When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Did you get the fork out? I'm wondering if one of the spring finger tabs is messed up. That's about the only thing I can think of in the vicinity of the outer edge of that bent flange.
I discovered this morning that I need to take off the bell housing to get the fork out, another messy job (lying under the car) I am working on as we speak
here are some pictures of the fork &ball stud. My analysis is that the fork must have slipped over the ball (spring has very weak clamping pressure, fork very loose over ball stud), pushed agains the flange of the bearing and bending it. You guys agree? Other opinions ?
And is a strong(er) spring clamping force how it should be when new?
Looks like the rivet did bend the flange at the back and probably that pressure contributed to the galling at the front. Probably need an adjustable ball stud and move the fork forward in the opening. From your pictures and damage it looks like the fork was probably too far back in the bellhousing. Most forks I have had, have fit reasonably snuggly on the stud - at least enough to keep them seated on the stud. The TO bearing typically rides fairly far out on the fork fingers.
I think I would start over with a new throw out bearing, clutch fork and an adjustable stud. I like to adjust the ball stud so that the fork is fairly far forward in the bell housing. It shouldn't be right up against the front of the opening because you need room for the rubber boot and also as the clutch disc wears, the outer end of the fork will move forward. That should put the arm of the fork inside the bell housing leaning forward a just slightly so that as the pedal is stroked, it moves to perpendicular to the drive line and then a little over towards the back. That way the fork should not bottom on the rear of the bell housing or allow the spring finger rivet to hit the TO bearing.
I'm not sure how tightly the fingers (fork and spring tabs) should fit in the width of the groove. The last time I did my clutch the fingers left about 1/8 inch play in the TO bearing groove. I bent the spring fingers so that there was just the slightest play left inside the bearing groove. Could be wrong, could be right, may not matter.
after reading all comments made, and compare that with my findings on the car my conclusion on the root cause of the problem is a way too weak ball stud spring. No way the fork could be held in place reliably.
Eventually the ball stud height needs tweaking, although I use all standard GM components from pressure plate to clutch to bell housing etc. So it should work as set up.
Will check anyway, found an excellent explanation about how to find perfect fork pivot ball height here
after reading all comments made, and compare that with my findings on the car my conclusion on the root cause of the problem is a way too weak ball stud spring. No way the fork could be held in place reliably.
Eventually the ball stud height needs tweaking, although I use all standard GM components from pressure plate to clutch to bell housing etc. So it should work as set up.
Will check anyway, found an excellent explanation about how to find perfect fork pivot ball height here
GM made at least two different dimension ball studs. And even though you may use all GM parts, clutch assemblies and TO bearings that are allegedly OEM equipment change over 50 years. An adjustable ball stud is good insurance towards getting the fork geometry dialed into an acceptable state. You are apparently right that the spring wasn't securing the ball stud in the fork pocket or the fork could not have otherwise moved inward to where the TO bearing was deep inside the fork and where the rivet could come in contact with the TO bearing flange. I would think that fork moving around would play havoc with clutch release and pedal action.
Last edited by DansYellow66; 11-11-2018 at 07:30 AM.
GM made at least two different dimension ball studs. And even though you may use all GM parts, clutch assemblies and TO bearings that are allegedly OEM equipment change over 50 years. An adjustable ball stud is good insurance towards getting the fork geometry dialed into an acceptable state. You are apparently right that the spring wasn't securing the ball stud in the fork pocket or the fork could not have otherwise moved inward to where the TO bearing was deep inside the fork and where the rivet could come in contact with the TO bearing flange. I would think that fork moving around would play havoc with clutch release and pedal action.
I have done the ball stud height measurements as specified in the document I referred to in my previous post, and all is according to spec. Now, I may as just wel add an adjustable ball stud when ordering the fork.
I think that your fork popped off of the ball stud and moved inward into the bell housing. That is the ONLY way the rivet could have hit and bent the rear flange on the bearing. When positioned properly, the inner ends of the fork should be centered on the bearing. The picture you took (shown below)
is probably the positioning your fork had when the damage took place. The fork, when properly positioned should be about a half inch further upward in that picture, so the ends of the fork are centered on the circumference of the bearing. I think if you get a new fork & bearing and make sure the fork is secure on the stud, you will be OK. After you put it all back together, just check to see that when the clutch pedal is in the full up position, that there is some room between the rear flange of the throw out bearing and the front face of the transmission front bearing retainer. The throw out bearing rear flange should never touch the front transmission bearing retainer---If it does, then you risk the possibility of the fork being pulled off of the stud.
GM made at least two different dimension ball studs.
Yes, and here's the story - design/drafting error that caused the mistake and the Engineering action (taken a year after the problem was solved) to reverse the mistake:
Yes, and here's the story - design/drafting error that caused the mistake and the Engineering action (taken a year after the problem was solved) to reverse the mistake:
great reading, the document says correct height should be 1 1/2", mine is exactly that. Correct data is key for remedying problems. The culprit clutch fork is the one that came with the car when I bought it. Should have replaced that too considering I have replaced about every critical part in the car anyway
Glad to read from you John once in a while, my car incorporates many of your invaluable insights. Keep going, at the pace that suits you best!
As I mentioned above the GM fork ball would be great to always rely on if diaphragm clutches were manufactured today to the exact same specifications as 50 years ago. I don't remember ever seeing a dimension on where the stock diaphragm clutch fingers should rest on a clutch bolted down to the pressure plate with a new disc. If you can find that dimension somewhere and compare it to the installed finger height of a modern replacement clutch, then you could be assured of meeting the geometry that GM intended with a GM fork and GM ball stud Without that information, that's why I prefer to install an adjustable ball and verify the geometry myself before buttoning it up. But it does look like the main issue here was the fork moving in and out on the ball stud.
I tried to understand how the pedal force translates to fork force. If my analysis is correct, by geometry of the fork there will always be a force (green arrow) that tries to push the fork towards the center of the clutch. (the red pedal force decomposes in a blue and green force). So if the fork ball retainer spring is loose, it ends up where mine ended up. Don't think that changing stud height will be able to balance that out, but it does help to avoid the problem as described in John's document.
Last edited by alexandervdr; 11-14-2018 at 10:19 AM.
No, the height of the ball stud would have no significant effect on the component of force applied that pushes inward on the fork. But the height of the ball stud can affect if the fork bottoms on the rear of the BH opening before it releases, and to the geometry of the bell crank and upper rod adjustment (or possible lack thereof) since the lower rod is not adjustable to compensate. I like to have the part of the fork in the bell housing leaning forward at the outer end slightly so that when it's depressed it moves over center in the arc of movement as the TO bearing moves forward. If the movement is balanced from under center to over center - that results in marginally less pedal stroke needed to disengage the clutch. If the movement is all over-center then that translates to slightly less TO bearing movement for the same degree of fork rotation around the ball. Of course it gets even more complicated at the pedal due to the angles of the arms on the bell crank and where they are at rest. If you can diagram out the entire clutch system from the pedal to the fork, taking into account pedal arm rotation, bell crank arm rotation and vectors, and fork rotation and vectors - we could really come up with something.
I don't believe the clutch fork is the problem IF it is the correct fork for the other clutch parts installed. The damage on the fork did happen on one drive. It has had an issue for a while. I still think it Is the result of incorrect geometry, not the fork itself. Is it the correct flywheel? Has the flywheel been over-machined moving the pressure plate mount surface too far forward? Are the pressure plate and disc correctly manufactured? Does your pressure plate have raised curved fingers or flat where the release bearing contacts it? Can you post pics of your other clutch parts? It may help.
There we go - that's something to behold. Your chart actually explains something about the z-bar design that I've not understood before. I always thought the upper arm should tilt slightly back towards the firewall so that it moves through vertical to forward past vertical. In other words I thought the upper and lower arm should be 180 degrees apart with the upper arm leaning towards the firewall with the pedal released. But from your diagram I suspect they set the angle and position of the upper arm to maintain a uniform angle of movement of the upper rod through the firewall penetration as the pedal is stroked so that it doesn't rotate or move up and down and hang up on the firewall hole or wear the boot out.
Last edited by DansYellow66; 11-14-2018 at 07:02 PM.
11-10-2018, 07:33 AM
