Clutch not fully releasing - and shifting problem

[Scud]

Here are the current symptoms (none of which were problems before the new clutch):

• Slight drag: with the bike in gear and clutch pulled in, I can move the rear wheel by hand and I can push the bike. But this is just a little bit harder than when in it's neutral.
• Missing shifts: Most of my shifts are excellent, but I have a recurring problem with 3-4 and 4-3. I frequently miss the upshift and the downshift, which yields a false neutral, followed by an unpleasant gear-mesh noise on shifting out of the false neutral (up or down).
• Difficulty getting into neutral: If I come to a full stop in gear, it is difficult to find neutral. It requires more effort than it should, and mostly toggles between 1st and 2nd. This is not a problem with the engine off, only while running.
• Clutch lever release point: I have to pull the clutch all the way to the handlebar to get release.

 

Here's what I've done recently:

• Replaced clutch friction discs with SD-TEC plates sourced from MG Cycle.  I think this could be the source of my problem. When I ordered, Gordon at MG told me they were a little thicker than normal and sometimes cause clutch to not fully disengage. I notice that MG has recently updated their product description to indicate that the plates are 8.3mm, compared to 8.0mm factory spec. That means 2 over-thick plates adds 0.6mm to the clutch stack.
• Replaced intermediate plate and springs.
• Rebuilt shift pre-selector (springs, polish, etc.) and added Lucky-Phil-Chuck magic shift extender.
• I drained the Redline Shockproof Heavy last night and replaced with synthetic gear oil. No chunks of metal, but there was a little metal powder on the drain plug magnet. This seemed normal - in fact, it was relieving after the bad noises I've been hearing.

 

Thinking about next steps:

 

I have a new throwout bearing (I didn't replace it when I did the clutch). Gordon suggests that I use an extra washer between the throwout bearing and the piston (that would be three total washers instead of two). Supposedly, this is the equivalent of having a longer pushrod and may compensate for the additional thickness of the clutch parts. He said that has helped with manual (cable) clutches, but has never tried it with a hydraulic clutch. And he wants me to tell him if it works... which makes me the test-monkey.

 

So - It's kind of big job to get at the clutch slave cylinder. What do you guys think about the chances of this working?

 

Some relevant pics....

 

This is the SD-TEC clutch as currently installed.

 

 

 

These are the clutch pushrod assemblies from other bikes.

 

 

 

EDIT:  I had LGC plates, not SD-TEC. They are about the same thickness as OEM plates. See next page for measurements.

Edited July 2, 2017 by docc

[MartyNZ]

The symptoms you describe all suggest that the clutch is dragging, and a small amount of wear may fix it. Can you slip the clutch under power, with the front wheel nudged into the corner of the garage? Just bedding the friction material may be enough to help. Don't do that for more than a few seconds at a time to avoid overheating, because warped plates will make things worse.

If that doesn't work, then adding a washer sounds like a fine idea. 

 

An off the wall thought, can you temporarily fit the clutch hose to the front brake master cylinder? The greater volume of fluid from that larger cylinder may push the slave an extra 0.6mm, so you can test the extra washer theory. Although bleeding the brake and clutch again may be more hassle than just adding a washer.

[Scud]

I also thought that a small amount of wear could fix the issue. I think I have about 500 miles on the new clutch - including one short ride where I tried slipping the clutch a lot - but I don't like riding that way. 

 

Adding a washer = 5 minutes.  Getting to the point where I can add a washer = a few hours (plus reassembly =  a day).  And it just seems so close to totally releasing. Maybe I'll put the bike on the front wheel chock (and that against a wall) and try to bed the friction material as you described.

 

In fact, that sounds like fun... maybe a few smoky burnouts in the garage while I'm at it. I will have to be very strategic about the timing of this "maintenance procedure."    On a serious note, that also seems risky... like I could engage the clutch and lose control of the bike.

 

I rode it to work today. Maybe I'll abuse the clutch on the way home - but if the problem is really 0.6mm, that's going to take a long time to wear off.

[gstallons]

If you have no "free travel" in your clutch lever , the washer trick is not going to change things . Your master cylinder is going to give you X amount of fluid for the slave to move. If you add too much thickness the slave piston will bottom in the slave cylinder and will maintain constant pressure on the release mechanism. The metal surfaces the friction material contacts has to be perfect. If the discs are larger or smaller in circumference you will have a problem . Think of the o/s edge of a 33 1/3 LP . I hope you can . If the metal surface looks like this , the friction material will drag here and cause you problems.

[Scud]

Well, I decided that trying to wear the friction discs out was not my style... so in we go. 

 

 

Fortunately, you only have to remove a few parts to get access to the clutch slave... such as the lower subframe...

 

And here is an interesting development. 

• Thickness original (2 washers and throwout bearing) that I removed:  3.88 mm
• Thickness of new (captured bearing and 1 washer):  3.48 mm

So that's 0.4mm thinner, and I figure the clutch plates are 0.6 mm too thick, which pushes the clutch button (and pushrod and bearing) 0.6mm further toward the flywheel... which means the clutch slave starts 0.6mm overextended.

 

So now I have to make up 0.4 + 0.6 = 1.0 mm. And the washers I removed are 0.91...     And to top it off, I was actually helping my daughter with her math homework in the garage while working on this. Who says you don't use algebra in real life?

 

Note the difference in the new bearing (lower) from the old. It's captured into a plate so it does not need a backing washer. It's 2 pieces, vs. 3 for the original. I installed the parts as lined up on the bottom, with one extra washer.

 

 

Before removing the slave and piston, I put the bike in second gear. With the clutch pulled in, I could turn the driveshaft by hand, but I could tell there was resistance.  After installing the thicker stack of stuff with the new bearing, I could turn the driveshaft freely.

 

I am cautiously optimistic and gonna go put it back together.

[Chuck]

Sounds like you are on to something. Good info for the collective..

[Scud]

Ok, ride report. I did a 4,000 foot ride. That's net elevation increase... took 120 miles round trip to do it. I followed a red, white, and gold Aprilia Dorsodura for while. That is fine looking and sounding supermotard. Then he got aggressive with passing over double-yellow lines... bye-bye.

 

The thicker stack of material, including the new bearing and extra spacer (washer) has definitely released the drag on the clutch. I can get into neutral easily at any time. And this is the first time since I installed the new clutch that I have heard the ching-ching-ching with the clutch pulled in, so I know the plates are moving freely when the clutch is in. There was no ching-ching-ching before.

 

Shifts are improved, though I still need a full squeeze on the clutch lever and I miss shifts if I get lazy. However, the make-up shift (following the miss) goes in better - it previously made painful re-engagement noises after missed shifts. I suspect that I may still have a little problem with the pre-selector, as I had a hard time with a few bits when I rebuilt it. That can wait...

 

All in all - totally ridable now. I just have be precise in my shifts. This is difficult coming off my other bikes, because they are all easy-shifters and I can get away with sloppy/lazy shifting (even the Scura).

 

I think the key problem here is that the clutch plates are too thick. It seems that I could probably have put in two spacers instead of one. But I think I'll just ride it for a while and see how it goes. Maybe a little plate-wear will improve things even more.

[footgoose]

I'm glad it was "easy" for you. I appreciate the running updates. As usual you are concise and informative. As Chuck said, good info for the collective.

[gstallons]

I assume you did not install the release bearing dry ?

[Scud]

1- If you have no "free travel" in your clutch lever , the washer trick is not going to change things . Your master cylinder is going to give you X amount of fluid for the slave to move. If you add too much thickness the slave piston will bottom in the slave cylinder and will maintain constant pressure on the release mechanism. 

 

2 - The metal surfaces the friction material contacts has to be perfect. If the discs are larger or smaller in circumference you will have a problem . Think of the o/s edge of a 33 1/3 LP . I hope you can . If the metal surface looks like this , the friction material will drag here and cause you problems.

 

1 - I think the master cylinder piston may have been extending too far, because the too-thick friction plates would push the little cup forward. The cup will gradually move backwards as the friction material wears. The complete pushrod, bearing and spacer assembly is now 0.5mm longer than stock. I think this compensates for the extra material and is why the "washer trick" did end up fully releasing the clutch.

 

2 - I was trying to figure out what you meant here. I put in new friction plates and intermediate plate, but I re-used the pressure plate and the starter ring (both of which contact the friction plates). These did not have any noticeable lip on them.

 

I assume you did not install the release bearing dry ?

 

Correct. I dabbed some gear oil on the bearing and all the other little parts. Everything to the right of the O-ring in the above photo also gets lubricated by the normal operation of the transmission.

 

After all this, I regret buying the SD-TEC clutch plates. According to Gordon at MG Cycle, they've been the most long-lasting and trouble-free plates he's carried. However, they just recently manufactured a batch that was too-thick. So... I'd avoid those until they run a correct batch. If, at some point, I feel the need to pull the engine out again, I might try putting a 5-speed RAM in there (since I have a spare clutch hub to make it fit the six-speed). 

[Scud]

I'm glad it was "easy" for you. I appreciate the running updates.

 

...and maybe not so easy after all.

 

Friday, after about 75 miles of non-stop freeway driving, I hit the throttle to make a pass and noticed the engine revved a bit more than it should have - then caught. It was like the clutch was slipping. Squeezed the clutch and it was really firm - no freeplay at all. In order to get it to release, I had previously adjusted it so there was very little freeplay in the lever.

 

Pulled over and backed the adjuster out of the lever until it had freeplay again. To explain how it works OK cold, but caused a problem after a longer distance, I think that with the heat, things expanded a bit and that ended up putting full-time pressure on the clutch pushrod. After the adjustment, it was manageable for the rest of my trip (about 500 more miles). But it was difficult to get into neutral while idling and I missed a few more 3>4 shifts. I think this means the clutch is still dragging a little - even though I could hear the ching-ching-ching sometimes.

 

Last week (before my trip) I updated Gordon at MG Cycle - that the extra spacer did release the clutch, but that I am still not happy with the lever effort and the need for a full pull to the bar.

 

He said he had just learned that when the original springs became unavailable, then Moto Guzzi went through three phases of clutch springs for the 10-spring clutches:

 

First - all 10 softer springs (as I removed from LeMans) - and too soft to allow full wear of the plates

Then - all 10 firmer springs (as I recently installed) - and too stiff for a comfortable lever-pull

Finally - alternating with 5 of each of the above - the just-right "goldilocks" solution.

 

So Gordon has kindly sent me 5 softer springs.

 

However, I don't think a spring change is going to solve the issue with lever travel. So... the clutch has got to come out - and probably needs the original-thickness plates installed.

[andy york]

Wow.... I hate to hear that. I'm on the road for about 10 more days. When I get home I'm going to check the spring pressure

on all the ones I have ....against a new set i ordered. I also ordered a set of stock plates..(.just for measuring purposes) lol.

Gut feeling here....I just don't think you loose that much spring pressure...or I don't....cause I am generally always moving .

My LeMans is definitely not a city bike. Sweetie says the slave cylinder is on the kitchen table....so I will be ready for some "garage" time when I get home.

I have done the job both ways ...crabbing the back    and pulling the motor.  Its a wash to me ..although if the clutch slave had

not been bad, I would have pulled the motor for the clutch ....then again, if it were not for the slave I would not be 

doing any of this 

later  

andy

[Lucky Phil]

A few things Scud, thicker clutch plates will slightly increase the lever pull effort. The springs are now .6mm more compressed so the lever load will be a fraction higher. It shouldn't however effect the amount of lever travel required to get the plates to release once you have shimmed the pushrod as you have done. Some thoughts on that, Firstly if the clutch friction compound is a little more aggresive than the std type or if the new plates aren't nice and flat as they should be then you will need a little more lever travel to create more friction plate clearance than you're used to.

I know its obvious but is the clutch bled properly? If the new plates are as mentioned then there will be less tollerance for air in the system than the std plates and you may figure it was ok before so it should be now, understandably.

All this will effect neutral finding and general clutch take up from a standstill but it wont have any real effect on shifting. A little bit of clutch drag doesnt really effect shifting much as the gearbox likes to spin to change nicely, its about real load when it comes to clean shifting not full and total release.

I'm very used to clutchless upshifting on the track and so also use it a lot on the road even with the Guzzi. So unless I'm tootling around town or bumbling along in the country with my wife on the back i never use the clutch from second through to 5th gear. I'll use it into 6th only because by that time I've hit cruize mode and it's like relax time:) I've never, repeat never missed a clutchless upshift on the Guzzi, or my track bike so as I said full and complete release isnt that important for shifting as long as the load is reduced its fine.

I really don't know why these aftermarket companies dont take the extra effort to just make the plates to the std dimensions, I mean the difference in thickness between a new clutch plate and a worn out one isnt that much and .3mm doesnt sound like a lot but is quite significant in the scheme of things when you consider it has to be multiplied by a factor of 2 in the case of the Guzzi. That .6mm may be just whats needed to allow for less than perfectly flat friction and drive plates.

 

Ciao 

[Scud]

Thanks Andy and Phil.

 

I'm starting to wonder if the 0.3 mm over-thickness applies to each friction surface, not to each complete disc. If that's the case, then the two plates are 1.2mm over-thick in total. Sadly, I did not record my measurement when I installed these plates - I just remember verifying that they were thicker than the ones I removed.

 

It's totally OK to suggest obvious stuff. I had a fleeting thought that I should re-bleed the clutch - so I will do that soon, maybe tonight. If it doesn't help at least it will rule-out something easy.

 

And yeah, I was also thinking about the springs being pre-compressed too far due to the thicker plates. The lever effort is therefore even harder than it would have been with 10 new stiff springs - and this puts more pressure on the seals in the hydraulic system, which can't be good.

 

As it's adjusted now, it clonks into first gear at a stop and the bike makes a little lurch. I did not have that clonk-and-lurch when I left for my trip few days ago, but then all the freeplay went away at operating temperature - so I backed off the lever adjuster (as I described earlier).

 

I'm working on my Popeye forearms.

 

This is pretty frustrating - I just want to put on some new tires and go riding.  But it does make me appreciate the engineering and fine tolerances required to make a clutch work correctly.

 

Assuming the clutch bleeding does not solve the problem, I'm gonna take the stock clutch plates out of the champagne project. That clutch is working perfectly with adequate free-play at the lever. I have a spare low-mile transmission too. I'll probably compare the two on the bench to see if they shift the same.

 

Man vs. machine.... 

[Scud]

...and my Scura (with the RAM single-plate clutch) shifts flawlessly and is quite happy with clutchless upshifts.

 

My current problem is on the twin-plate LeMans.

 

Phil - does your bike still have the twin-plate?  I think I recall that you were going to install a RAM.