1935 Buick starter solenoid troubles

[cascadia]

Hi all, so I have an interesting dilemma. The '35 Buick that I just got has a mysterious way of trying to start on its own!  Unsettling to say the least. Thankfully I was close by when it happened, the ignition was off, and it was out of gear.

The accelerator switch is adjusted properly, in fact, I disconnected the electrical lead from it completely, thinking that was the obvious suspect. Still tries to start! I put a quick disconnect on the battery until I get this figured out. 

 

Disconnecting the ground terminal (marked 'solenoid') on the voltage regulator will stop it from doing this, but then of course the car won't start when you want it to either. 

I've removed the relay cover and can't see any signs of obvious damage or burned areas. All of the car's wiring is original, and doesn't appear to have ever been tampered with.

The interesting thing is that when I put a test light on the 2 terminals, both of them had power to them, whether the ignition was on or off. I'm thinking that there must be some internal short? Electrical diagnostics are not my strong suit.  I removed it from the car for the photo and further testing.

[Jim Nelson]

I have a 35-58 Buick.   First :  Do you have a factory electrical diagram ?   Second :  do you have a 40 series or a 50, or 60 series.   There is two very different schematics.   The 40 series has a layout that is  what I call a normal layout.   Starter on the pax side,  distributer mid way on the pax side,  the generator is mounted on the drivers side like 1936 and newer layouts.    If you have a 50 or later series,  you have the starter mounted on the drivers side,  the generator, distributer and water pump all on the pax side.   Thus, wiring was changed accordingly.    Studying the differences will show you that the starter solenoid wiring with the vacuum start switch on the 40 series.   Same with the 50 or 60 series wiring.    If you are not going to 'show' the car where examining it for any alterative changes,    by passing the vacuum switch is very important.   By putting a hidden push starter button,  it will keep you from ghost operation.     This switch is in series of the starting circuit,   with this mod.    the starter will NOT function unless you push the button.    I have a '38' where my starter would engage while the engine was running (accelerating from a stop).     It ruined my starter ring gear after a few errant starts and it caused me to do an expensive replacement.    Check your service manual for the method to remove the flywheel.    (On the 40 and 50 series engines the flywheel removal is a bitch.    The big engine (322) lets you remove the flywheel so much easier - - )    The ring gear  on the small engine is fairly complicated and a replacement gear cost me +$300 plus cost a partial tear down.     The ring gear design is for '35' thru '38'.   (hard to find now,  call me for my supplier).   Then they changed it to a design that was fixed until 1953.     ( I ended pulling the engine and did a  complete rebuild  -  damn vacuum start system ).    I drive my Buicks and dependability is very important.    I have eliminated the details to getting there.    A very expensive fix....

[cascadia]

Yes, thanks, I have the shop manual (it's a 50 series with the later engine) and have studied it closely. The vacuum system is not at fault, I've determined that.  I'd rather not make any modifications to the factory design, but am more interested in finding out the reason for the fault, and bringing the car back to its normal operating condition.

[Oldtech]

If there is voltage on the starter side connection the solenoid part itself is sticking or something is melted. I would remove the contact cover and examine it. There is a big washer that contacts the 2 terminals when the plunger pulls in.  If possible remove the plunger and spring. It may just be mechanically sticking. I don't know how much you can get apart on that type of solenoid but it looks from the picture like it's fixable. 

There are settings for the relay air gap also. Contact point opening is .030 to .045. The gap between the armature and coil core should be .010 to .014.

Lastly. Simple things first: Is the spring on the arm on the starter pulling the arm back properly?

Edited April 14, 2020 by Oldtech (see edit history)

[Bloo]

You have a terminal marked solenoid?

 

What year is the engine, more to the point what year is the starter?

 

What year is the voltage regulator, and how many terminals?

 

On a 37 for instance, any series, power is provided to one of those terminals and ground to the other to make it crank. The cover can get bent and short one terminal, pay close attention when you assemble, but you would still have to have power on the other terminal to crank.

 

The 37 vacuum/throttle switch provides the power, and the 5-terminal regulator provides a ground. The regulator terminal that does this is marked "GND". It is not really the regulator ground, it is a contact that is grounded only when the engine is not running (system not charging).

 

For the engine to crank, the charging system has to be not charging (so the regulator provides ground) and there has to be no vacuum (so the vacuum/throttle switch provides power).

 

When you say both terminals were hot, do you mean the little ones? If there were no wires connected to them, I'm pretty sure they should have been dead.

 

I agree with Oldtech. The solenoids are rebuildable. I have done it. The only possible showstopper is an open winding.

 

 

 

 

 

 

Edited April 14, 2020 by Bloo (see edit history)

[cascadia]

It's a 5 terminal VR, and the engine, starter, and all parts are stock 1935.

The terminals on the solenoid that tested hot were the smaller ones on top. They tested hot with no wires connected to them.

The starter arm pulls back properly, and the plunger on the solenoid itself is working. The starting mechanism all works freely, it's the starting on its own part that's the problem. 😄

 

 

 

[Bloo]

I have not seen one like that. Thanks for posting it! I gather it must be a simple cutout, combined with the horn relay?

 

Anyway, it appears that automatic starting works the same way a 1937 does, and that the relay terminals should not be hot. I believe the core that the winding is wound on IS hot, or at least the frame of the relay is. I would remove it and look it over really close and see if you can find a short external to the winding that can be cleared.

 

I have set up those relays using the information in the 1937 shop manual. 1935 may have been a little different, however the 1937 manual says:

 

Quote

Note: the solenoid relay calibration differs from that used on previous models. Past relays are not to be used on 1937 cars. However it is recommended that the 1937 relay be used to service past models.

 

Calibration specs from 1937 manual:

 

Cut-in voltage ................................. 1.9 max

Opening voltage.............................. 1.0 - 1.2

Point opening ......................... .025" to .045"

Air gap (points closed) ......... .010" to .013"

 

Generic specs from a Chilton manual covering 1938-1947:

 

Cut-in voltage ................................. 3.2 - 3.6

Opening voltage.......................... 2.0 or less

Point opening ......................... .025" to .030"

Air gap (points closed) ......... .007" to .009"

 

But that only matters if you can clear the short. Air gap is the gap between the core and the moving piece with the points closed. Cut in and cut out are set by the spring tension (there is a bendable tab putting tension against the spring IIRC). They both set with the same spring, but the relationship between the two should be close enough if the air gap and points gap are right.

 

Edited April 15, 2020 by Bloo (see edit history)

[cascadia]

Thanks for the information.

I think I figured out what was wrong: The big terminal that the battery cable attaches to was shorting to the case internally. I added an insulation washer and it solved the problem. But now the plunger doesn't want to move. With 6V power to the large and small positive terminals, and ground to the case and small terminal (on the top left in my photo), the relay will click, but the plunger doesn't move. I removed the plunger and cleaned it up with 00000 steel wool, and adjusted the point gap and air gap to my shop manuals' specifications. Any ideas from here?  The frustrating part is that it was definitely working before, but turning on by itself. Manually pressing the plunger with power on does make the opposite terminal (the one that goes to the starter) read hot.

 

Here's the 6V tester I was using, maybe it doesn't have enough amps to get that plunger moving.  I'm going to reinstall the solenoid on the car and I'll report back.....

 

Edited April 14, 2020 by cascadia (see edit history)

[cascadia]

Back on the car, and unfortunately, no difference. 😑

What can be causing the plunger not to move?

[Bloo]

Is the starter motor connected to the other side of the solenoid?

 

The frame of the little relay should be battery hot if I remember correctly. It's connected to the battery cable. The moving point is connected to the frame and is also hot. When the moving point touches the other point, it energizes the solenoid.

 

I don't think that 2.5 amp tester will do much with the starter assembled. You would have to use a car battery.

 

The solenoid itself has 2 coils inside, a "pull in coil" and a "hold in coil". Both coils have one end connected to the non-moving point on the relay. The pull in coil (the big one) is connected to the hot side of starter motor windings for a ground! (more on that in a minute). The hold in coil (the smaller one) is simply grounded at the far end, through the solenoid case.

 

When the little relay snaps down, it energizes the pull in coil and the starter motor, connected in series. (It also energizes the hold in coil because the hold in coil has it's own ground).

 

The pull in coil pulls in the solenoid engaging the gear, and knocking the copper washer in the back of the solenoid against the two large copper bolts. In doing so it shorts out the pull-in coil (remember it was connected between the non-moving relay point and the starter windings).

 

With the copper washer knocked back against the big copper bolts, the starter motor is "on", cranking, and the hold in coil (still having it's own ground), holds the starter engaged until the little relay opens and removes power from the hold in coil. The the solenoid disengages.

 

Back to your little relay, it sounds as if it has an internal short between the frame (which should be hot), and the coil (which should float, and be connected only to the 2 little terminals). It is probably defective. The coil is made of magnet wire and wrapped around fish paper or something similar, and should never touch the frame, Tiny wires should be coming out of the relay coil and connecting to the two little terminals. If you cannot find some place the wire has become bare and touched something it shouldn't, you will probably have to replace the relay.

[cascadia]

Thanks for the help Bloo. When I put it back on the car I made all the usual connections, including the starter motor.

I took a closer look at the magnet wire. I can see one (they're tiny) connected to the small ground terminal up top. I can't see a 2nd one, but I do see a small portion of an insulated wire coming out of the center back of the frame before it disappears back into it.  Maybe that's the one that goes to the positive terminal up top? Either that or the positive wire's snapped off, although I don't see any remnants or broken leads.

I have power to the moveable part of the relay, and it does transfer over to the fixed part of the frame when it closes. It definitely snaps closed when power's applied to the positive terminal up top. Just no action from the coils. 

But one more clue could be that when I apply power to the terminal up top to close the relay, I noticed with my test light that whole case of the solenoid assembly is energized.  That can't be right.

[Bloo]

Only the relay frame (and the movable point) should be hot The solenoid case needs to be grounded to the starter case (gets its battery ground from the starter case). It sounds like the solenoid case isn't grounded.

 

Could paint be preventing the solenoid case from grounding?

 

 

 

[cascadia]

Good thinking. I reinstalled it on the car, and added a ground strap for good measure. No go.

I think it must be some internal that I can't get at or see. Maybe just the relay like you said.  I'll take it in for service and let you all know how it ended up.

Thanks for everyone's comments.

[cascadia]

Hi everybody, I took the solenoid into a service shop and he was able to get it working again. There was an internal short. I took it to an old-time repair shop here in Portland named Philbin Manufacturing, he does a lot of distributor work for flathead guys. Thank goodness that there's still folks like that out there, guys with that type of experience. They don't deal with computers, they just put it on the bench and start testing it!

😊 Grateful.