Garth S

Before one tosses big $$ for an intermediate plate in the '78 - '86.5 dual disc clutches when doing a clutch job after finding the IP chewed up ... there is hope

I recently finished a job on an '86.5 that suffered a badly dragging clutch: the rear disc was warped, causing it to bind on the shaft, and constantly rub on the PP and IP surfaces when disengaged.

The IP flat springs and adjusters were in good shape; however, the rear surface was badly grooved & polished mirror smooth.

Long story, short version .... sent the IP off with the flywheel to my local machine shop to see what he could do. The set up was a bit tricky, but he nicely trued both sides with minimal grind - the IP was as new!

I preset the 'H' adjusters 1mm above the rear surface prior to installation, which turned out to be a good guesstimate as the IP was some 40 thou (1mm) thinner ... release & take up were perfect without further adjustment.

No pics, but the surface was identical to the flywheel done for my car last summer - shown in post 18
here.

Botton line, I haven't previously read of grinding the IP .... perhaps it's common practice that escaped my backwoods niche, but if not - they can be saved.

Perhaps this can add to the current chatter about release bearings ( re NLA futures), and to one of the best dual disc clutch threads ever.

Mark Anderson

Glad to hear it can be done. I've always worried the the pressure plate could make up the difference.
BTW I have a new intermediate plate minus ring gear for $295.

mark kibort

as mark said, the pressure plate has to make up the difference, but with a flywheel grind, it doesnt matter. dont preset the intermediate plate to anything , even 1mm. that might not be enough or too much. It can easily drag, and that could have been your issue before. before you put the lower cover on , look at the gaps . they shoud be just about .5mm to .75mm gap from a visual check. very very slight gap visually. its just enough to allow the front disc to pull away from the flywheel. too much of a gap and you risk it pulling back so far that the rear discs touch the pressure plate. this is the direction that the H tabs move toward naturally due to vibration and rotational forces on the flat springs under High rpm and fast rpm changes with the clutch in
mk

James Bailey

There is some reduction of the clamping force as the stack of two discs and the intermediate ring is less due to the material removed. Not much of an issue with new discs but might lead to some possible slipping as they wear down. As Kibort mentions when you grind the flywheel you also grind the mounting surface which keeps the clamping force the same. If one were concerned it is possible to have the pressure plate spring fingers setup with more clamping force. I was thinking about the double disc clutch issues and how the dealers and most shops years ago tended to replace EVERYTHING on every clutch job now with many hobbiests doing their own work, many things are often reused as "good enough". So you have clutch short shafts with worn splines, intermediate plates with worn adjuster rivets, flywheels with grooves or cut thin, pressure plates which have weak spring fingers ,etc. etc. Often, the only real reason these things are not being replaced is because they "cost too much" .....

Garth S

My first concern also ... although thought that the new disc thickness would compensate: the potential issue of slippage would be deferred until the discs wear, as noted below.

I went this route, for as mentioned, the springs and adjusters were as new .... so what a waste to toss it.

Were the new clutch assembly to have slipped, my next step would be to have the flywheel step ground ( as in several BMW clutches) so as to leave the working surface on a plateau higher than the bolt ring annulus a distance equal to the total thickness ground from the IP ..... that would work as long as the adjusters has sufficient movement to compensate, and it appears they have that range.

To Mark K .... no dispute on the sensitivity of setting the adjusters - my point is that I've done enough of them to have a 'feel' for the presetting of an IP prior to installation: the final installed measure is what counts.