48 DeSoto New problem. Starter pinion grinds and clashes half of the time.

[marcapra]

I had an electric shop rebuild my solenoid switch a couple of months ago.  The starter was rebuilt back in the 80s, and they said it worked fine.  But in the last few days, the starter pinion gear has started to clash with the ring gear about 50% of the time.  I checked the two bolts that hold the starter and they are tight.  I guess I will have to take it to the electric shop again.  Has anyone had this problem before?  

Edited May 4, 2023 by marcapra (see edit history)

[Andy J]

Yes,I have. Thinking my Bendix was bad,I replaced it and nothing changed. I noticed the Bendix wasn't fully engaged with the teeth on the ring gear.I replaced the solenoid and that fixed it. My starter is a Delco and is on my '51 Pontiac 8 cylinder.The solenoid is mounted on the starter.You didn't specify what type yours is so it may be different from the one I'm describing.

Edited May 4, 2023 by Andy J (see edit history)

[marcapra]

I have an Autolite starter and the solenoid is mounted on top of the starter like yours.  How did you see that your pinion gear was not fully engaged with the ring gear?  My solenoid was just rebuilt and the starter was rebuilt in the 80s, but never used until now.  Was your problem intermittent like mine, or did it grind every time?  

 

[Bloo]

Are you sure the gears are clashing? Generally speaking, cars with a solenoid mounted on the starter have no Bendix, but have a one-way clutch inside the starter's drive gear. The one-way clutch can fail, and the starter makes a sort of grinding BJEEEEEEEKKKKkkkk!!!!!!!!!!!!!!!!!!!!!!!! noise when the one way clutch fails to engage. It is often intermittent. If that's the case, replace the drive gear.

 

If the gears are truly grinding, take the starter off and look at the starter's drive gear. Usually the teeth have a ramp shape at the tip to help engagement when the teeth are not aligned perfectly. See if the front of the gear teeth are ground flat from failing to engage. Replace if so. Then, take the spark plugs out and rotate the engine with a wrench to look at the ring gear teeth. The wear will not be even. The engine stops in just a few places (three or so) when you shut the engine off, so the starter will be engaging in those same few places most of the time. You'll have plenty of good teeth to look at, and if the teeth that are actually used all the time by the starter have a lot of damage you will need to replace the ring gear, or in some cases you can flip the ring gear upside down.

 

[marcapra]

I looked at my ring by taking the starter out and rotating the engine with the fan.  Just like you said certain areas look untouched and other areas of the ring gear are chewed up.  This just started to happen.  It was fine before.  I'm thinking of replacing my ring gear, but I don't have the special tool that the shop manual says you use to get the ring gear within .020" out of round.  Can someone who might have the tool loan it to me?  I put the existing ring gear on back in the 80s with the help of a metal shop teacher without the tool.  That always worried me a bit.  

[Andy J]

Mine was intermittent also.I took the flywheel cover off and got my wife to work the starter while I watched from underneath.It was easy to spot the problem.

[marcapra]

My starter mechanic told me I could try to file the scarred teeth on the fluid drive, but I don't how that would solve the problem.  There might be an alignment problem?  Odd that this starter worked just fine for about a month and then started grinding.  I wonder if you really do need that special tool for checking out of round when you weld on a new ring gear, or if there is an alternative method that could be used.  Seems like you might be able to set up a dial gauge off the side and put it on the ring gear, and then turn the fluid drive around on its driving plate to check that there is no more than .020 run out.  

[Oldtech]

Is the bendix gear coming out all the way when the solenoid engages? Some starters you can slide the solenoid to position the gear all the way out when the solenoid is engaged. 

[marcapra]

I can can pull the pinion gear out all the way easily.  I think it works fine.  I'm going to dremel the chewed up teeth on the ring gear and see if that helps.  

[Bloo]

 I

19 hours ago, marcapra said:

My starter mechanic told me I could try to file the scarred teeth on the fluid drive, but I don't how that would solve the problem.

Look closely at the shape of the good teeth. Is there anything special about them? More cars than you would think have flat faced teeth from the factory on the ring gear. There might not be much to do. If there's burrs hanging out into the gap between the theth it couldn't hurt to file them off.

 

19 hours ago, marcapra said:

There might be an alignment problem?

Hard to say without any pics of what we are working on. USUALLY Chrysler products have a machined hole for the starter to locate in, and mounting bolts that run parallel to the armature. If Chrysler got things machined right in the first place, and they almost always did, there is nothing to do. My knee jerk reaction is there is not going to be an alignment problem.

 

On 5/3/2023 at 9:41 PM, marcapra said:

 But in the last few days, the starter pinion gear has started to clash with the ring gear about 50% of the time.

Look at the drive pinion and compare it to a new one. Pinions usually have tapered teeth on the engagement end. That's important. If they were tapered and they are ground flat now, it would explain intermittent engagement. They WILL look all beat up. That's normal, and part of the job. Look mainly at the shape of the tips, on the side that crashes into the ring gear.

 

I hate to belabor the point, but how sure are you the teeth are grinding? A slipping starter clutch makes a horrible grinding noise, and almost always intermittently works when it fails. Fast forward to 2:20, and don't ever test by locking up a starter like this.   He's lucky it didn't throw that prybar and break his jaw or something. It might have if it engaged. Also, that's not a Bendix in the video.

 

 

[Andy J]

I just thought of something that helped me in the past. The solenoid on my Delco starter('51 Pontiac 8 cylinder) has elongated holes on the mounting  bracket. Loosen the mounting bolts and slide the solenoid as far away from the drive end as you can and retighten the bolts. This causes the solenoid plunger to extend the Bendix a fraction further onto the ring gear.This has worked for me in the past.  YMMV

[pmhowe]

I have had this problem happen to me twice. The first time, the mounting bolts were loose. You said that is not the problem you face, so rule that out. The second incident was due to lack of lubrication of the Bendix. I think that might be your problem, as you said it has just begun. Worth a try to lubricate the Bendix and see if things improve. Good luck.

Phil

[Bloo]

Now I'm confused. Does this 48 DeSoto have a Bendix? I assumed not because @marcapra said his solenoid was mounted on top of the starter. I guess that's possible if there is no linkage and the "solenoid" is just a relay like the one on a Ford, and happens by coincidence to be mounted on the starter.

 

It *cant* have a mechanical linkage from the drive gear up to the solenoid *and* have a Bendix. The two are mutually exclusive. They would fight each other. Which is it? Can we see pictures?

 

Edited May 6, 2023 by Bloo (see edit history)

[marcapra]

You can see my starter on page 25 of this MTSC booklet.  I don't have a Bendix.  They were on the cheaper cars, Dodge and Plymouth.  I have the Positive Drive (Solenoid type)  Autolite starter.  I definitely have clashing of the gears.  I took off the starter and have beat up teeth in three locations on the ring gear at about 12, 4, and 8 o'clock.  I have just finished deburring the teeth with a dremel tool.  I don't know if this is good, but I wiped a thin layer of grease on those deburred areas.  The booklet goes on the say that the pinion should clear the housing by 1/16".  I' m not sure what they mean by that.  Not sure what "clear the housing" means?  https://www.mymopar.com/downloads/mtsc/014.pdf

[billorn]

Someone on here had there starter rebuilt and they added a small surprisingly light spring on the bendix to help push it out of the way after it starts. I dont remember who it was but they said it fixed the problem.

[Bloo]

54 minutes ago, marcapra said:

The booklet goes on the say that the pinion should clear the housing by 1/16".  I' m not sure what they mean by that.  Not sure what "clear the housing" means?

I don't know how you adjust that for sure on Autolite, but on the nearly identical looking Delco's, that means the clearance between the tip of the pinion and the starter nose casting with the solenoid pulled all the way in. For instance, the measurement is 1/8" on some mid 30s Buicks.

 

If this is similar, there are threads on the inside of the solenoid plunger. You pretty much have to have the boot removed to check it. You pull the solenoid plunger all the way back, pulling on the outside edge, not the linkage. With it all the way back, measure from the tip of the pinion to the inside of the nose casting using a round gauge like a drill bit. Take out or disconnect the plunger to screw/unscrew the threads and recheck until it is correct, then reassemble with the boot.

 

EDIT: From the book link you posted, yes it is the same.

 

P.S. Not to keep harping on this but.... I'm still harping on this. The starter motor shouldn't be able to come on until the gear is engaged. The switch is at the back of the solenoid. How can it grind? I guess it could, especially if that linkage described above was way out of whack. I am still extremely suspicious of your starter clutch.

 

EDIT 2: This is a Buick Illustration, but shows where the measurement is taken. Look at the lower left.

 

 

 

Edited May 6, 2023 by Bloo (see edit history)

[marcapra]

After deburring my damaged teeth, I installed the starter to see if that helped.  It didn't.  It worked OK a couple of times, but clashed most of the time.  And I did notice that every time it clashed, there was an explosion of sparks coming out of the remote switch connection on the upper left terminal as shown in my last pic above.  Also, I should mention that the specialist who just rebuilt my solenoid switch reverse wired the two top terminals of the solenoid switch.  The hot connection is supposed to be on the upper right and now it's on the upper left in the below pic.  That's why I had to take my starter back to the shop a couple months ago to say the starter didn't work and the remote start wire got hot.  They said I was connecting to ground.  I said but I always connected to the upper right terminal.  So they put it on a table and bench tested it for me to prove it worked, and I thought boy, my memory is getting bad.  Do you think with all those sparks coming out of the remote start terminal, there could be a short there?  

[Bloo]

I'd set that clearance. I don't think the starter motor should be coming on until the gear is engaged.

 

Where does the small black wire from the right small terminal connect?

 

I'd have to see inside that relay to make any really good guesses about whether having those wires reversed matters or not. As for the sparking, maybe it has hit the contacts befure the gear is fully engaged and sparks? Other possibilities are that a winding is burned out in the solenoid. It can spark like that but wouldn't be intermittent. The contacts might just be worn out too. Probably not a short though.

Edited May 7, 2023 by Bloo (see edit history)

[marcapra]

Thank you Bloo for the Buick diagram!  Because of it, I knew how to measure my gap.  My gap is about 1/4" instead of 1/16"!  Maybe that's the problem?  That black wire on the right connects to ground, a bolt that holds the solenoid on the starter.  I think I'm going to take the starter back to the shop and have them adjust the gap at least.  I guess I should check my ring gear with a dial gauge to see if there is no runout more than .020"

Edited May 7, 2023 by marcapra (see edit history)

[marcapra]

I did a bench test and it looks like the pinion comes out very fast and bounces off the end and comes back about 1/2 " or more and spins there.  I don't know if this is correct.  The manual says that during the bench test, use battery current to hold the pinion in the engaged position while adjusting the plunger stud linkage.  The strap connecting the solenoid to the remote starter button should be removed so the pinion doesn't spin.  Then you push the plunger into the engaged position by hand where it will remain in the proper position for making the pinion clearance adjustment.  The shift lever adjusting stud can now be adjusted so that there will be 1/16 inch clearance between the end of the pinion and starting motor drive housing.  I'm going to see now if I can do this adjustment myself.  

[billorn]

I found the link to the starter with a spring. He was having the same problem where the starter gear was bouncing back into the flywheel after it retracted. A spring seems to have fixed it.

 

 

[marcapra]

here is a video of the bench test I did to show the pinion in action.  Does it looks too far back to turn the ring gear?  I tried to do the pinion adjustment as i described above, but when I pushed the plunger in and the pinion moved out, it still started when it was all the way pushed out.  The directions from the shop manual said it wouldn't turn, but just stay engaged.  I don't know what the stud is or how to adjust it, so I'm going to take it back to my electric shop tomorrow.  

 

 

[Bloo]

That ain't right. In fact, it's so far off I am wondering if there is a problem with the linkage or something is incorrectly assembled? Maybe a fork (that moves the gear) has worn through?

 

If the spec is 1/16", then that is what you should see. I agree, it is more like 1/2" and cannot be right.

 

The adjustment is inside the solenoid plunger. Using the Buick diagram again (and partly ignoring the meaning of the legends....).

 

 

The threads are in-between the arrow that says "adj stud linkage" and the one that says "shift plunger".

 

The linkage must be disconnected at the arrow that says "shift lever" and then the piece labeled "adj stud plunger" can be screwed in or out of the "shift plunger" to change the length.

 

There is built in slop in the linkage, they have even drawn some of it right by where it says "shift lever" so to check the 1/16" dimension you want the "shift plunger" all the way back (but all the linkage running free!) or you wont get the right measurement. I mentioned holding the outer edge of the "shift plunger" all the way back with the boot removed. It sounds like they instead want you to engage it with electricity. That's even better but you have to remove the strap so the starter motor doesn't run. The strap is the one directly southwest of where it says "switch contacts".

 

Correctly adjusted, the gear should come all the way out (except for the last 1/16" so it doesn't hit). It should pull more or less all the way back in when released. I don't think I see enough movement there to accomplish that. I wonder if something could be worn out or misassembled where the fork (the part labeled "shift lever") intersects the overrunning clutch and drive gear assembly directly north of the first "R" in "overrunning clutch"?

 

 

Edited May 8, 2023 by Bloo (see edit history)

[marcapra]

I will print this out and tell it to the electric shop who said the starter was fine when they gave it back to me.  

[EmTee]

11 hours ago, Bloo said:

That ain't right. In fact, it's so far off I am wondering if there is a problem with the linkage or something is incorrectly assembled? Maybe a fork (that moves the gear) has worn through?

That's what I'm thinking also.  It sounds as if something in the linkage may be badly worn.  By the way, what is the condition of the armature bushing at the gear end of the shaft?

[dodge28]

I read the comments and found almost all are blaming the starter motor. The solenoid on the starter motor has 2 windings. 1 for pulling the gear and one for holding the gear in place . They are all like that. The problem lies in the ring gear. All engines , when turned off stops at 1 particular spot . Most flat head engines usually stops at the # 3 cylinder position. So every time the engine is started the nearest 5 teeth on the flywheel at that point in the ring gear gets beaten the most. NO WHERE ELSE ON THE RING GEAR. To prove it ,if that is a problem, turn the engine by the fan a 1/4 turn and then hit the starter. There will be no grinding of the teeth. AMC in the 60's were famous for premature worn teeth. Poor material.

Remove the starter motor and use a screw driver  and slowly turn the flywheel and look for worn teeth. The dentist will not help. The solution is to replace the ring gear. "YOU CANNOT FIND A RING GEAR ? Mark the position of the offending teeth in relation to the flywheel. Remove the existing ring gear by heating it with a torch. Just enough to get it loose and change the teeth position. Let the ring gear cool on its own.  NEVER OVERHEAT RING GEAR.

[marcapra]

Yes, the ring gear is worn in three spots, but isn't it clear that the pinion is not coming out far enough?  

[dodge28]

Previous owners must have done the same I sugested . It was quite a common practiced  in the early days for economic reasons. In the absence of a torch home mechanics heated the new ring gear in a bon fire. NEVER HAMMER OUT A COLD RING GEAR . BY DOING SO THE TINY RIDGE ON THE FLYWHEEL GETS DAMAGED ,HACK IT OFF 

[dodge28]

To check the starter gear do a bench test. Hold the starter in a vice, put voltage to the solenoid and look at the gear. The solenoid  will pull the gear to the nose cone with a heavy "clack". 

[marcapra]

Did you see my bench test shown above?  It's supposed to come out almost to the end of the shaft with a 1/16" gap.  It's more like 1/2".  

 

I took the starter to the electric shop today.  I tried to explain the problem with the pinion gear not coming out far enough, but the counterman couldn't be bothered with that.  I gave him the papers from the shop manual showing the directions for adjusting the pinion to the 1/16th spec, but he said he didn't need that.  Makes you wonder if it's going to be done right or not.  

Edited May 9, 2023 by marcapra (see edit history)

[Oldtech]

There is definitely a problem with the engaging fork or the drive. I hope the electric shop guy checks it out, but that's why I like doing it myself. I used to think I could screw up a few things and still be ahead. 

I suspect the fork business is broken or the pins aren't engaging the drive or ...  ? It definitely should hold the gear out to the 1/16 spec. 

[marcapra]

I wonder if I could have damaged something in the starter when I used it to break the crankshaft jaw when I had to remove the hub so I could fix a timing cover oil leak?  I put a socket on the jaw and a cheater bar, but i couldn't break the jaw because I was just turning the engine over.  So I put the cheater bar in part of the front frame and cranked the starter to break the jaw and unscrew it.  The starter balked a couple of times and then broke out the jaw, or crankshaft bolt.  

[dodge28]

On application the pinion must never hit the nose cone. There must be a gap, There could be a problem with the solenoid itself. The floating terminal could be worn. Loose the nut about 7 threads , push the terminal back and turn 180 degrees and re tighten. It is a common trick. In the old days repair kits were available. That terminal can be repaired using silver solder to build up the worn part. I done it many times. I have never seen forks on small starters made adjustable. On large starters the forks are adjustable.  

[Bloo]

9 hours ago, marcapra said:

I wonder if I could have damaged something in the starter when I used it to break the crankshaft jaw when I had to remove the hub so I could fix a timing cover oil leak?  I put a socket on the jaw and a cheater bar, but i couldn't break the jaw because I was just turning the engine over.  So I put the cheater bar in part of the front frame and cranked the starter to break the jaw and unscrew it.  The starter balked a couple of times and then broke out the jaw, or crankshaft bolt.  

On more modern cars that is a fairly common technique. I've never heard of anyone breaking a starter that way.

[marcapra]

7 hours ago, dodge28 said:

On application the pinion must never hit the nose cone. There must be a gap, There could be a problem with the solenoid itself. The floating terminal could be worn. Loose the nut about 7 threads , push the terminal back and turn 180 degrees and re tighten. It is a common trick. In the old days repair kits were available. That terminal can be repaired using silver solder to build up the worn part. I done it many times. I have never seen forks on small starters made adjustable. On large starters the forks are adjustable.  

I should mention that my solenoid switch was rebuilt by a specialist in Maine a couple of months ago.  

[dodge28]

marcapra, now that the solenoid is good , I am almost sure the problem in the ring gear. Can you find a new one ? A good machinist can build those damaged teeth back to like new.

Hobbies are not without problems. This is a costly one. I AM SUBJECT TO CORRECTION. 

[marcapra]

I have new ring gear on the way.  If it's the ring gear, how do you explain the pinion gear not coming out far enough?  

[c49er]

Those ring gear welds have to be carefully be cut to remove the ring gear.

Also the ring gear needs to be carefully positioned and re-welded.

Welds equal to maintain FD balance.

Be sure you need a ring gear to solve your problem.

[dodge28]

marrkapra, There is no way to tell how far the pinion goes in because once the starter is installed you cannot see inside the nose cone. The only thing you can do , before  installation  is to paint the pinion gear with some kind of coloring, remove the feed wire from the starter to the solenoid and use a jumper feed the solenoid. Let the solenoid push the pinion a few times in and out, remove the starter and look to see the impression the coloring will leave on the pinion gear. Also look to see how far back the solenoid is pulling back. It is a painful exercise, not foolproof but at least you will get close.

[Bloo]

That might work, but the insufficient engagement problem has already been well pretty documented in this thread. The measurement from the gear to the nose casting is 1/16" as documented in the PDF of Chrysler's own service documentation that @marcapra linked in the 14th post of the thread. Then in post 22, he posted a video showing that the gear is more like 1/2" back when engaged. That is approximately 7/16" less engaged than it should be. How thick is a ring gear?