the circuit under test without a complete circuit. There are complex potential meters that do all manner of things all manner of other ways and cost more than 5 of these bikes, but I am still lost on the simplest of things here. Please elaborate and enlighten me on the error in my ways with regard to that. We could talk about how radar affects the reading of a meter, cable or what have you, but it does not answer what I am eluded by here and likely does more to confuse others that are very inclined, intelligent and capable but perhaps have not quite the book-smarts some do.

Talking tank circuits is fun, but telling someone about it and Tesla when they have no spark likely doesn't help them understand what is going on with their bike at the time... LOL ---> What me? Do I have a tendency to do such things>>> No, never...

I am sure it is easily explained in a simple sentence and I am somehow overlooking the obvious...

Not sure how to explain it better than I already have , but it does work with my $8 multmeter.

Sure does, it assumes the solenoid's ground is A-OK...

If this were AC we would be talking different stories, take your meter to L1 and L2, you'll measure the potential (now-a-days with digital DMM's likely the RMS...) there and see 240 VAC. I am still lost how DC potential can be shown without an assumed ground, if this were true, and the test is valid - meaning not making the assumption that the solenoid ground (and contacts that make the circuit for it inside...) than, the test should work the same with the cable unhooked from the solenoid terminal and we all know it won't for two reasons.

1) Bye, bye starter switch actuating solenoid and influencing the current flow through both cable and meter.
2) No connection to ground - nothing happens at all...

Furthermore, if you unhook the cable, well there is your series connection - do it on the other end of the cable at the big solenoid terminal. Do you all see where this is going? There are now a combination of tests that can be run to check the cable and the starter each REALLY well if that is where you unhook at, but I digress... Duane had the simple answer. Inspect, if it looks bad it likely is. If it doesn't it very well may still be - test it; if it's old and the OP was going to spend big money on designer cables, well Heck, just replace it anyway! LOL...

This is a valid test. It's really what to do, but not how. You use a known load and a known system to test the cable OFF OF THE BIKE. They make such things... I just think making the assumption one's solenoid is not involved by running anything in parallel is not the way to go, but what do I know??? You don't want to invest in a set-up to test cables. Next we could start discussing Wheastone Bridges and other cool stuff - again lots of learning going on, little BMW fixing happening...

Again, anyone, anyone? Help me out if I am losing my mind here... I don't mind a bit being wrong; actually, since I don't have a bad system at this moment, I would LOVE TO BE WRONG as I would REALLY LEARN SOMETHING TODAY!

dwire wrote:I am aware of these things... Much will educate others and as well remind me of a thing or two too.

You still have yet to answer how the volt meter measures the (potential) difference that you are speaking of that would be indicated without already taking for granted that the solenoid is making proper contact to ground. To date I have yet to posses a (simple) DC meter that can read DC from the circuit under test without a complete circuit. There are complex potential meters that do all manner of things all manner of other ways and cost more than 5 of these bikes, but I am still lost on the simplest of things here. Please elaborate and enlighten me on the error in my ways with regard to that. We could talk about how radar affects the reading of a meter, cable or what have you, but it does not answer what I am eluded by here and likely does more to confuse others that are very inclined, intelligent and capable but perhaps have not quite the book-smarts some do.

Talking tank circuits is fun, but telling someone about it and Tesla when they have no spark likely doesn't help them understand what is going on with their bike at the time... LOL ---> What me? Do I have a tendency to do such things>>> No, never...

I am sure it is easily explained in a simple sentence and I am somehow overlooking the obvious...

Doug, are you asking Garnet, or me?

Don't care friend, I don't think I even saw your post when I wrote that.

Does not matter, would love to understand (and be wrong today...) and it matters not who breaks it to me. Pride is not a thing I have in matters I simply cannot understand...

Hmm... So the starter, cable, etc. then is the "known load???" Even though it is unknown? I don't question the viability of the test, I just do not see how the test does what one might want... See my other likely flawed post...

Perhaps I should sketch this out and see if I am losing it by plugging in the values...

Yes, Duane, after re-reading your post (that I had not seen...) I understand, but with an ammeter the shunt in the meter is known, what exactly is "known" with the voltage drop test, we've lost our control have we not? Would not the starter, cable, connections etc, now be the "known's" for the measurement, yet they are exactly what one would test because they are unknown -see why I got confused? I've seen this done and done it myself for other sorts of things, can't see how it works well here, but I'll default to those that have used this method on THIS system time and time again and received a valid answer to the question - "Where is the weak link?"

Thanks for your time and patience...

dwire wrote:
Again, anyone, anyone? Help me out if I am losing my mind here... I don't mind a bit being wrong; actually, since I don't have a bad system at this moment, I would LOVE TO BE WRONG as I would REALLY LEARN SOMETHING TODAY!

To test weather the solinoid contacts are OK:

Attach the + lead of the multimeter to the battery cable stud of the solinoid (not the battery cable). Attach the - lead of the multimeter to the other big post of the solinoid where the heavy power wire goes to the starter motor (both are 8mm studs), again to the post only. At no time disconnect anything.

Hit the starter button and crank the engine and observe the meter. You should see a reading of about .2 Volts. The solinoid disc connects the two heavy studs and should have a drop of .1v at each conatct. If you have a much higher reading on the meter, your have bad conatcts inside the solinoid.

It maters not if the volt meter is gorunded.