Yesterday I started up the car and heard an odd scrapping noise after I pushed the clutch in and engaged first without releasing the clutch. I drove the car off in first gear and then put it back in neutral when I hit a stop sign. After that I couldn't get it into any gears with the clutch pushed down. I started in first to get out of traffic, put it back in neutral and was able to get 2-5, but not 1st or reverse. The next time I started it with the clutch pedal pushed down the car shuddered as if the clutch was trying to engage. I let out the clutch, drove a few feet, pushed the pedal back down and it did not shudder.

I adjusted the clutch pedal to remove the slack and nothing changed.

Today I dropped the car off at my mechanic's place. He had someone push the clutch pedal down and looked at the clutch through the service window. With the clutch engaged the disk does not make contact with the flywheel. My mechanic removed the transmission and inspected the clutch fork and it does not appear to be bent or broken. The brass snap ring also appears to be in good shape.

Does anybody know what is wrong with my car?

 

Still sounds like the clutch fork. I would have them check again to be CERTAIN. Even when broken you still have to look closely to see the crack. They will still operate somewhat when broken but they don't have as much travel as they should.

There's a pic in this thread that shows where all of them break. While still installed though all you will make out is a crack where the pin is. A broken one can be easily missed if not studied close enough.

broken clutch fork

 

how about a master/slave or clutch line problem/leak. i would definitely check the clutch line near the ends. that's where they start to leak when they go bad and can cause problems similar to what you're experiencing.

 

sounds like me when my clutch fork broke too. its only $20 so replace it anyway while its apart. good luck.

 

Thanks for the input guys. I'm also not ready to give up on the clutch fork. I spoke to the mechanic again tonight and this time he said he did NOT remove the transmission to inspect the fork. My question is can the fork be adequately examined by looking through the access windows as he did?

I found a couple pictures of the view through the access windows in another thread posted by Section8. Dust aside, it looks like it's pretty easy to see.


Tomorrow I'm going to go over to his shop and inspect everything myself.

Re the hydraulic system, there are no visible leaks, bleeding the system had no impact on the problem, and the pedal does not stick to the floor. Is there anything else I should check before ruling this factor out?

 

i think that covers any potential hydraulic problems. good luck tomorrow.

 

It could be a loose slave cylinder, too. If the bolts are loose, it might "grab" the bolt close enough sometimes and not close enough (to disengage the clutch) other times. Check the bolts that hold the slave cylinder to the bell housing to make sure they are tight. This happened to me once.

-Max

 

If this problem occurred suddenly, then I would have to agree with the clutch fork diagnosis. My FD did the same thing a few years ago, and I was just able to nurse it home. A runner up might be a failed pilot bushing whereas the needles get lodged in between the crank and the input shaft. Sometimes you can clear the latter problem (I ran for a long time with NO pilot bearing), but a clutch fork will require replacement.

Mazda screwed up in the design of the fork. For some reason, they felt compelled to put a roll pin (that isn't really necessary IMO) through the fork into the pivot shaft in a highly stressed area (in a casting, no less). That roll pin (where the crack will start and progress from) and one possible fracture point is visible in the above pics. Depressing the clutch should cause any crack to open since it is under tensile load. If it has cracked on top, you may not be able to see it (the crack), but the fork may not exhibit even pull on the fingers.

In my case, I repaired the part by TIG welding it using phosphor-bronze rod (not NI rod, as it wasn't compatible with the base material) AND peening it while it was still hot in an iterative process. The reason I choose to repair was so that I could reinforce the piece and never have another failure of this part. So far, it has held up. If you repair or reinforce this part be sure to: 1) Not overheat it, 2) that the bushing is re-charged with oil, and 3) check and true up the fork fingers if necessary (disk grinder--lightly--as the ends are hardened).

Hope this helps.

 

I just got back from my mechanic's shop. I got under the car and looked at the clutch fork through the window in the bell housing. No cracks were visible. When he depressed the clutch pedal the disk pulled away and the arm did not appear to flex. The slave cylinder bolts seem tight and the piston appears to be in the cup of the clutch arm.

The mechanic noticed that the hydraulic fluid was very dirty (my fault) so he replaced it with new fluid by sucking the old stuff out of the reservoir, topping it off with new fluid and bleeding the fluid in the line out through the port on the slave cylinder. Now when I press the clutch pedal it goes to the floor without any resistance until the last 1/8th of travel and stays there. If I pump the pedal I get resistance for the last 1/3rd of travel.

I thought there might be air in the system so I bleed it with him by pressing the pedal to the floor, opening the bleed on the slave letting fluid out, closing it, raising the pedal with my hand, and then doing the same thing over. We tried this about 6 times, saw no air bubbles, and felt no change in the pedal.

The mechanic believes the master cylinder is bad and the new clean fluid made the problem worse, because the clean fluid can get past the piston seal easier than the dirty fluid. Frankly, that sounds far fetched to me. I don't see how new fluid could have that much of an impact on the problem. As I said at the beginning of the thread, the pedal felt normal when the problem started, aside from the fact that the engagement point has always been a little low. There is no fluid leaking out of the master into the pas compartment or on the motor side. As far as I can tell the slave is also dry on the outside. Is it possible for the master cylinder to be bad without leaking fluid outside the system? If so, where is the fluid going?

 

so you cannot get into any gear even with the car off? yes the older fluid in a weak master will make more pressure than new due to the viscosity. same thing in old neglected automatic transmissions if it changed they have a tendency to start slipping. shift fork?

 

If you are certain that the clutch was fully releasing (before you messed with the hydraulics), that would tend to rule out the fork or the slave cylinder assy. The only other connection (FW to input shaft) is now via the PILOT bearing. When mine failed, it made a scraping and ticking sound. I eventually "cleared" it, by forcefully putting in and out of gear while it was running and the broken pieces that were wedged in there fell out.

Another condition worth mentioning is clutch disk failure. This is where the clutch lining (face) has torn loose and rolled up between the disk and the FW or pressure plate. This could prevent it from fully disengaging as well.

In any event, and based on your descriptions, I don't think that the slave or hydraulics had anything to do with this problem, and so that was just an exercise. Unfortunately, you will have to repair it to get back to where you where--not to mention that it has added another variable to the equation.

 

I agree. A problem with the hydraulic system does not explain the scrapping noise I heard right before the problem first started. On the other hand a pilot bearing problem doesn't explain why the pedal sinks to the floor after the fluid was changed.

My hypothesis at this point is that a damaged pilot bearing is why I couldn't get it into gear, and the hydraulic system may either just have air in it or was already weak and changing the fluid pushed it over the edge.

Below is a post by Howard Coleman on the function of the pilot bearing that I think is worth reading for anybody that is not familiar with how it operates.

“in order to go from neutral (clutch pedal in) to a drive gear the trans must disengage from the motor. the pilot bearing in the crank, er eccentric shaft, has roller bearings so that when you push the clutch in even though the crank is turning as the motor runs, the input shaft and with it the clutch disc stops turning allowing you to change gears. when the pilot bearing goes bad the input shaft keeps turning and you have to crunch to get into gear.
if your slave and clutch cylinders are good it is your pilot bearing.
howard coleman”

 

"Now when I press the clutch pedal it goes to the floor without any resistance until the last 1/8th of travel and stays there. If I pump the pedal I get resistance for the last 1/3rd of travel."

Mine felt the same way when the fork went. Before this problem, did you by any chance get the impression that your clutch pedal felt like it was feeling more effort when pushing down? Mine did a little bit, but it was ever so gradual.

 

My fork went over a week or so period. I looked at it no less than 5 different times through the lower inspection port. It never showed any problems until it went completely. I switched out the master and slave and it actually got a little better. During the final bleed of the clutch the fork finally went and then I could see the crack.


The one thing I think I learned through the ordeal was dont do the rebuild for master/slave. Go ahead and get new ones.

 

I'm definately going to order a new fork when I buy the rest of the parts. If I don't need it I can always send it back.

Thanks for the input. I was on the fence about which way to go with this.

 

I'm not sure I understand the question. Prior to all this bs the resistance seemed to be linear through most of the pedal travel but would ramp up a bit near the bottom of the pedal travel. Is that what you were asking about?

 

I will switch out my fork at least every other clutch change (or more) from now on. I figured that it was my upgrade to the ACT XT clutch that killed the original one. The part is already known to have problems and there is no upgrade to take the higher weight pressure plate/ clutch combos so go ahead and put that new one in no matter what.


If you replace the wedge collar, fork, master, slave, and bleed the fluid there really is nothing else it could be.

 

Does anybody have a picture of the pilot bearing installed on a motor?
I have a picture of an uninstalled bearing, but not one where the bearing is installed AND clearly visible.

 

if he cannot get into any gear when the car is off i dont think the pilot is the prob. a new clutch fork is a good idea but the shift fork or select rods may be bent

 

It goes into gear with the car off. When the car is on with the clutch pushed in it won't go into gear.

 

ahh... then i am in the pilot bearing boat also

 

Pics attached of various damage: Bad pilot bearing in rear of crank, input shaft, and the infamous clutch fork.

I would highly recommend reinforcing the clutch fork--even a new one--to prevent recurring failure.

 

Thanks for the pics Speed of Light.


UPDATE:

The mechanic replaced the master and slave cylinders, and I still have the same problem (surprise surprise), but it is, and may have always been intermittent.

When the car is cold everything works fine. The longer I drive it the harder it gets to put into gear. Eventually it stops going into gear all together. However, when I shut the car off, push the clutch in, and start it in first everything seems to work fine again for several shifts.

Can I rule anything out based on this new info?

 

If you have replaced the master and slave then it is something in the tranny.

Most likely the fork.

Troubleshoot the entire system:

1) Pressure is applied to the pedal which is transmitted via metal rod to the master cylinder.

2) Fluid transmits pressure to slave.

3) Slave to fork.

4) Fork to throwout bearing.

5) Throwout bearing to wedge collar.

6) Wedge collar to clutch.


You replaced 3 of the 5 things it could be. (master/slave/fluid)

Now you have wedge collar and clutch fork left.


Sorry to hear you havent solved it...

Shawn

 

It's got to be something turning the input shaft (with the clutch disengaged): either the clutch assy or the pilot bearing. The fact that it's more prevelant when it warms up suggests that thermal effects--expansion and/or coefficient of friction--is playing a part.

Are you absolutely sure that the clutch disk is getting good clearance?

Has it been measured with a feeler gauge?

Does the clutch pedal feel springy at all, especially toward the end of travel?

If there is any appreciable amount of torque applied to the input shaft while you are shifting, the syncro will not be able to slow or stop the input shaft. Therefore, you will not be able to get it in gear.

Hate to be the bearer of bad news but... It's going to be either the clutch and/or pilot bearing. Disassembly and inspection is indicated. From my perspective, you will have to pull the tranny to fix this one anyway.