29 replies to this topic

[gerglmuff2]

this isnt my spec miata, another 1.6 i have though for autocross competition. 

i was out last night taking some of the mold release off the new tires for this weekends event. previously the clutch was slipping so i put a new clutch in it 2 weeks ago. 

so i hit a pretty heavy launch, and hear and feel a bang, and then the car seemingly has no drive. so i get my truck and trailer and pull it back to the garage (so handy). i expect its the diff (1.6 vlsd), so i jack up the rear, and run it, in gear, and the wheels turn, no clunking etc etc. i move them by hand, no issue, driveshaft spins etc. ive blow diff before, and this one looks good. 

so i read on the internt that a good test is to put it in gear, on stands, and then push the brakes. the clutch/everything else is work fine if this stops the engine. so i do this test, and the engine keeps running. this has me really curious, so i put in on the ground and now it will like kinda creep (?) in gear. like the transmission goes into all the gears, and when you let out the clutch, the car just kinda slowly rolls in whatever direction the transmission is (ie, reverse, its backwards, 1-5 its forwards). pushing the gas at this point revs the engine, but doesnt increase thrust. 

so i think the engine is good, i think diff is good, i think the transmission is good ... that leave the clutch right? 

so i pull the transmission this morning (less than 2 hours woot woot), and the clutch looks .... okish? like you can see some worn spots, some dust etc, but it looks pretty good. i was expecting it to be broken somewhere or something. 

i have a spare ACT 4 puck laying here, worked when i pulled it (one of those "while i was in here" things). im considering putting it in with that, but im also concerned there is something im missing. 

anything im missing? its gotta be the clutch right? its not crankwalk is it? 

[Michael Novak]

Could be something broken in the transmission OR CV shaft that looks like it is all together, but is broken inside one of the boots.. 

[gerglmuff2]

fluid came out of the transmission clean, and with no chunks. but ill be trying to turn it by hand in each of the gears here when i go back outside and work. ill also see about the CVs. i looked at the CVs and didnt see anything, and when testing the diff they seemed fine, but ill double check em. 

[Steve Scheifler]

You say the clutch disk is intact but what about the pressure plate? That’s what produces the clamping force. Did the clutch pedal feel any different? The lack of any conspicuous sound during your jackstand and creep tests is interesting but a loud exhaust can drown out a lot. I assume no new vibrations during those tests?

[gerglmuff2]

clutch felt the same, but cant really tell when it "grabs" because it doesn't really grab. but it does go up and down fine, no issues. pressure plate looks good, has some black marks (already?) around the edges. this is with less than 100 miles on it. 

no new vibrations in the testing, no knocking or other rhythmic sound in testing. 

i suspect the clutch is just slipping massively because the clutch that was in it, was a 6 puck, and i put in a more streetish one, an exedy stock replacement. the car makes probably in the 120s rwhp, and revs to 8000, racing beat header, megasquirt, all that stuff that we can't do in spec miata, so powers up 10-20%. i figured a stock replacement would be fine, but maybe not. 

EDIT: cvs good, and trans works in all gears and reverse when turned by hand. still no noise from the diff too. 

[Bench Racer]

A VLSD is made up of thin perforated plates that are suspended in a silicone fluid. The plates shear the fluid as the wheels rotate at different speeds. This shearing can equalize the different wheel rates thus limiting slippage. The VLSD only provides a 15-20% locking function when new, so they are not really meant to be a performance item. I have found that the locking action of the VLSD takes a moment to activate when one wheel starts to slip so it usually takes affect after you have already exited the corner. They are really meant to help you get the car moving from a dead stop in slippery conditions and not a performance aid. The VLSD works just fine for it's intended street usage but it will effectively stop limiting slip, from a performance point of view, at some point as the plates can break, the silicone fluid can leak and the tolerance between the plates can wear. For more detail information on how a VLSD operates check out the links page for the BMW and 4x4 articles.

 

The shop manual states to test a VLSD you put a torque wrench on the hub (wheel in the air) and turn 90 degrees with 11 ft lbs of torque (car in gear). This should take no less than 4 seconds. If less then you have either an open diff or the VLSD is toast. Another test for a good VLSD is to jack up the rear with both rear wheels in the air. A good VLSD will turn both rear wheels in the same direction when manually turning the drive shaft. The open unit (or toasted VLSD) will turn the wheels in the opposite directions. There is no rebuild kit for the VLSD and the unit is completely sealed so you can't even get inside it short of cutting it

[gerglmuff2]

A VLSD is made up of thin perforated plates that are suspended in a silicone fluid. The plates shear the fluid as the wheels rotate at different speeds. This shearing can equalize the different wheel rates thus limiting slippage. The VLSD only provides a 15-20% locking function when new, so they are not really meant to be a performance item. I have found that the locking action of the VLSD takes a moment to activate when one wheel starts to slip so it usually takes affect after you have already exited the corner. They are really meant to help you get the car moving from a dead stop in slippery conditions and not a performance aid. The VLSD works just fine for it's intended street usage but it will effectively stop limiting slip, from a performance point of view, at some point as the plates can break, the silicone fluid can leak and the tolerance between the plates can wear. For more detail information on how a VLSD operates check out the links page for the BMW and 4x4 articles.

 

The shop manual states to test a VLSD you put a torque wrench on the hub (wheel in the air) and turn 90 degrees with 11 ft lbs of torque (car in gear). This should take no less than 4 seconds. If less then you have either an open diff or the VLSD is toast. Another test for a good VLSD is to jack up the rear with both rear wheels in the air. A good VLSD will turn both rear wheels in the same direction when manually turning the drive shaft. The open unit (or toasted VLSD) will turn the wheels in the opposite directions. There is no rebuild kit for the VLSD and the unit is completely sealed so you can't even get inside it short of cutting it

 

a vlsd failing to open would not cause any of these issues. 

[gerglmuff2]

welp, put the 4 puck in with a new pressure plate. so we'll try this. 

[gerglmuff2]

was not the clutch, finished the job here tonight ... and nope, still the brakes will not kill the engine. something still slipping. suspect the diff or the transmission. when the transmission was out, i tested it in all 5 gears and reverse and everything turned properly and easily by hand, so i was pretty sure that was fine. the diff ... well ... i suspected the diff when it happened, but jacked up, everything turns right, no noises, no clunks etc. i have a hard time thinking its the diff too. 

i am paranoid, could this be a form of crankwalk? the bang when it happened didnt seem crankwalky to me, and crankwalk typically is not being able to get into or out of gears, not lack of engagement. 

[Steve Scheifler]

Not sure what you mean. The flywheel is bolted to the crank, the clutch plate bolts to the flywheel. When clamped by the plate the splined disk transfers the power to the transmission input shaft and so on. If the engine is running, the crank is spinning and if the clutch is engaged then the transmission input shaft is spinning (unless clutch or shaft splines are stripped). Something downstream has failed and although you are having difficulty detecting it there aren’t many possibilities. Your initial impression that it’s the diff still seems the most likely.

[gerglmuff2]

yeah im referring to the test of having the rear wheels up in the air, and then having it in gear and running (so the rear wheels are spinning) and then pushing the brakes. this should simulate coming to a complete stop, and thus kill the engine ... but it doesnt. means it still has little/no drive. 

i have a spare diff, i can throw that in, its a quicker/easier change than the transmission, and i can check the hubs/half shafts at the same time even better. i'll also have a second set of hands tomorrow to help, maybe worth doing the test with someone pushign the brake petal so i can lay under it and see whats happening. maybe hear the diff better? 

[gerglmuff2]

so i just did "the test" in all 5 gears and reverse. feels fine in 1-2-3 and R and the brakes do not stop the engine. in 5th they do. and in 4th and 5th the car starts to have a pretty serious vibration. 

i also noticed, the speedo works, and reads properly. that means that the transmission output shaft is spinning at the right rpm for the engine/gear .. which means the problem is farther back. 

its the diff. its gotta be the diff. 

[Steve Scheifler]

Now you have me sure it’s the transmission.
Saul?

[FTodaro]

Just how are you testing this what kind of load. If it was the trans there would be metal involved on the magnet and I would expect one of the gears would not work. Before i started pulling all this out i would get out a bore scope.

[SaulSpeedwell]

I was peer-pressured into ringing in.  The transmission cannot slip.  The differential cannot slip.  Both contain gears in constant mesh.  "Slipping" would mean "stripping" the gears, and that won't last for long, and will be accompanied by hellacious noise and shrapnel on the drain plug.  In any case, 4th gear is "direct drive", meaning the gears are carrying no load at all.  You could shift the car into 4th gear, throw the entire countershaft through the glass door of a Target store in Minneapolis, get up to 88 mph, and go back to a time when plastic straws were a newsworthy problem.  In 4th gear, all of the torque is carried through the main shaft and the "coupling" that joins it to the input shaft when you are in 4th gear.  

A new clutch disc is a clutch that is not broken in, and a likely contributor, or perhaps the sole cause of, the slippage.  It is the only thing in the drivetrain designed to, at times, slip.  It is the only thing in the drivetrain whose intended end-of-life failure mode is slippage.  Nothing else would "slip" without the tearing of metal.

Consider "the test you read about on the Internet" as, perhaps, not being a great, or even good, idea - although I have certainly run cars on lifts and jackstands for various diagnostic reasons - such as pinpointing sources of noise.  If you were getting vibration running on jackstands in 4th and 5th, it may be because the CV joints were at their least happy angle with the rear wheels being in full droop, and you were most likely spinning everything >4X faster than when you were testing in 1st/2nd.  

Clutch slippage always shows up in higher gears (numerically lower ratios) more readily, because the engine doesn't have the mechanical advantage of the lower (numerically higher ratios) to get the 2300 lbs moving very quickly, thus the the torque "across" the clutch's friction couple is higher for a longer period of time.  I've seen a lot of "new clutch is slipping" Miatas, usually boosted cars running aftermarket clutches, and every time we take it apart, the new clutch is using only the outboard ~1-2 cm of the friction surface because Boosty McBoosterson didn't take the time, or, to be fair, didn't have the time, to break in the new clutch before engaging in whatever competitive endeavor they were headed to.  I don't have an explanation for why new clutch discs don't seem particularly "flat", but I suspect it may have to do with aftermarket quality in general, since new OEM cars don't seem to require clutch break-in.

It is probably the clutch.  The "creeping" observation is a disconcerting clue.

Crankwalk is pretty easy to determine, but I can't imagine it being bad enough to cause clutch issues.  Convince yourself.  Find a non-essential worker sitting at home, invite them over to press on the clutch pedal while you look at the crank pulley.  Try it stationary, try it running.

 

It is probably the clutch. 

Depending on how much you value your time, and how much income inequality you will address when compensating your clutch-pedal-pumper, it may make sense to replace the flywheel, disc, and pressure plate, all at once.   The clutch's ability to resist torque is a function of the friction between the flywheel and the disc, and the pressure plate's pressure ring and the disc.  That means you need 4 individual surfaces happily gripping each other, whereby those four surfaces are compressed together by the very short stroke diaphragm spring in the pressure plate.  If the flywheel has been "resurfaced", you will have lost some necessary preload on the diaphragm spring, unless the amount of material removed was compensated on the clutch assembly.

Not that you or anyone cares, or should care, but I always hated the clutch-drop abuse of parking lot cone racing and drag racing.  I was pleased when Car and Driver started doing rolling start 5-60 mph times alongside the "Rev to 4500 and dump the clutch if you want to get the best-case published 0-60 times for this turbo AWD car".  Then again, outrunning WRXs with a Sienna or CX-5 Turbo is fun in its own right.  

It is probably the clutch.
 

[Steve Scheifler]

That’s the Saul we know and love!

I was hoping the clutch was now ruled out even though it was easily the most likely culprit. But as Frank and Saul have said, even if you managed to knock all the teeth off a gear in one clean wipe, there should be chunks at/on the drain plug.

Another reason it’s the clutch: Rule #1 A new sudden problem is usually related to the last thing you worked on.

You had recently changed the clutch, and you just swapped parts around again. When bolting on the plate did it require “pulling down” significantly all the way around as you tightened the bolts in sequence through several rounds? Point being, if you have the right parts the disk will hold the edges of the plate well off the flywheel (visualizing here so approximately) a good 1/8 of an inch or more. Then as you torque in stages it pulls down, loading the clamping force onto the disc. If you have the wrong plate or too thin of a disc, or cleverly put lock washers between plate and flywheel (yes, I’ve seen that) then the plate isn’t going to clamp and everything spins.

The bang you heard still isn’t explained though. Is there a chance that the throw-out fork and/or bearing or pivot are f’d up resulting in negative free-play (always some pressure against the pressure plate like a chucklehead who uses the clutch pedal as a footrest)? Get under there and make sure that you can manually move the fork back & forth enough to feel/hear the bearing slap the plate fingers and that the slave isn’t sticking partly out.

[LarryKing]

Just out of curiosity did you manually wiggle the pinion flange where the drive shaft u-joint bolts on?

 

You could bolt everything back together and do another burn-out to see what blows-up.

 

I went to school with Boosty McBoosterson. Great guy!

[gerglmuff2]

I was peer-pressured into ringing in.  The transmission cannot slip.  The differential cannot slip.  Both contain gears in constant mesh.  "Slipping" would mean "stripping" the gears, and that won't last for long, and will be accompanied by hellacious noise and shrapnel on the drain plug.  In any case, 4th gear is "direct drive", meaning the gears are carrying no load at all.  You could shift the car into 4th gear, throw the entire countershaft through the glass door of a Target store in Minneapolis, get up to 88 mph, and go back to a time when plastic straws were a newsworthy problem.  In 4th gear, all of the torque is carried through the main shaft and the "coupling" that joins it to the input shaft when you are in 4th gear.  

A new clutch disc is a clutch that is not broken in, and a likely contributor, or perhaps the sole cause of, the slippage.  It is the only thing in the drivetrain designed to, at times, slip.  It is the only thing in the drivetrain whose intended end-of-life failure mode is slippage.  Nothing else would "slip" without the tearing of metal.

Consider "the test you read about on the Internet" as, perhaps, not being a great, or even good, idea - although I have certainly run cars on lifts and jackstands for various diagnostic reasons - such as pinpointing sources of noise.  If you were getting vibration running on jackstands in 4th and 5th, it may be because the CV joints were at their least happy angle with the rear wheels being in full droop, and you were most likely spinning everything >4X faster than when you were testing in 1st/2nd.  

Clutch slippage always shows up in higher gears (numerically lower ratios) more readily, because the engine doesn't have the mechanical advantage of the lower (numerically higher ratios) to get the 2300 lbs moving very quickly, thus the the torque "across" the clutch's friction couple is higher for a longer period of time.  I've seen a lot of "new clutch is slipping" Miatas, usually boosted cars running aftermarket clutches, and every time we take it apart, the new clutch is using only the outboard ~1-2 cm of the friction surface because Boosty McBoosterson didn't take the time, or, to be fair, didn't have the time, to break in the new clutch before engaging in whatever competitive endeavor they were headed to.  I don't have an explanation for why new clutch discs don't seem particularly "flat", but I suspect it may have to do with aftermarket quality in general, since new OEM cars don't seem to require clutch break-in.

It is probably the clutch.  The "creeping" observation is a disconcerting clue.

Crankwalk is pretty easy to determine, but I can't imagine it being bad enough to cause clutch issues.  Convince yourself.  Find a non-essential worker sitting at home, invite them over to press on the clutch pedal while you look at the crank pulley.  Try it stationary, try it running.

 

It is probably the clutch. 

Depending on how much you value your time, and how much income inequality you will address when compensating your clutch-pedal-pumper, it may make sense to replace the flywheel, disc, and pressure plate, all at once.   The clutch's ability to resist torque is a function of the friction between the flywheel and the disc, and the pressure plate's pressure ring and the disc.  That means you need 4 individual surfaces happily gripping each other, whereby those four surfaces are compressed together by the very short stroke diaphragm spring in the pressure plate.  If the flywheel has been "resurfaced", you will have lost some necessary preload on the diaphragm spring, unless the amount of material removed was compensated on the clutch assembly.

Not that you or anyone cares, or should care, but I always hated the clutch-drop abuse of parking lot cone racing and drag racing.  I was pleased when Car and Driver started doing rolling start 5-60 mph times alongside the "Rev to 4500 and dump the clutch if you want to get the best-case published 0-60 times for this turbo AWD car".  Then again, outrunning WRXs with a Sienna or CX-5 Turbo is fun in its own right.  

It is probably the clutch.
 

this is why i was so sure it was the clutch too. when i did the clutch 2 weeks ago i used a resurfaced flywheel (including spring plate surfaces) and brand new everything else (plate etc), and when i toss in the extra clutch yesterday, that was a good act 4 puck (was fine when it was taken out), with a new pressure plate, same (less than 100 miles) new flywheel. same symptoms though. 

the only symptom that doesn't sound clutch related was the initial symptom: a big bang. when clutches go bang its also a huge mess, and the clutch that came out looked ... ok. so the bang probably did not come from the clutch. 

i also agree that if there was a problem with the trans it would be a nightmare of grinding, and nothing would work. the transmission is two shafts with like 12 gears that all have to line up, if anything was bent, or broken in it, it would be a mess and none of it would work. never mind working fine when moved by hand, and easy to get into all 6 positions etc etc. 99% sure the trans is fine. 

i'll have more info tonight, and likely change out the differential. i still think its the diff.  the diff failing is 1. super common, 2. explains the bang, and 3. its easier than doing a 3rd clutch job in 2 weeks. i did not shake the diff pinion assembly last night (should have though, good tip). 

maybe the creeping is like a tiny amount of tooth engagement, or the fluid in the diff being spun and making a slight amount of force. the creeping is slower than if you push the car by hand. its not a strong force. i would not describe the car as having drive. the diff does "work" though when pushed by hand, which is also really freakin weird. 

[Ron Alan]

There is a lot in these post so apologies if I missed this...but does the driveshaft spin when you put in gear while the car is running?? Yes look behind, no look in front! 

If it is the diff I'm sure you are kicking yourself for not really diagnosing first before you broke out the wrenches...so I'll kick first...doh  

 

Dont discount a broken half shaft...I had this happen once...I wish i could remember which end but It broke at a snap ring. Super un clear now but I think it was not installed correctly(snapped in) or the snap ring somehow did not hold and the axle backed out of the correct position slightly. The last 1/2" broke off(I think this was all that was engaged)and lost drive.

[chris haldeman]

Didn’t take the time to read all the posts. If the car creeps in gear with the clutch pedal pushed and also doesn’t stall or move with authority the disk is installed backwards.