Marbles in a coffee can.

1981 XJ650 Maxim

ok, so I was on vacation and didn't ride for two weeks. Went to start bike to run to the store; push button, starters says, crank - stop, rinse lather repeat for half a dozen times. Crank and then nothing . . No marbles yet. Finally cranks over and fires up.

I did recently change the oil and used this: http://www.valvoline.com/products/brands/valvoline/motorcycle-atv-oil/13

Got to the store did my thing, came out and the 'marbles in the starter' noise was all I got, a whole lot of it . . . enough to make me think, "this is going to be a long walk home with a 30lbs. bag of dog food". Finally tapped the starter a couple times (not sure it did anything or just pure luck). Bike finally started, got home, shut bike off, tried starting a few minutes later and the marbles were there.

So, I know the starter is 34 years old, figure I am do for this: http://www.xj4ever.com/starter brush replacement.pdf
Ass-u-me that the brushes are not providing enough umph to get things engaged?
Anything else?
Could this be the starter clutch failing?

 

yes, but "could" is the key word

 

More likely either the starter is in need of cleaning/servicing, or your battery is going

 

Mileage (actually number of starting cycles) matters more than age when looking at the health of starter clutches. Charge that battery up and see if the clutch grabs better. What condition are the starter brushes in?

 

I'm in the middle of repairing the starter clutch on my 550 Seca. Got the case split and removed the clutch and it would work in my hand. As I removed the starter shaft I found black gunk on it. It was causing enough drag to keep the clutch from releasing. I wonder if adding Seafoam before an oil change would have prevented this from happening. I use Rotala 20W40 oil. So if you can get your bike to start I would try some Seafoam in the oil to see if this fixes your problem.

 

50K miles and as far as I know no one (PO or myself) has looked at starter brushes . . . . guess I know what is next for me.
Noted that there is a minimum dimension in the manual - will reference that.
Battery was new last year and spent winter on a trickle charger - h20 levels are good - easy enough to stick a meter on it.
Len had a write up about cleaning the clutch with Seafoam, drain and replace oil (as lostboy mentioned). http://xjbikes.com/forums/threads/xj-750-starter-clutch-going-out.59675/#post-450899
Just a shot (1 oz.) of Seafoam, or load it up with a couple glugs?

Here comes the questions that may make me sound stupid . . . but I always prescribed to asking the dumb question prior to doing something dumb.

The starter is removed with the two bolts on the left hand side. The starter clutch is not a part of the starter but inside the motor case - yes?

Replace the brushes, clean commutator (600 grit emery ok? wet or dry?), clean gears and re-grease is the extent of starter 'servicing'?
The manual mentions measuring the depth of the grooves between the commutator contacts, since I will have it in my hands I will do that also.

 

use crocus cloth or an eraser . the seafoam can tells you how much to use I think its 3oz

 

If the commutator cuts are too shallow they can be carefully recut with a thin hacksaw blade. Be sure to clean the motor out with electrical motor cleaner. Other solvents can damage the lacquer on the windings.

 

My Maxim's starter got the marbles... Rebuild... My Seca had a slippy starter clutch... PO split it's cases but it still happened and I think that's why he sold it. Changed the oil a few times with Castrol 20w50 dino-oil and it hasnt slipped in two years. Seca then developed a case of the marbles but luckily my maxim starter was handy (RIP). I'd get a new set of brushes and have at the guide, not a bad job. Just don't break it all apart on the planetary (gear) side as there's no need. That reminds me, I have to change a set of brushes myself

 

Actually there is no way to tell if the starter gearbox needs service without opening it. Grease gets old and chunky. Old and chunky is not good. Old and chunky causes drag and accelerated wear. Fifteen additional minutes of effort yields another 30 years of service life.

 

UPDATE:
Thanks for tips!

Yesterday had time while wife was working in the garden to pull the starter. Prior to taking wrench in hand I checked voltage on battery, then tried to start the bike - fired up immediately! WTF? Checked voltage again on running bike.
Well starter is 34 years old and my intent is to service it. Start with two mounting bolts, then went to the positive lead connection . . . . . It was loose!

Grease was old and halfway between smooth and chunky, brushes still have .10" of wear left on them (measured .43"), cleaned commutator, checked depth of cuts between contacts - used an old mandolin string (.024") as a go/ no go gauge, this was a sketchy way to check and the string would sit higher in a couple slots. Used a hobby file to deepen slots so the string fit flush. Bushings were good, o-rings ok. Added fresh grease, put it back together and starts right away.

The lead connection was loose but not sloppy loose . . . electricity is magic to me- - Does it make sense that a less than solid connection will not allow enough juice to go to the starter creating the 'marble effect'?

 

Yes it makes sense. Good physical contact is critical in an electrical connection. A loose connection has the effect of increasing resistance and reducing the amount of current that can move through that point.

 

My bike is doing the exact same thing. when I engage the starter the thing sounds like a mechanical mess. now, that’s with the original *and* another, rebuilt starter. It isn’t the starter. It’s either the clutch or possibly, *possibly* bad voltage at the starter itself, but I’m leaning more toward the clutches cuz it sounds awful.


Last thing I want to do is split the cases.

 

Two days ago, she started right away.

Today push starter, she tries to go and then stops - stops dead like. Disassembled the starter switch and the contacts were dirty on the button and the 'receiver', in fact there was a small dimple on the receiver. Filed it flat, added die-electric grease and put it back together. Same symptom, crank and stop (occasional marble noise, but nothing like before).

Finally got it to start checked voltage at battery:
12.0v at battery with bike off
12.2v at idle
13.3v at 4K rpm

Battery is a year and a half old (wet cell/ lead) was kept on a trickle charge during off seasons.
According to the manual I should be seeing 14+ volts at 2K rpm (ass-u-me the same holds true at 4K).

Ass-u-me that the alternator/ "A.C. Generator" is amok.
In the manual it says:
"Resistance test of field coil and stator coil. Check resistance between terminals. If resistance is out of specification, coil is broken. Check the coil connections. If the coil connections are good, then the coil is broken inside and it should be replaced." Then there is a chart with Field and stator coil resistance (4.0 & 0.46ohms respectfully).

As stated before - electricity is magic to me . . . .

1. What/ where are the terminals I should be checking across?
2. Should the battery be connected when performing this test? (battery is on the trickle charge and shall remain overnight - showing full charge now)

Oh gods of all things electric please impart any further wisdom on this. My desire is to become wise of the magic which is called 'electricity' and to make its sometimes evil ways work in my benefit.
amen.

 

This is going to sound weird but......check your oil level.

Your charging voltage is indeed low. Have you gone through and cleaned all of the electrial connections on the bike? Loss due to oxidation (increses resistance) could explain the 0.7 volt loss that you see at the battery.

There are two sets of wires coming out of the alternator housing. One set is three white wires. Those are the alternator coil wires.

 

Yep, you are correct, that sounds really weird- but I have been here long enough to embrace all that is weird.

Will check the oil level.
Prior to checking ALL electrical connections I would like to verify the alternator.
How do I know which wires to test for resistance if they are all white?
Manual says - Green / Brown for Field Coil and White / White for Stator. . . . .

 

check resistance between green and brown. check resistance between white and white.

 

And for the stator, there are actually 3 white wires (call them W1, W2, and W3, although there is no specific "order" to them to be concerned about) and you need to check the resistance between each set:

W1 to W2
W1 to W3
W2 to W3

 

Oil level is good.

Battery was fully charged this morning (12.9v), bike started up with choke full on. Bike idled with choke on for about 5 minutes as I searched for my riding gloves (turns out they were in my riding jacket pocket instead of the helmet where I usually keep them, jacket was hanging on the fence so . . . yada yada). I digress.

Get on bike, back off on choke and she stumbles a wee bit. Put it in gear and start to give it throttle and she dies!

Now she won't start, same as before, cranks /stops, need to wait a moment prior to hitting the starter button as nothing will happen. Battery is now back down to 12.0v, pointing towards a loose/ corroded connections somewhere???

As soon as the time allows I am off to check the alternator and then all connections.

Any other thoughts?
I am deeply puzzled by the bike conking out when I gave it some gas .

 

Does the neutral light dim when it won't crank? Does the oil light illuminate when you push the start button and it doesn't crank? Do you hear the solenoid click? When you check the battery voltage place the meter leads on the cables, not the battery terminals to verify the connection. Make sure voltage stays above 9.5 volts when starter is engaged.

Clean up the battery terminals and cables with a wire brush if you haven't done so in a while, including the negative lead that connects to the engine.

If no cranking again when the start button is pushed, do steps 1 through 4 in Gamuru's how to guide:

http://www.xjbikes.com/forums/threads/how-to-test-your-starter-circuit.7828/

Caution: bike on center stand in neutral when doing steps 2 and 3.

 

I recall seeing the oil light dim when cranking, can't recall what the neutral light was doing. Solenoid is clicking.
Will go through Gamuru's most excellent write up (thanks for the link). Crossing fingers that I don't end up at step #9!

 

Yes, step 9 is a very bad place to end up at.

If the solenoid is clicking and the starter is not turning then either the solenoid or starter is bad. Since you just rebuilt your starter then odds are the solenoid is on its way out. Hopefully, doing the testing suggested by Gamuru will help you isolate it for sure.

 

Should have added if the battery and connections are in good working order as verified with a DMM.

 

Went through Gamuru's write up and stopped at #7.

Hooked up the continuity tester to the blue w/ black stripe wire - not the Sky Blue as directed (at least it didn't look sky blue to these old eyes) and the negative battery terminal. Had -.028 reading on the meter in both neutral and in gear. Write up says that circuit should be 'closed' when in neutral - to me and my limited 'lectricty knowledge - closed continuity is when there is a reading on the meter - 'open' is OL (overload?)

According to the write up, failing step #7 means that the neutral switch should be inspected, cleaned or replaced. Now - the neutral light works, when in neutral it stays on, when in gear it shuts off. The neutral switch from what I can tell is underneath the engine near the left foot peg but obstructed by the frame . . . I couldn't see any sign of it (1981 XJ650 maxim - which does NOT have a side stand switch) - I did see what I ass-u-me is the oil switch, about the size of a quarter, appears to have an epoxy coating and held in place with two small bolts with a wire running into the epoxied center. Did NOT see any other wires on that side of the underneath except for the wire running to the oil switch.

So where/ how do I get to the neutral switch and by what means should it be tested/ inspected?
As mentioned the switch appears to work (at least for turning the light on/ off on the instrument cluster).

I am glad that I didn't make it to step #9 but am still puzzled.
Meanwhile I did pop the cover off of the alternator and was greatly surprised to see the brushes look great - cleaned contact surface and brushes (not very dirty). Have not checked resistance on it yet. Was lost in "can't find my neutral switch land".
TIA

 

here's where it is on a 750, i imagine a 650 is the same

 

It is, good picture. Note that Polock has an aftermarket exhaust on the bike, so that switch appears pretty easy to get to.........on bikes with the stock exhaust system, that switch is pretty well hidden and a beast to get to.

 

i also have oil and dirt on the bottom of my engine..........i'am so ashamed

 

It appears you were troubleshooting a working system, the joys of electronics and intermittent failures. Gamuru's step 2, 3, and 4 would be good advice when you have the situation of you push the start button and nothing happens.

Only blue/black wire I can locate is connected to the Headlight relay, and since it also powers the headlight and indicator bulbs, 028 ohms to ground is likely normal. Where was the blue/black wire that you were checking?

If your neutral switch was bad as in shorted all the time, the neutral light would stay illuminated all the time. Also, if the neutral switch was defective (open) deploying the safety circuit, then the solenoid would not click when the start button is depressed. Hopefully, you can get back to the failed condition of pushing the start button and the starter does not spin. If you can:

Verify battery voltage at the battery cables when the starter button is depressed, not the battery post
Perform steps in Gamuru's guide to isolate to starter, solenoid, or starter switch.

 

So, neutral switch would require removal of exhaust . . . . BUT may not be the problem. ( thanks for pic Polock)

Rooster, the blue /black wire is under the side cover on the left hand side (opposite battery) on a pigtail that was plugged into a relay (small cube). Total of 4 wires on the pigtail.

I did verify voltage at cables, but did NOT push start button to check any voltage drop (Gamuru's write up has continuity checked when starter button pushed). Will need my assistant (wife) as my small Radio Shack meter has only points and not alligator clips

 

you can pick up alligator clips for a couple bucks at autozone! i did that and modified my meter and i like it a lot better now. i made the ground suuuper long so the thing would be more mobile and it's worked out great.

 

Checked voltage at battery cables whilst the assistant pressed the starter button - first crank voltage dropped to 9V and then turned over. Did it again and dropped to 10.5v and turned over . . . . . The bike will now crank and turn over (start)!!! I'm cautiously happy as the issue appears to be resolved . . . but not sure why.

However when I started on the second chapter of this I cleaned the battery cables/ connections and even went as far as disconnecting the negative cable where it mounts to the motor . . . when I pulled the looonngg shoulder bolt out of the motor it was covered with a white chalky type residue - similar to a build up that sometimes occur on electrical doodads (the magic of electricity again!). Although this residue appeared isolated to the shoulder - saw nothing on the 'spade' connector nor the underside of the bolt head and nothing on the end of the threads.. I did clean this all up and smeared it really good with die-electric grease. (is that grease conductive or is it just a corrosion inhibitor?).

Need to get out for a test ride, raining now, and the weekend has a honey do list that needs to be modified and reprioritized to have "Take XJ for a test ride" as #1 . . . . .

 

I didn't realize "beaver falls" My dad went to Beaver High, Callaghan(age 78),from Monaca.
Marbles: I was reading old thread on the sound. The only time I hear the effect, is when my battery is getting low. A fresh battery cranks over with no "marbles in a can" sound.
When working on the bike , I often forget to turn off the key so the light goes out. I am guessing during sync or carb adjustment with lots of restarting that it is easy to run the battery down.
Am I right in this thinking?

 

Correct. Anything that would cause the battery to have low voltage will make the starter not spin fast enough for the starter clutch to engage properly.

 

What regenerates the voltage in the battery, simply riding around, correct? Idling does nothing, and the light draws power down too.
I guess this also means I have to read the starter revitalization thread. We actually have a small shop in town that does windings on alternators, starters,
I think he stays very busy. If anyone needs such info I can get you his number, so you can send your clunky old thing, and get a rewound clunky new thing.

 

Almost always the starter just need cleaned out, new grease in the gearbox, and new brushes. Rewinding is only called for if the commutator has a dead spot, indicating a shorted winding.

 

+1 on kmoe
Also, a cheap or old battery will draw down quick as it looses its reserve capacity fast.