OM617 Flywheel missing teeth ... cause?
 

I just rebuilt the starter on my '84 300D and when I went to check for broken teeth on the flywheel I found some missing - so I am trying to figure out why.
What I found interesting is that the missing teeth correspond to the top of the compression stroke on at least one of my cylinders. I advanced the cam 2 degrees and injection timing 3 degrees years ago. I am wondering if anyone else has had broken teeth possibly caused by advancing the cam or IP.

I cannot see any reason for advancing the cam or IP to break teeth off the flywheel. If you have broken teeth there I would suspect a misalignment of the starter. or possibly a starter with the wrong number of teeth on its gear.

How long have you had the car? Do you have any trouble with the starter slipping during cranking? The only thing I can imagine would damage the flywheel is the starter... Aside from an errant screwdriver impact...

Unless the motor was kicking back against the starter, your adjustments didn't cause the failure. And even if it was kicking back, the starter drive typically fails first by splitting in half.

Now for a detailed description on how a starter works.

When an engine stops, it is typicaly against compression and the flywheel will roll back a bit. This leaves a small range of flywheel teeth that are at the starter engagement point.

When a starter is commanded to operate, a sequence of events must occur to prevent grinding. For a solenoid operated starter, there are 2 coils on the same bobbin. One is a high current strong pull in and a low current weaker hold in coil. ( Have a look at a good wiring diagram that shows the internal workings of a starter and the below will make more sense. )

The current path for the pull in is from the battery , ignition switch, small cranking terminal on the starter, pull in winding. The other side of the pull in winding goes to the starter motor. What this does is allow the starter motor to slowly rotate as the drive is moving towards the ring gear. The hold in winding is also on the cranking terminal and the other side to ground.

Once the drive bottoms out, it triggers a set of contacts that apply full power to the starter motor bypassing the pull in winding. Since the hold in winding is across the cranking terminal and ground, it keeps the drive pulled in.

When the tips of the starter drive wear, the timing between pull in slow rotation and full power rotating becomes incorrect causing the starter drive to rotate at full power and impact the flywheel teeth. This will fatigue the teeth causing them to break.

There are a couple of other things that will cause teeth to shed. Cranking, letting off the key then recranking before the engine and starter have come to rest, this is a sure way to damage parts. A slipping starter drive one way clutch is hard on teeth too.

The other is that the flywheel was plain fatigued and needed replaced. What started out as a fuel tank cleaning on my IH backhoe ( German Neuss engine ) turned into a ring gear, engine bearing and seal replacement. ( Pulled tank to clean, easy access to starter, pull starter to inspect drive, found broken teeth on the flywheel, pulled motor, gave high wear items a freshining. )

Your explanation of the solenoid is pretty nice - it also explains of the "no click, no action" type failures experienced by starters.

Walgamuth: The starter and flywheel are original, so the number of teeth has to be correct. The transmission was replaced about 3 years ago, and the alignment could be an issue, but I saw no problems when pulling it out to indicate that.

Vstech: I have been driving the car for 4 years now, and the starter has been making an occaisional grinding-screeching "slipping" sound for the last 2 years. After the first time it made that sound it wouldn't start and I called a tow truck. The driver suggested trying to turn the crank a bit to see if the starter would engage and his trick worked. I used an adjustable wrench on the power steering pump nut to nudge the crank. That wrench stays in the car all the time now...

SL320: Wow... that was detailed. Thanks!! - I have hunch that you are correct about the timing between the solenoid pull-in and the motor full power cycle being off. When you think about, there are several wear points that could affect this timing. The drive gear teeth are indeed worn, but the nylon solenoid plunger link, the yoke pivot, the slip ring, etc. are all worn. I wonder if it would be possible to shim or otherwise adjust one of these components to cause the gear teeth to engage earlier... :rolleyes: Don't get me wrong, I am perfectly willing to order a rebuilt starter, but I can't help wondering.

the starter has been making an occaisional grinding-screeching "slipping" sound for the last 2 years

This was your signal to act right away.

I sometimes can drive for days without the problem occurring. Not sure the problem has been made worse for the driving. The very first occurrence mandated a new flywheel and starter anyway. I have it up on jacks right now, so it's do or die time... :D

The real answer is "bad dental hygiene".

Dan

Shimming where the starter bolts to the engine would make things worse from a timing standpoint. Shimming the solenoid would help delay full power.

A your welcome to those that appreciate the detail. There is way more going on in seemingly simple systems that it appears. I've been in and around the automotive business for 40 years and always have looked at things from an engineering perspective.

To expand on this: As you figured, no click no action can be caused by the starter motor portion not having continuity. When this occurs the pull in windings don't do anything and the hold in windings are too weak to pull the drive / solenoid contacts.

Sometimes you will get a partial / weak click usually with a hot engine. This is caused by an otherwise good starter not getting enough amps through the starter trigger wire. A very common fix ( and many times a factory design ) is to use a Ford style remote solenoid ( relay ) to trigger the starter. This is also called a "hot start kit". The Ford solenoid takes much less current than a starter trigger.

If you get a single click and no crank ( with a good battery ) , this is usually the large battery power wire at the solenoid or the solenoid contacts failing.

If you get a violent chatter when trying to crank with a good battery and cabling, the hold in coil in the solenoid is bad causing the drive to pop in and out of engagement when the pull in coil is switched out of the circuit.

oi... that's it!
 

I get a weak click with a hot engine, followed by the screeching/grinding but no engagement (not always). I was going to mention the hot engine variable... I was wondering about the start trigger wire supplying enough amps to the pull-in windings. I have a dc solid state relay (ssr) that I can use to switch the current ( I measured 11 amps) and just use the starter trigger as a low level signal. That should improve the hot start issue.
To improve the pull-in / motor start timing issue I fabricated a .090" stainless shim that slips into the nylon link: https://photos.google.com/album/AF1QipM_A_7XykROfbuGEj9paTlH6sotdJFSe0FakD-r - I am hoping that this will not put undue stress on the nylon link...
I am also making a new 2/O cable from the front battery post (battery moved to trunk) to the starter to minimize voltage sag under load.

A couple quick things.

Inrush to the starter trigger wire will be high, many amp clamps are heavily dampened so you might not be seeing the peak. ( RE: you might just be seeing the hold in coil. ) I haven't measured pull in so can't give you an exact # but do know it is reasonably high.

I use SSRs on an industrial basis but would not use in your situation, they are generally used when switching is frequent and a mechanical contactor ( relay ) would wear out rapidly. Electric resistance heating is a prime example.


When a SSR fails it usually fails on so your starter will remain engaged. In a heating app there is a mechanical contactor in series triggered by main power / high limit sensor. Even when turned on, the SSR has some voltage drop across it and you will loose a bit of voltage.

A SSR generally does not like inductive loads so a flyback diode across switched terminals will be needed. When a coil is energized and the power removed, the magnetic field collapses. This energy needs to go somewhere so the coil now becomes one side of a transformer giving a high voltage spike.

The pic does not work for me, google wants me to make an account.

With a remote bat it is OK to put the neg terminal to the body but be sure to bond the engine to the chassis. This is how MB does it from the factory.

Don't stray too far from factory parts in good condition as the timing will take care of its self. If the FW is missing teeth, change that before going any farther as the situation will degrade rapidly once another tooth is lost.

I am trying to paste the images directly into this post. We'll see if the moderator will tolerate this...

https://lh3.googleusercontent.com/FO...Y=w288-h511-no

https://lh3.googleusercontent.com/ou...M=w288-h511-no

OK - hope that worked.

Thanks for your thoughts on the SSR. I forgot that they fail to a shorted state - not what I want.

You are right about fixing the flywheel, that has to be next.

Appropriately sized pictures are OK in a post.

On your starter I don't know what bottoms out first, the drive or the solenoid core. If the drive is what bottoms out first, there may or may not be enough solenoid core travel to trigger the starter motor to turn. If this occurs you will just get a click ( and that would be a good time to measure pull in coil amps, just don't hold the key in the crank position for more than a second or two as the winding will become overheated. )

As it sits on the bench right now the starter operates perfectly with the shim installed. What I don't know yet is what it will do when re-installed on the vehicle. I am planning to re-install it today to check and see if engagement with the flywheel prevents the solenoid from pulling in the motor contacts. First, I am upgrading the start signal circuit using a 30A Mbz relay.

BTW... I checked the Mbz starter wiring diagram and surprised to see that there is no relay in the circuit. The driver's key switch handles the full load! I also noticed that the start signal current path is directed through the neutral safety switch on the transmission before it finally terminates at the starter. I checked my wires and noticed that there is worn insulation down near the transmission, as well as old and degraded fiberglass insulation pretty much everywhere under the hood. It looks as if installing a relay to "amplify" the start signal is good preventative medicine...

I also did the math to determine if upgrading to a 2/O starter cable would be worth the effort and found that it would reduce energy lost in the cable by about 20 watts, which isn't much for a motor drawing around 2kW - so decided to just clean the terminals, use a corrosion inhibitor, and leave it stock.

Yeah, the keyswitch contacts are fairly robust, and the starter solenoid is a big relay, so not much load on the key...

problem fixed!
 

I measured 11 amps on the hold-in coil and 43 amps on the pull-in coil. The pull-in only activates for a fraction of a second, but still draws a lot of current. I saw degradation of the start signal wiring circuit in 3 places that most likely caused problems with the speed and timing of the starter pull-in. This ultimately cost me 2 lost teeth on the flywheel.

The three problems with the start signal circuit path that I witnessed were:

1.) the wire bundle near the neutral safety switch on the transmission had lost some insulation due to friction & vibration. I do not think that it had shorted to ground before I discovered it.
2.) on the passenger side front fender well, the white start signal wire is bundled with the motor power wire and passes through a plastic wire support grommet about 10 inches from the starter motor. That wire had lost its insulation and some copper due to vibration within the bundle at the location of the grommet. It was not visible or apparent until I removed the wire bundle completely.
3.) the motor windings themselves complete the start signal circuit path and accumulated dirt & grime on the brushes & commutator created a higher-than-normal impedance in the ground path.

The start signal path is long, maybe as much as 10 ft. of wire going through 7 or 8 junctions. If any of the wire or junctions are degraded due to age, oxidation, grime, vibration, etc. then the signal path will have a total impedance that is probably out-of-spec according the the original design. The higher-than-normal siganl path impedance caused a slow and/or late engagement of the starter pinion gear which ended up breaking 2 teeth on the flywheel.

What I did to fix the problem:
- I rebuilt the starter motor. The commutator was in good condition and only needed to be reconditioned on a lathe using a file and a scotch-brite pad. The brushes were in good condition. The motor was cleaned, lubed, and reassembled with a .090" shim installed in the solenoid plunger nylon link to advance the pinion gear engagement timing (see photos in previous post)
- I installed a 50 amp relay in the start signal circuit to "amplify" the signal close to the starter motor.
- Since I removed the starter wire bundle, I ended up replacing the #2 motor power cable with high quality 1/O welding cable.

How it performs now:
The starter engages and spins the motor with authority now. The 2 missing teeth do not seem to affect anything unless the motor comes to a stop with the gap lined up with the starter gear, which happens now and then because the gap is located just shy of TDC on the cylinder with best compression. This is not the best solution - but it will hopefully get me by until I can find the time to go pull a good flywheel, get it match balanced to mine, and replace it.

Yep, 43 A is a lot and why some use a "hot start" relay. What did you use to measure current and did that include the starter motor as installed on the engine or running free on the ground?

You have a good thought process going in order to solve problems.

Thanks!
I used a Fleabay DC clamp-on ammeter.
I measured the solenoid pull-in coil by energizing it with 12V pos. connected to the small screw start signal terminal on the solenoid and 12V neg. connected to the copper braided link between the solenoid and the motor. The motor itself was not energized.

The ford solenoid as an add on is cheap. For some reason many German cars in my experience seem to have a lack of a good strong trigger current. As they age in general this situation usually does not improve.

After I measured 43 amps going to just the pull-in coil, I reached into my box-O-relays and grabbed one of the beefier ones. It is a bit of a mystery as to out how the engineers at MB decided on such a tenuous start/trigger signal supply to this obviously hungry load, maybe they reasoned that the duration of the demand was so short that it didn't matter... or perhaps they were intending the overall impedance to act as a sort of current limiting "ballast" of sorts.

I question the 43 amps measurement of the pull in coil. You used a cheap ebay DC clamp on meter as I recall. I am not sure how accurate that meter can measure a signal that is a few micro seconds long. I know the hold coil is around 8 amps (measured in series with a digital ammeter), so my guess is the pull in coil is not much more than that. Maybe you can measure the resistance of the pull in and hold coil and calculate what the currents should be?

See his post # 21.