Clutch master cylinder - gradual failure or sudden?

[OddDuck]

Do (can) clutch master cylinders fail gradually over time, by progressively shifting less and less fluid? I.e. clutch disengagement very gradually gets worse?

Or is it more usually a sudden, clear failure one day?

Is there an accepted mileage at which it's generally considered a good idea to preventative-maintenance the master by replacement?

Asking since the 900SS is now over 100,000 km's / 22 years old and wondering if it's prudent to simply replace the clutch master - it's had normal street bike use. The clutch master is still working but am having trouble with clutch disengagement. Fluid's old, I'll do a bleed shortly. The clutch line is braided stainless (salvage off a wrecked bike, not new) and I'm running an Oberon clutch slave, installed at approx. 65,000 km.

[AllanB]

Last one I had go (years back) was progressive until it started leaking. Japanese bike so a repair kit fixed it.

Suspect lots of the modern stuff is like everything now days and you replace, not repair.

Which is interesting in a world of plastic bag hate, enviro this and that. Vehicles are becoming single use items.

[SaferRides]

Masters usually fail gradually.

It's well worth changing the fluid regularly, especially if the slave is mounted somewhere that gets hot.

[neels]

There's not much in a master cylinder that can spectacularly fail, unless the seal completely disintegrates. Wear on the internals can result in worse performance due to not generating the pressure it should be, and as it doesn't move a lot of fluid there may not be an immediately visible leak.

Reconditioning is dependent on a few things, if the bore is worn that usually requires boring out and sleeving, at which point it's probably easier to replace* the whole thing.





*By replace, I mean upgrade......

[WALRUS]

Dry clutch, yeah?

When was the last time you serviced/replaced your clutch plates and springs, etc?

[OddDuck]

Dry clutch, yeah?

When was the last time you serviced/replaced your clutch plates and springs, etc?

Last week - brand new basket and friction plates. I'm re-using the steels since I've put a few hours in deburring the spline teeth on these. The steels were cleaned of old plate residue via impromptu surface plate and wet'n'dry paper before reinstallation.

The springs are original stock, and you have a point there - I haven't checked for equal lengths uncompressed or length of compression under load. A skewed pressure plate wouldn't help.

The issue I've got is poor clutch disengagement - I believe that this will kill the new friction plate tangs in short order if it isn't sorted out. The tangs will get hammered on every gear change unless the engine is blipped correctly during the shift. I think I need to get more lift on the pressure plate.

At this stage the likeliest culprit is 2 y.o. clutch fluid (yeah I know but I've had other bike issues to sort out, see the Winter Layup thread) and / or a badly worn bore / seal in the clutch master cylinder, which has 100+ km's and 22 years on it. Both are believable. While the bike's apart it'll be worth checking the springs, as you point out.

The general feedback (if I read the replies right) is that the master is probably getting tired. Thanks everyone, makes my decision a bit easier...

Brand new direct replacement is affordable (approx $180 NZD), the nearest upgrade was worth a look but at approx $700 is probably too good for this particular bike. I've always been OK with the stock master anyway, I'm not hammering the bike on track days or similar.

[Drew]

Check your chain tension.

[WALRUS]

Fair enough. Well yeah, have a squizz.

Whenever I'm ordering parts for the 996 I go through these guys.. Decent prices (over here, at least) and good service.

https://ducati-gowanloch.com/product-category/clutch-2/

If you have to go down the way of replacing parts, hopefully this site could help :)

[OddDuck]

Fair enough. Well yeah, have a squizz.

Whenever I'm ordering parts for the 996 I go through these guys.. Decent prices (over here, at least) and good service.

https://ducati-gowanloch.com/product-category/clutch-2/

If you have to go down the way of replacing parts, hopefully this site could help :)

Thanks for that - I usually go with these guys

https://stein-dinse.biz/

The parts diagrams can be useful even if you aren't shopping.

Drew - not a challenge or anything but why would chain tension lead to clutch disengagement issues? Genuinely curious, it's not something I've encountered before. I think tension is correct but am shortly off to the garage, I'll have a look while I'm down there.

[AllanB]

2 years is not much on clutch fluid - general servicing is three years.


Reminds me, I've had mine three years now. Where the heck have I stored that speed bleeder after changing house.........

[OddDuck]

OK, am back. Chain tension OK, at the slack end of the workshop manual specification.

Clutch bled and fluid changed.

Travel before: 1.1 mm (!)
Travel after: 1.3 mm

These were measured via vernier calipers between one of the spring caps and the pressure plate.

Master cylinder piston travel: approx 9 mm. Quick calculation to work out expected travel at pressure plate:

Vol Master = Vol Slave (assuming incompressible fluid)
13mm diameter master
30mm diameter slave

Area x Travel (master) = Area x Travel (slave)

Area = PI/4 x Diameter squared

so crunching the numbers I expect to see 1.7mm, if the system's perfect. There are 7 friction plates, that gives 14 interfaces... 1.7 / 14 = 0.12mm gap between plates, with the clutch fully pulled in. That isn't much leeway for warped plates.

While doing the travel measurements I finally noticed something subtle: it looks like the clutch hub is moving, soaking up some of the travel as the pressure plate moves outwards. It's only a fraction of a millimeter. I'm running the stock 2-piece hub, the one with the rubber cushes. I saw this by eye, by looking vertically downward in the plane of the spring caps, while working the clutch.

Le sigh. I'll carry on in the Winter Layup thread.

[Drew]

I asked about chain tension because it's the most common cause for stiff gear changes and not being able to find neutral.
Does your clutch drag at a stand still, or is it just shifting gear that is an issue?

[OddDuck]

I asked about chain tension because it's the most common cause for stiff gear changes and not being able to find neutral.
Does your clutch drag at a stand still, or is it just shifting gear that is an issue?

Both - drag while stopped and choppy gearchanges. Finding neutral currently not possible with engine running. Chain tension checked and found OK.

[Voltaire]

Slave cylinders seem to be a common upgrade for Duclattery clutches.
I prefer a wet clutch to the sounds of a can stamping factory....

[Grumph]

so crunching the numbers I expect to see 1.7mm, if the system's perfect. There are 7 friction plates, that gives 14 interfaces... 1.7 / 14 = 0.12mm gap between plates, with the clutch fully pulled in. That isn't much leeway for warped plates.

While doing the travel measurements I finally noticed something subtle: it looks like the clutch hub is moving, soaking up some of the travel as the pressure plate moves outwards. It's only a fraction of a millimeter. I'm running the stock 2-piece hub, the one with the rubber cushes. I saw this by eye, by looking vertically downward in the plane of the spring caps, while working the clutch.

Le sigh. I'll carry on in the Winter Layup thread.

Clutch hub wear is a prime cause of lost travel on the lifter mechanism. Somewhere there may be a figure for endfloat - but less is generally better.
The 80's Honda fours are very bad for it as the alloy hubs wear on steel thrust washers. In extreme cases the back of the hub may need machining and a shim fitted. Wouldn't surprise me if high mileage Ducs suffer the same thing.

[OddDuck]

Clutch hub wear is a prime cause of lost travel on the lifter mechanism. Somewhere there may be a figure for endfloat - but less is generally better.
The 80's Honda fours are very bad for it as the alloy hubs wear on steel thrust washers. In extreme cases the back of the hub may need machining and a shim fitted. Wouldn't surprise me if high mileage Ducs suffer the same thing.

I got into the hub in the Winter Layup thread - the finding (in short) was that the hub was flexing on its own cush rubbers and also was possibly misaligned on those same rubbers. The cushes can skew on assembly, if one or two bind up when the hubs are fitted together. It also looks like Ducati have undershimmed its position, unless that's a consequence of the vertically split cases and the way that the gearbox shafts assemble.

The design simply uses a nut and washer as an end stop on the hub's motion away from the clutch slave. There's no stop on it moving back toward the clutch slave, but that isn't needed. Ducati may have wished to avoid yet another fiddly shim stack and simply allowed a generous assembly clearance.