New headlight problem

[Barney Eaton]

Since the first of the year, I have seen this problem 3 times.  Before that I only remember one.  Not sure why it is happening but here is the problem.

For reference the first picture shows the big white gear and the motor output shaft.  The "rollers" are the normal problem, they crack and then get ground into powder, their purpose in life is to connect the inside of the white gear to the shaft.

Picture 2 shows the output shaft and the grey casting that is pressed on the shaft.  Note the shaft has a knurl on one end (the part pictured has the threads broken off the right end)

The knurl locks the shaft to the casting when the shaft is press into the casting.

The new problem, the shafts are coming loose....either the knurl is not as course as it should be or there is just too much torque.

The motor will run, the gear will turn but the gear is spinning on the shaft so there is no output to the bellcrank.

There could be several ways of repairing this but I chose to drill a hole (1/8 in) as shown in the third picture.  The hole goes thru the casting, thru the shaft and out the other side of the casting.

I have 1/8 X 3/4 steel roll pins and this to me seemed like the simplest and most professional looking solution.  

[DS_Porter 89 Burgandy]

Looks like a simple and good solution. Early in my Reatta ownership days (1998 I bought my first one) I used this method to pin the output shaft to the bellcrank. That was before I discovered this forum and the availability of improved bellcranks. Thanx for all your work in this area

[KDirk]

 

I suspect it is a matter of age and fatigue of the casting from so many mechanical cycles rounding out around the knurlings on the shaft. I suppose this will be yet a other problem we need to deal with as these cars advance in age. Someone who has machining skills and gear might consider offering casting/shaft upgrades as you've engineered here on an exchange basis. Might be a few dollars to be made, as many will not have the skills or tools to roll their own.

[2seater]

Looks good to me. Interesting to see apparently there are two designs, one with a flat area and one with all rounded?

[Barney Eaton]

Good catch on the different castings.......don't know that one is better than the other.

Also note the output shaft.....one has a undercut at the base of the flats.......the other does not. That is probably two different vendors ....different process

 

7/19 - also note that the shaft with the with the undercut has a hole drilled in the flat area and the threaded end tip is missing.   I suspect the hole was drilled to pin the

bellcrank to the shaft,  unfortunately the hole was misplaced and too close to the base of the threads and at some point the threaded shaft broke off because the

drilled hole had weakened it at that point. 

Edited July 24, 2019 by Barney Eaton (see edit history)

[Ronnie]

The roll pin is a good fix but not everyone will have the tools and skills to do that. I would think if you cleaned the knurled part of the shaft and gear really well, then applied JB Weld before pressing them back together that it would last for many years...

[89RedDarkGrey]

1 hour ago, Ronnie said:

I would think if you cleaned the knurled part of the shaft and gear really well, then applied JB Weld

 

Or- a good mechanical and chemical cleaning (emery cloth & acetone) then flux both, assemble, then heat and sweat solder together. There are a few ways to remedy this. I'm surprised at the poor engineering of such a critical and costly safety system. Cast aluminum isn't something you normally rely on for torsion or clamping.

 

Most likely the steel shaft was insert molded to the cast aluminum, to eliminate any damage from having to assemble by pressing it in.

[Ronnie]

6 minutes ago, 89RedDarkGrey said:

Or- a good mechanical and chemical cleaning (emery cloth & acetone) then flux both, assemble, then heat and sweat solder together.

 

If a person had the skills to do all that they could probably drill a hole and drive in a roll pin...

[harry yarnell]

Sweat solder aluminum?

[89RedDarkGrey]

12 minutes ago, harry yarnell said:

Sweat solder aluminum?

 

Yes, easier than measuring, vise, good drill with bits, drilling two separate pieces- or together, hoping the bit doesn't snap off, find a roll or regular pin, and punch (all supposedly very easy for everyone to do)

 

Or- use a propane or MAP gas torch and THIS that you could have around for other stuff, too. <5 minutes later, done.

[Ronnie]

Are you serious about using a propane torch to solder a piece of aluminum to a steel rod and expect it to hold? And are you expecting someone to do this who has problems drilling a hole? I'm not being argumentative. I'm just wanting to know if that is what you are recommending.

[2seater]

Just guessing from the color, the cast piece is some sort of pot metal but no evidence to back that up. There is more than one grade of aluminum, at least the hardness or ductility level, but I would guess this a lower grade casting. It doesn't seem to be the best design with a smaller clamping surface than it could be with the counter-bore on the lobed side, leading me to believe it was a press fit, not cast in place,.

[DS_Porter 89 Burgandy]

Gee you guys are making me feel proud of my drilling skills (without a drill press)

[Ronnie]

38 minutes ago, 2seater said:

It doesn't seem to be the best design with a smaller clamping surface than it could be with the counter-bore on the lobed side, leading me to believe it was a press fit, not cast in place,.

 

I think you are correct. When  you knurl a part it increases the diameter of the part in the knurled area. The shaft,including the knurled surface, was probably made with an automatic screw machine that spit those things out like popcorn. The parts could have been easily pressed together with a small hand operated arbor press since one of the parts was soft metal.

[Barney Eaton]

I feel sure the shaft is pressed into the cast part (which is probably zinc)

If the shaft were loaded into the molding machine, there would be evidence of a parting line on the cast part around the shaft and/or flash around the area where the shaft goes into the casting.

the cast part has the parting line  around the large diameter and you can see the ejection pin marks.

I am just posting this to let you know you should check these parts if you repair your headlight motors......feel free to repair them any way you like.

 

(revised to try and clarify the description...)

Edited June 20, 2017 by Barney Eaton (see edit history)

[89RedDarkGrey]

1 hour ago, Ronnie said:

Are you serious about using a propane torch to solder a piece of aluminum to a steel rod and expect it to hold? And are you expecting someone to do this who has problems drilling a hole? I'm not being argumentative. I'm just wanting to know if that is what you are recommending

 

Yes. Just as leadding cast iron pipe together, or the principal of Loctite thread locker- all it has to do is flow into the crevices and harden. The Brazing system mentioned easily does this. Muggy Weld has been around a while now, and has many videos of their different products. There is no need to "weld" the two parts together, and aluminum and steel aren't homogenous any way.

 

I'm not saying that I definitely know the shaft was insert molded- it just made sense to me, to not press a knurled steel shaft into a brittle cast aluminum part. The easiest way to tell- look at an unrepaired one; you should see tracks on one side where the knurls dragged into the hole as it was pressed in.

[Ronnie]

Thanks for pointing this out Barney.

[89RedDarkGrey]

Yes, thanks for pointing this out Barney.

[NCReatta]

I've seen this problem about 3 or 4 times in the over 350 motors I have rebuilt. To date my solution is to reoamce it with one that does not wiggle. I test them all to make sure they're tight before installing them back into a motor. 

 

I have also noticed a correlation between this part failing and the motor having been previously rebuilt with different designs of bellcranks on the outside. Every single one I've gotten back as a core with this issue has had some other form of bell crank installed. 

 

I believe the reason this happens is because at the end of the travel cycle, the end of said bellcranks do not hit squarely on the bracket giving them a little bit of wiggle room that the original was not designed to have. This ultimately leads to slop in the mechanism that will slowly eat away at the splines 

Edited June 20, 2017 by NCReatta (see edit history)

[89RedDarkGrey]

After doing some research (without examining an actual part for knurl cut marks) I am 95% certain the shaft was insert molded into the aluminum-zinc hub.

 

The 3 circles where the Delrin rollers sit are marks from ejector pins, used to push the part out of the mold. There is no "part line" in a poured mold, unlike a 2-piece clamped mold.

 

Look familiar?

   More on this is found at  http://www.alcastcompany.com/tubeinsertaluminumcastings.html

 

A fluted steel insert, in a low pressure 1-piece mold is quicker in production than pressing in the shaft. 1 production step is faster than 2.

[DS_Porter 89 Burgandy]

It is worth noting that at the end of the bellcrank travel it it stops against a rubber pad to soften the stop as the current ramps up and the electronic control shuts it off.

[Barney Eaton]

Had another motor fail because of the shaft turning in the cast part.    At this time it probably does not matter if the shaft was insert molded or pressed into the part.

J.B.Weld, Locktite, and other things might work but pinning the casting to the shaft is a permanent fix.    If the problem is from the sudden stopping when the bellcrank hits

the stop....you would think the plastic rollers would crack or absorb the shock before the shaft would break loose.   GM just did not think we would be driving these cars

30 years after they were made.

I will even do them for free if you pay the shipping.

Edited July 20, 2019 by Barney Eaton (see edit history)

[Barney Eaton]

I think I've discovered the reason for these failures.  

When removing or installing the nut on the end of the motor shaft,  to get it on or off,  owners are stripping the shaft out of the cast part.

The nut is a locking nut,  it takes some extra torque to install and remove.

Owners are relying on the internal parts of the motor to hold the shaft when they install/remove the nut. 

This action (torque) is stripping some of the shafts.

The solution......hold the bellcrank with one hand while tightening the nut.   This method allows the bellcrank to take all the torque.

In cases where you are replacing the bellcrank because it has failed,  the bellcrank will not hold the shaft and you need to check

the shaft to casting fit before reassembling the motor.     This can also be a problem when the owner is just replacing the bellcrank

and not opening the motor to replace the rollers....the shaft could be separated from the casting and the owner may not notice.

[padgett]

Makes sense. Did anyone try just a schmeer of JB Weld and assemble ? Did it work ?

[Barney Eaton]

Padgett.......... there is a lot of torque especially when the headlight hits the open/closed stop position.   in addition there is maybe 3/8 inch where they mate,   epoxy, glue etc might last  3-4 cycles.

When I run out of projects I will bond one and see how long it last.

[padgett]

Anymore I think in terms of fractals (what makes a 500 mph 1/4 mile possible) & considered forming a hard surface between the splines rather than a bond.