Intermittent dead short while starting

Getting very close to having my 84 GPz750 back on the road but dealing with an electrical issue that's kicking my butt. More than not, when pressing the start button the motor will start to turn over then the battery will go into protection mode and have no voltage. All electronics will be dead including lights and gauges.

Battery is Lithium with the protection mode triggered by high current draw among other things. Had a low CCA lithium battery in when the issue started but it did not start happening immediately with this battery. Got the battery warrantied and it still kept happening. 

Replaced reg/rec with one from Ricks that is compatible with lithium batteries as I saw it overcharging the battery. Then got new starter solenoid and thought the issue was fixed as the original was corroded. Few days later, happening again. Each time it happens have to put the charger on the battery to get it out of EMS protection mode.

Bought a larger capacity lithium battery, went over my power wires from the battery down to the starter motor and found no signs of shorting or wire wear. The ground wire goes from the battery into a loom but I cannot seem to find where the ground wire splits off and grounds to either the frame or engine.

Two days ago I pulled off the starter motor to check it over and give it a good clean. Brushes were in good shape and other than it being dirty, everything seemed okay. Gave the armature a good polishing and reassembled. 

Anyone have other recommendations on what to check? Just saved a comment from @Wookie58 on how to detect a bad ground or positive line issue that I will try with my next chunk of spare time. 

Are you using a custom harness? I have not seen one where the ground goes into the harness. It should be a heavy gauge wire, same as the positive cable, and i think it should go from the battery post to the engine case, or perhaps the frame.

The harness is stock as far as I can tell, but the ground wire is electrical taped with other wires starting about 2 inches from the battery. I cannot tell where it runs to because this bundle runs under frame over the engine and I do not see the ground wire splitting off. I do know that this bundle of wires contains the rear turn and brake light wiring as it splits off at the battery and continues to the rear. I'd grab a photo but I am not at home at the moment. I can get one in the AM and post it up. 

I have found the best method of diagnosing elec faults is to start with a wiring diagram, look at "what you do know" and eliminate everything that can't be effecting the fault.

The fault only occurs when the starter button is pushed - eliminate everything before the starter button
When the starter is operated the main fuse doesn't blow - eliminate everything up to the solenoid on the switched side
The fault only occurs when the solenoid engages - eliminate input to the solenoid (main battery cable)
The battery safety circuit operates due to excessive current flow - eliminate ground issues which would "reduce" the current flow and cause the starter to run "slow" or "not at all"

By following the above you can eliminate all but the output cable from the solenoid to the starter and the starter itself. I suspect the fault is internal to the starter (possibly insulation breaking down in the armature) this can only be fully tested with a machine called a "growler" but you can remove the starter, connect an ammeter from the starter body to the positive terminal on the starter and slowly rotate the starter by hand. You may find the meter will drop to zero when a specific armature loop lines up with the brushes (this is not a conclusive test as there is no "load" on the motor during testing)

The below shows how much of the wiring diagram is left after the elimination process

What a great way to diagnose electrical issues!

Until  total peak amp draw is established via instrumentation, troubleshooting may well be shots in the dark.  If you know anyone who has an AC/DC clamp meter that can sense & peak-hold DC inrush current, that would be helpful.  Note we highlighted the DC, as most cheaper clamp meters are incapable of reading DC current, only AC.

We'd try a conventional flooded wet cell battery.  The protection circuit on LiIon batteries can be finicky.

If it truly is a dead short to ground somewhere, at least one fuse should be blowing instantly.  Of course, that won't be true for the + battery starter circuit.

There's a rule of thumb way of reading blown fuses.
A)  If a the link is just melted into a small gap with rounded ends& no additional discoloration, it's a simple over current that exceeded to ability of the link to pass current & it simply melted under slight over current.
  If the link has simply disappeared, with little bits of link metal are scattered inside of the envelope, along with an associated blackened starburst pattern, that indicates the fuse opened while under very high current (dead short) with the resulting massive spark.

Good Ridin'
slmjim & Z1BEBE
 

Good advice as usual guys ref interpreting failed fuses, I think in this case getting an inductive reading may be tricky given that when it faults it trips pretty quickly but it does point towards a faulty starter on the evidence presented.

Thanks a ton for taking the time for the write ups @Wookie58 and @slmjim+Z1BEBE. Thus far, no fuses have blown when this happens. I've been perusing the internet looking for a replacement starter motor but a new one runs $600 which seems steep. OEM PN is 21163-1053 if anyone has a recommendation, I know it was shared by the early Zr-7 and ZR750s.

One thing I should have mentioned in my initial post (as if it wasn't long winded enough) is that the last time this scenario happened, I had just installed the higher rated battery and did not completely lose power. This time however, the starter spun for about half a second then stopped. Following this when pressing the start button the starter solenoid clicked rapidly for a few seconds then I did have a dead (or protected) battery with no power output.

Dan

The ground wire should be connected to the motor. If you look where the clutch cable connects its just above that real easy to see. Check and make sure the connections clean . Seems best thing to do with starter is send it out and have it rebuilt. Not sure if there's anyone in your area .This guys good did my friends turbo starter,      Rebuilt Kawasaki Starter Motor 60_230 from Rick's Motorsport Electrics

84_GPz750 wrote:

Thanks a ton for taking the time for the write ups @Wookie58 and @slmjim+Z1BEBE. Thus far, no fuses have blown when this happens. I've been perusing the internet looking for a replacement starter motor but a new one runs $600 which seems steep. OEM PN is 21163-1053 if anyone has a recommendation, I know it was shared by the early Zr-7 and ZR750s.

One thing I should have mentioned in my initial post (as if it wasn't long winded enough) is that the last time this scenario happened, I had just installed the higher rated battery and did not completely lose power. This time however, the starter spun for about half a second then stopped. Following this when pressing the start button the starter solenoid clicked rapidly for a few seconds then I did have a dead (or protected) battery with no power output.

Dan

I agree with Riturbo ref rebuilding the original starter (much better option than another 40 year old part of unknown origin) couple of things to look at

• Do you or a buddy have another motorcycle you can try the battery on to eliminate it as a possible problem source ?
• When the fault occurs have you tried to turn the engine manually to eliminates the engine "nipping" as a possible issue (the starter can't turn a locked motor but it will try)

Wookie58: wrote:

• When the fault occurs have you tried to turn the engine manually to eliminates the engine "nipping" as a possible issue (the starter can't turn a locked motor but it will try)

   That's very good insight that completely escaped us. Kudos!

   Intermittent problems are a bane of any Tech's existence.  It's very likely the device will work properly in the Tech's presence.  Example:
   A copier or printer jams an average of once every ~3,000 pages.  When routine volume is 30,000+ pages per month, that's a lot of jams.  Most modern devices have diagnostics that indicate a probability (not a certainty) where the jam originated, but not why.  There will likely be multiple sheets of paper scattered along the entire paper path to complicate things.
   \ end rant on past career.
   A Tech, whether they be copier, automotive, motorcycle, whatever, never really knows if that last repair attempt really fixed it.  Not until an indefinite period of time has passed.

Good Ridin'
slmjim & 1BEBE

@Wookie58 I purchased a lawn tractor battery from my local farm store a few weeks ago to help in my troubleshooting. I successfully started the motorcycle 4 or 5 times and let it run for about a minute each time. No issues with the instant loss of battery charge but then again, I can do the same thing with the lithium battery as well at times. The issue is very random, sometimes it acts up the first time I hit the starter and sometimes its after a few starts.

I will get the lawn tractor battery hooked back up this weekend and see if I can do some more testing. As for testing the battery on another motorcycle, I don't have any friends who ride. The same issue has happened with three separate batteries, two of which were the same model and one that was a larger capacity. I have the rear wheel up on a paddock stand so I'll try to turn the wheel next time it happens to see if the engine is the source.

I found the ground point @riturbo thanks for pointing it out. The ground wire didn't actually run into the loom, it was just taped to it close to the eyelet and I didn't realize it was hiding behind the frame. Grounding point on the motor looked clean.