Topic: Clutch slipping when cold. Does that make sense? UPDATE 2 - PICS

[grumpy]

I installed new EDC frictions plates in the 750K3a few weeks ago.
Also added Barnett HD springs.

It worked fine for about a week - maybe 250mi or so.
Then one morning on the way to work it slipped a bit on the freeway on ramp (air temp was about 48deg). It was fine about 30 minutes later - after it was up to operating temp.
Worked fine on the way home (temp was about 80deg).
Next day, same thing - only it slipped a little more.
It has gotten progressively worse to where now it won't hold 65mph on level ground in 5th.

Cable & free play are fine so I'm gonna have to go in & see what's up.

Question is - why would temp matter?

[trueblue]

The oil is thicker when cold, and also the thickness of the clutch pack changes because it expands when it heats up, not a lot, but I find with my 650 if I adjust it when it is cold, it will slip when it gets hot.

[HondaMan]

Check the thickness of those new plates. I got a set of them, and they were all .131" thick, which is the worn-out spec for Honda's OEM plates. When they are too thin, the clutch pack is too short. This prevents them from being squished between the pressure plate and basket in enough strength to squeeze the oil out, so they slip.

OEM plates ranges from .144" to .141" new, considered worn out at .131" or less.

The modern sportbikes with 8, 9, and 10-plate clutches use the same ID/OD as these bikes, and I suspect the vendors are getting careless in their 'cross reference' applications. The way I have been telling them apart in a glance: the thin ones have a groove across the base of the metal tab that engages the clutch basket (this is the "plate dog" to the basket). The proper ones have no groove there.

In the early bikes (K0/K1), Honda had an extra steel plate at the base of the stack, used 6 friction plates instead of 7 as the original clutch. In the K2 after 1/72 or so, this changed to a 7-plate friction stack, with the steel one gone and the rear hub surface ground flat to meet the cork one. Today, you could possibly just add a steel plate to the stack (on the bottom) to make up the missing thickness with those plates. It comes out pretty close to 1/2 a plate thickness, overall. Use the Honda springs, though, as they will be compressed further, so HD springs will strain the hub when tightened this way.  Also, make sure the lifter does not bind when the clutch is pulled in: the adjustment will be close. It will get better as the plates wear in, though.

The slipping is most likely happening before the oil gets thin enough to let the plates grip with only the light contact they have now.

[KRMK3]

My K3 has developed a similar problem with the clutch. Both friction and steels are new yet within 500 miles it is intermittently slipping. Springs are new Honda ones and lifter bearing is also new. Clutch adjustment has been checked and is fine and have also adjusted slack in cable when problem occurs just to make sure it wasn't that causing the problem. Thanks for the information Hondaman will check my new EBC friction plates closely when I next pull it apart and if their thickness is the problem I'll try sticking an extra steel in the back of the stack as you suggest.

[750resurrection]

Check the thickness of those new plates. I got a set of them, and they were all .131" thick, which is the worn-out spec for Honda's OEM plates. When they are too thin, the clutch pack is too short. This prevents them from being squished between the pressure plate and basket in enough strength to squeeze the oil out, so they slip.

OEM plates ranges from .144" to .141" new, considered worn out at .131" or less.

The modern sportbikes with 8, 9, and 10-plate clutches use the same ID/OD as these bikes, and I suspect the vendors are getting careless in their 'cross reference' applications. The way I have been telling them apart in a glance: the thin ones have a groove across the base of the metal tab that engages the clutch basket (this is the "plate dog" to the basket). The proper ones have no groove there.

In the early bikes (K0/K1), Honda had an extra steel plate at the base of the stack, used 6 friction plates instead of 7 as the original clutch. In the K2 after 1/72 or so, this changed to a 7-plate friction stack, with the steel one gone and the rear hub surface ground flat to meet the cork one. Today, you could possibly just add a steel plate to the stack (on the bottom) to make up the missing thickness with those plates. It comes out pretty close to 1/2 a plate thickness, overall. Use the Honda springs, though, as they will be compressed further, so HD springs will strain the hub when tightened this way.  Also, make sure the lifter does not bind when the clutch is pulled in: the adjustment will be close. It will get better as the plates wear in, though.

The slipping is most likely happening before the oil gets thin enough to let the plates grip with only the light contact they have now.

Jeeze! Did that all come out of your brain like water ? ( Did I mean like taking a whizz... who knows)

I am voting that for Post of the Month ( there isn't one whatever...)
I have to say I was impressed.

[Old Scrambler]

Mark has the practiced art of telling us the hidden factors behind the occasional problems we encounter with our beloved machines............hopefully someone is going to archive his words of wisdom beyond what he conveys in his published books............carry on

[grumpy]

Yeah. Hondaman knows his #$%*e.

The quality of Hondaman's wisdome is not strain'd,
It droppeth as the gentle rain from heaven
Upon the place beneath. It is twice blest:
It blesseth him that gives and him that takes.

Or something.

[grumpy]

OK - update

The EBC friction plates were  .134 - .136. Not dead but nowhere close to what they should be for new.

I have a set of new Cycle X frictions. I measured them just to see: .136 - .137

I also have a set of frictions that saw at least 20,000mi. Make unknown. .1365 - .138.

Here's the kicker and the likely culprit...
The shoulders* on the lifter and the basket are visibly worn.
I checked them against another set (I have two motors) that shows almost no wear.
The difference was .015 on the center and about .01 on the lifter.
Add those together and it makes a new set of plates function like a dead set.




*the surface where the friction plates contact the lifter / center.

[HondaMan]

OK - update

The EBC friction plates were  .134 - .136. Not dead but nowhere close to what they should be for new.

I have a set of new Cycle X frictions. I measured them just to see: .136 - .137

I also have a set of frictions that saw at least 20,000mi. Make unknown. .1365 - .138.

Here's the kicker and the likely culprit...
The shoulders* on the lifter and the basket are visibly worn.
I checked them against another set (I have two motors) that shows almost no wear.
The difference was .015 on the center and about .01 on the lifter.
Add those together and it makes a new set of plates function like a dead set.




*the surface where the friction plates contact the lifter / center.

I think I'm having a drugged-brain moment: I'm not sure where this spot is that you're describing? It sounds reasonable, just not understanding where it is...

I have recently seen a case where notches were worn into the 'fingers' of the outer basket, which made the plates stick to them. This had slip when cold, but it also had difficulty disengaging when hot. The plates would grip those notches and drag on the adjacent steel ones until the throttle was blipped with the clutch pulled in. Then the clutch would rattle slightly if sitting still and pulled in, but would release. Had to file off a bunch of notches (love that Dremel!) to fix it up. (This was in May, I think?)

[HondaMan]

Check the thickness of those new plates. I got a set of them, and they were all .131" thick, which is the worn-out spec for Honda's OEM plates. When they are too thin, the clutch pack is too short. This prevents them from being squished between the pressure plate and basket in enough strength to squeeze the oil out, so they slip.

OEM plates ranges from .144" to .141" new, considered worn out at .131" or less.

The modern sportbikes with 8, 9, and 10-plate clutches use the same ID/OD as these bikes, and I suspect the vendors are getting careless in their 'cross reference' applications. The way I have been telling them apart in a glance: the thin ones have a groove across the base of the metal tab that engages the clutch basket (this is the "plate dog" to the basket). The proper ones have no groove there.

In the early bikes (K0/K1), Honda had an extra steel plate at the base of the stack, used 6 friction plates instead of 7 as the original clutch. In the K2 after 1/72 or so, this changed to a 7-plate friction stack, with the steel one gone and the rear hub surface ground flat to meet the cork one. Today, you could possibly just add a steel plate to the stack (on the bottom) to make up the missing thickness with those plates. It comes out pretty close to 1/2 a plate thickness, overall. Use the Honda springs, though, as they will be compressed further, so HD springs will strain the hub when tightened this way.  Also, make sure the lifter does not bind when the clutch is pulled in: the adjustment will be close. It will get better as the plates wear in, though.

The slipping is most likely happening before the oil gets thin enough to let the plates grip with only the light contact they have now.

Jeeze! Did that all come out of your brain like water ? ( Did I mean like taking a whizz... who knows)

I am voting that for Post of the Month ( there isn't one whatever...)
I have to say I was impressed.

Well, not really...it came from the Clutch section in my book.

[grumpy]

I think I'm having a drugged-brain moment: I'm not sure where this spot is that you're describing? It sounds reasonable, just not understanding where it is...

Drugs? Anything good?   


Caliper is pointing to the worn surfaces I was talking about

.

[HondaMan]

Ah, I see!
That top one (pressure plate) looks pretty beat up, like a friction plate crumbled off its cork in days gone by. So long as it's not warped, it will still work OK. This brings up a thought: if the surfaces are worn that far back and the new plates are already thin, maybe just plunk a steel plate on top of the stack to take up all the slack? It would let the top plate grip something other than the beat-up aluminum of that pressure plate.

Every AM I get to try something from the doc, to see if it helps my back for the day. I was injured when I was just 6 years old, and the arthritis in my back today can be killer on some days! So far there have been about 15 different things to try: some actually work! Most of them just make you stupid: gave that stuff up when I was in college.

[grumpy]

I thought about adding a steel. I didn't know if  it would work. I was afraid it would grind the aluminum away...
(but as I really think about how the clutch works I'm realizing the steel won't grind anything. it's the friction plates that could)
What's the max thickness the entire stack can be & still fit? Any idea?

Oh, and that surface really isn't that beat. That's just dirt from the garage floor stuck in the oil.
It wiped right off.

[HondaMan]

I thought about adding a steel. I didn't know if  it would work. I was afraid it would grind the aluminum away...
(but as I really think about how the clutch works I'm realizing the steel won't grind anything. it's the friction plates that could)
What's the max thickness the entire stack can be & still fit? Any idea?

Oh, and that surface really isn't that beat. That's just dirt from the garage floor stuck in the oil.
It wiped right off.

The total thickness of the stack when new is based on max cork plates of .144", which was the K0 OEM plates in 1969. So, 7 of those comes to 1.008" (25mm). The steel plates, I don't have the dimension on hand, but they are usually 6 in the stack (post-K1), so measure one and add that up.

So, if the new plates are only .131", then you are short by (7 x (.144-.131)")=0.091". If one steel plate is anywhere near this thickness, it would work. I seem to remember they are 1.5mm (.080"), but don't have one right here...

[trueblue]

I thought about adding a steel. I didn't know if  it would work. I was afraid it would grind the aluminum away...
(but as I really think about how the clutch works I'm realizing the steel won't grind anything. it's the friction plates that could)
What's the max thickness the entire stack can be & still fit? Any idea?

Oh, and that surface really isn't that beat. That's just dirt from the garage floor stuck in the oil.
It wiped right off.

The total thickness of the stack when new is based on max cork plates of .144", which was the K0 OEM plates in 1969. So, 7 of those comes to 1.008" (25mm). The steel plates, I don't have the dimension on hand, but they are usually 6 in the stack (post-K1), so measure one and add that up.

So, if the new plates are only .131", then you are short by (7 x (.144-.131)")=0.091". If one steel plate is anywhere near this thickness, it would work. I seem to remember they are 1.5mm (.080"), but don't have one right here...

Hondaman, 1mm is near enough to 0.040" so 1.5mm is only 0.060" not 0.080"

[HondaMan]

I thought about adding a steel. I didn't know if  it would work. I was afraid it would grind the aluminum away...
(but as I really think about how the clutch works I'm realizing the steel won't grind anything. it's the friction plates that could)
What's the max thickness the entire stack can be & still fit? Any idea?

Oh, and that surface really isn't that beat. That's just dirt from the garage floor stuck in the oil.
It wiped right off.

The total thickness of the stack when new is based on max cork plates of .144", which was the K0 OEM plates in 1969. So, 7 of those comes to 1.008" (25mm). The steel plates, I don't have the dimension on hand, but they are usually 6 in the stack (post-K1), so measure one and add that up.

So, if the new plates are only .131", then you are short by (7 x (.144-.131)")=0.091". If one steel plate is anywhere near this thickness, it would work. I seem to remember they are 1.5mm (.080"), but don't have one right here...

Hondaman, 1mm is near enough to 0.040" so 1.5mm is only 0.060" not 0.080"

Right you are! Thanks.

[grumpy]

My steels measure out to .078