Starter Motor Issues

Good evening,

So I wrote two months ago about some issues that I was having, and with everyone's help, they were resolved. Unfortunately, on Friday I was on my way home after work, I stopped and got some groceries. When I left I got about half a block and the bike quit on me. I was turning right and I heard of pop, then a whiz/squealing. At first I thought maybe one of the grocery bags got caught in the wheel somehow, but the engine cut out and I coasted into the parking lot of a gas station. When trying to start it back up from the parking lot, it sounded terrible again, and there was terrible screeching from the front end, if I had to make a guess as to where exactly it was, I would say piston 1 or 2. I pushed the bike home from there, trying to start it occasionally. When I got home I put the bike in 5th gear and turned the engine by manually turning the rear wheel just to make sure it wasn't seized, and it wasn't.
The last time this happened it sounded the same, but after a few days I tried starting again and it seemed like the problem sound moved from the front of the engined to where it was the starter clutch not disengaging. On a member's advice I got in to where I could spray some carb cleaner in the starter clutch's general direction, and was able to successfully get the motor running fine, and I have put several hundred miles on it since. So I did the same thing this time, took the starter motor off and sprayed some carb cleaner in the hole to the starter clutch, I also took the alternator cover off so I could turn that to try and get an even coating on the clutch. I changed the oil out and saw a few flecks of metal on the bolt, but none really suspended in the oil. After I sprayed the starter clutch a few times and let it sit for an hour or two, I put a little seafoam in the oil, and buttoned everything back up and tried to start the bike up - same terrible noise. Just to be clear, the bike starts, and will sort of stay running if I twist the throttle, but I didn't want to let it run for long, just enough to get the oil with seafoam spread throughout the engine so I could let that sit for a bit as well before changing the oil again. Well, I guess it must have been a little too long (not entirely sure why, I wasn't on the starter button at the time, as I recall). The bike stopped and wouldn't start back up.
At this point I was just getting a click from the solenoid. I let it sit for a few hours and then came back, tried to start again, but now got a fast click from the solenoid. I tested the battery and it was at 9.5 volts when not under load. So I put it on the charger, then tried to start again when the voltage was back up - back to the single click. I tested for continuity based on Gamuru's guide and determined the issue was the starter motor. I just got done pulling it apart and quite a bit of stuff came falling out when I took the caps off. I have a refresh kit for it, but after taking it apart it looks like the copper on the non armature side is shredded on one side. Am I going to need a completely new starter motor at this point? I don't think the repair kit covers this...

 

Time for a replacement..........

 

Yes. That is an ex-starter.
A used replacement will be the most economical, but that one might not be past being rebuilt by an electric motor shop (which would be my second choice since these are commonly available used).

 

Thanks for your responses! I am ordering a Aftermarket/Rebuilt one from XJ4Ever. I will probably hold onto the starter for the time being and throw it in the "parts I could totally use if I needed to pile". Is it readily apparent based on the pictures what might have happened to the starter? Did I just fry it by being to overzealous with it? I didn't think I was being too rough with it, I wasn't laying on it for more than 5 seconds at a time. I guess the proof is in the burnt starter pudding though...

 

So that starter doesn't look burnt, it looks as though it contacted the stator magnets.

My question would be is it the starter clutch intermittently not free wheeling and turning the starter all the time, which would likely destroy a new starter if the starter clutch is not free wheeling? Or, is it electrical and the starter button or solenoid is sometimes sticking causing the starter not to disengage? The fact that the battery drained seems to imply that either the solenoid or starter button was sticking.

The terrible screeching was likely from the starter rotor rubbing on the stator magnets.

Does the end cap or bearing on the starter seem to have excessive radial wear?

Does the idler gear freewheel in one direction and not the other?

 

First off, please forgive my ignorance here.

The starter clutch may not have been moving at the time the incident described above occurred. I had an issue two months ago that presented itself pretty much exactly as this one did. At first I got the screeching, then it seemed, after some futzing with stuff, that the starter clutch was not disengaging after the engine started firing.


I don't think that it is the button, but I guess I am not positive. I was able to start the bike, then ride down the road a little ways before the problem happened both times this occurred, probably very close to the exact same distance too.


I'll have to upload some more pictures of the starter motor, I was taking quite a few as I was tearing it apart to make sure I didn't screw it up when I put it back together (I need to have everything spelled out for me the first time). I can't say I noticed any excessive wear, but I can't say I would know exactly what to watch out for. I can say that there was A LOT of crud in there, what I would think are ground up pieces of gears, but looking at my starter, and the one in the DIY starter refresh guide, it didn't seem as if I was missing anything. The idler gear (the one that sticks out of the one end cap, right?) did have difficult switching directions, but if I rocked it back and forth it would eventually change directions OK. However, when I took that cap off, the reduction gear assembly just sort of fell out. Those two smaller inner gears don't fit snugly on the posts that they should based on the pictures in the guide. I found, what looks like a single washer, that would fit in one of them (you can see this in the one pic above with all the pieces in it, the reduction gears are in the top right). Maybe what should be in there is what some of the crumbly bits are, in addition to the copper pieces.


The housing itself seems fine, and there were no (visual) signs of stress to the unit before I removed it from the bike and took it apart.


Not sure if it is a good idea (probably not), but if I were to try to pop start the bike with the starter motor off, and was successful without the screeching, would that help narrow down the problem?

Thanks!

 

With the gear reduction assembly removed, each end cap has a brass bushing to locate the rotor inside the stator. Excessive radial play in these bushings would allow the rotor to contact the stator.

This is the gear that the starter output gear mates with up inside the case. It should freewheel in one direction and lock in the other.

Yea, I see that now. So those brass bushings being destroyed would likely be caused by excessive spinning - or just from age, which would be the preferred answer.

Not sure about the pop starting. At the very least it may throw a lot of oil out the hole. Best to let one of the engine guru's comment on that.

 

Ok, I'll check the idler gear one I get home. I know that it was spinning freely in at least one direction earlier, I was using my finger to move it around to check for any broken teeth. I figured I would't want to run it too much because of the hole in the side of the engine, but thought maybe it might help to rule something out. Would I be able to remove the starter output gear and refit the starter for testing purposes?

 

To me it looks exactly like a starter that has excessive bearing wear. Too much axial runout will cause that damage.

You can bump start it, but the starter clutch will not come into play at all, and you will get doused in oil unless the starter mounting hole is blocked off. You can put the starter back together without the armiture and use that to block the opening. Keeping the planetary gears out of the gearbox might be a good idea too just to prevent rattling that may make you think that something else is wrong. This will be done so you can ride the bike, not for diagnosis because...

Get your new starter and use that to determine if the starter clutch has issues. There really is no other way to test it. The starter clutch only comes into play when the starter spins the idler wheel, which in turn engages the sprague drive on the clutch (via centripital force). Otherwise the starter clutch just sits there freewheeling.

 

K-moe, so you think I may not have tightened the starter motor down enough when putting it back in? Or just that the bearings failed. possibly from being too heavy handed on the starter button, and that caused the axial runout (my new vocabulary word for the day)?

 

OK, got your update. So when I get the new starter motor and put it in, if the problem persists, I should absolutely stop trying to start the bike via the starter and work on splitting the cases? Either that, or just start making sure I park at the top of a hill?

 

That all depends on what happens once you try to start the bike.
I am absolutely certain that your starter failure was caused by worn bearings, and not by the starter clutch or by your repeated attempts at starting. Remember that starter is over 30 years old and has been uncared for.
Your bike (if in a correct state of tune) is easy to bump start on flat ground (I've bumped mine a few times).

 

I don't have enough experience to say if you could ever wear out the bushings on a starter, but it seems a bit unlikely with a working system. That coupled with the fact that your battery drained makes it suspect that the starter was spinning continuously. I will add though that you should hear it spinning, as I just pressed the start button on my running bike at idle and could definitely hear the starter motor spinning.

So, when you get your new starter installed connect a DMM to the output side of the solenoid. During cranking the voltage should be around 10 to 12 volts. Once the bike starts it should drop to zero when the start button is no longer depressed and the solenoid opens up. If it stays at 10 to 12 volts, then something electrically is at fault not removing power from the starter. And, if it does not drop to zero but instead outputs a voltage that is dependent on rpm, then the starter clutch is not freewheeling, and has turned the starter into a generator - not sure what the voltage output would be in this case, but suspect it could go quite high at higher rpm's and anything other than zero would be an issue.

Also, if all seems OK at first, be sure to repeat the test after a good warm-up since on both occasions it seems you had issues after stopping momentarily on the way home.

 

OK, I will respond once the new starter motor is received and I get it installed on the bike. You are correct, I have had very little issue bump starting my bike. I still need to sync everything using the tool I purchased, but even completing the valve adjustments yielded a noticeable improvement.

 

The only other possibility (and one that I have also seen) is that the adhesive holding one or more of the magnets to the can let loose, and the magnet bashed into the armiture as a result.

 

Well,I just bump started the bike successfully. It was hesitant at first, but after about a quarter mile evened out. The idle is higher than normal, which was kind of weird. Usually once warmed up fully I'm idling right around 1100, but it is going at about 1800 now. Would that be due to not having the resistance of the starter motor, or the seafoam in the oil, out both?

 

I suppose if the starter clutch was stuck in a way that kept the starter constantly turning (unlikely but possible) that would account for your idle speed rising. That would also explain any bearing wear that may have occured in the starter.

 

I guess we'll see when we get the new starter motor hooked up. Rooster gave me some good things to look for, one of which covers that possibility. I wonder if this could have been somehow tied to the accident I had late last year. The mechanic I took it to said that he went out to test it, but they were having a hard time starting in so they jacked the idle up. They said they changed plugs afterwards too and were able to bring the idle down a little. Maybe the accident buggered up an already brittle starter or something. I don't know, I'm probably grasping at straws at this point.

 

Ok, I got the starter (thanks, Len!) And it is still being drug along. The voltage when measuring the output from the solenoid continues after the motor catches and the button is no longer being pressed. The voltage starts going up as RPMs increase, going up to near 40 volts I think (my analog multimeter goes from the 10 up to the 50 range). Do you know how to isolate whether it is the starter clutch or something electrical telling it not to disengage? Or do we already know the problem is the starter clutch? Another think I did was checking continuity from negative on the battery to the case, and I get the tone.

 

At this point I'm going to say that we know the problem is the starter clutch. The increasing voltage leads me to believe that the starter is being turned by the engine and turning it into an unregulated generator.

 

Well, that is a bummer. I tried the seafoam and that didn't work. Spraying with carb cleaner as well. On the xj4ever piece it mentions that you might need to try a solvent. Are there any suggestions for this solvent?

 

You have already used the most often reccomended solvents.

Here is my WAG. The starter clutch body may be cracked in such a way that the rollers remain engaged. I have never actually heard of that happening, but it is quite plausible given that there have been cracked starter clutch bodies found by several members.

 

Fair enough. I accept my fate.

 

Please document your findings for the collective.

 

I will make sure to take lots of pictures, I hope to get enough together to make a coherent step by step since this seems to be becoming a relatively common occurrence. The closest I've got to anything like this was testing down a B&S lawnmower engine for a small engines class, so it should be interesting. But, if I can do it, anyone can. That having been said, it will probably be a while until I get to it. I'm finishing up my major right now (I'll still have some gen eds to pick up afterwards), so I'll have a month off school, but I also have a 3 week old that likes a little of my attention (a do her 3 siblings). It may take a while, but I will get the bike going under is its own power again.

 

Hard to believe the actual starter clutch would fail that way especially if you can turn the idler gear in one direction, however just about anything is possible. Is it a possibility that the idler gear / collar is not being oiled properly and is trying to seize to the alternator shaft? Are all of those components lubricated by the oil nozzle?

Edit: So I just checked the lubrication diagram, and oil gallery B should lubricate the idler gear / collar via a hole in the alternator shaft. Looks like the nozzle is just for the primary chain.

 

Correct.
Like I said, this appears to be a new sort of failure. I'm very interested in seeing exactly what is going on.


Here is a thought. Disconnect the starter positive lead and insulate it. Bump start the bike and check for voltage between the starter post and case. That will tell us if the starter is really being spun by the starter clutch or not.

 

Ok, to make sure that I am understanding you correctly:
1. Disconnect the solenoid connection that is NOT going to the battery's positive terminal and insulate it.
2. Bump start bike
3. Put one end of the multimeter on the post on the starter and put the other end on another part of the starter?

I will say that I haven't had much luck bump starting the bike with this issue, and by that I mean no luck. The engine really struggles to deal with the extra drag. I had to really open the throttle to test the voltage at the solenoid earlier.

 

1. Disconnect the only wire that is hooked to the starter.
2. Bump start the bike
3. Place the positive lead on the starter post and the negative lead anywhere on the engine case or starter case.

If there is that much extra drag then forget bump starting it. Something is not right on the inside and the only way to find out what is to look.

 

Following up here since I realize this thread never had a "conclusion". The issue was the bolts in the starter clutch backed out, forcing it to be engaged all the time (causing the squealing noise mentioned in the thread). I had to split the cases and replace the starter clutch. Really, I probably could have re-used the old one and just replace the rollers and springs, but I figured I would like to prolong the need to do this again as long as possible. The whole thing can be seen documented here.