MKIII trouble shifting while engine is running

Hi all,

I have an issue finding neutral while the engine is running as well as a great amount of force required to shift from first up to second. A quick pop from second down into first just so will find a false neutral that doesn't register on the dash light, and once the engine is off it slips into gear. When the engine is off its generally very easy to shift through. And when the bike is at speed the issue of force to shift through the gears is greatly reduced. I've dug through past threads and the primary suggestions I've seen are to adjust the pawl spring, check primary chain tension, and to look for grooves cut into the clutch basket housing. Well, I checked and adjusted the pawl spring for a bit more gap on each side; that did not improve the issue. Since I have an MKIII I should be able to rule out primary chain tension since its running on an automatic tensioner, unless the oil I'm using (20w50) could be too thick and tensioning it up too much? And I just checked the clutch basket and everything looks pretty smooth (see pics). Does anyone have experience with finding a solution to this particular issue, or has ideas for other areas to check?

Thanks,
Anthony

How is your drive chain tension? Too tight a drive chain pulls the tranny which pulls the primary...

it should be 3/4" to 1" (19 to 25 mm) of up and down movement in the centre of the chain run WITH THE RIDER SEATED (and wheel on ground )

Thanks Gortnipper, I’ll do a rider seated test of the chain and report back.

gortnipper said:

How is your drive chain tension? Too tight a drive chain pulls the tranny which pulls the primary...

it should be 3/4" to 1" (19 to 25 mm) of up and down movement in the centre of the chain run WITH THE RIDER SEATED (and wheel on ground )

Click to expand...

Thanks Gortnipper, I’ll do a rider seated test of the chain and report back.

Anvill said:

A quick pop from second down into first just so will find a false neutral that doesn't register on the dash light,

Click to expand...

Sounds like the detent plunger isn't locating in the camplate notch (possibly due to play in the cross-shaft mechanism?) so rolls back into 2nd gear (or can jump to 1st)?

Anvill said:

When the engine is off its generally very easy to shift through. And when the bike is at speed the issue of force to shift through the gears is greatly reduced.

Click to expand...

Both of which suggest there's clutch plate or other 'drag'? Are the (5) friction plates the original bronze type?

Anvill said:

Well, I checked and adjusted the pawl spring for a bit more gap on each side; that did not improve the issue.

Click to expand...

I wouldn't expect it to as setting the pawl spring is more to do with missed gear selection, not neutral selection.

Anvill said:

Since I have an MKIII I should be able to rule out primary chain tension since its running on an automatic tensioner,

Click to expand...

Yes.

Anvill said:

unless the oil I'm using (20w50) could be too thick and tensioning it up too much?

Click to expand...

I haven't found 20W/50 to cause any problems. The bronze plates, however, can cause clutch drag unless cleaned periodically.
The is a dynodave clutch pushrod seal which should prevent gear oil contamination of the plates.

You may find it easier to select neutral (either from 1st or 2nd) just before you roll to a complete stop.

L.A.B. said:

Sounds like the detent plunger isn't locating in the camplate notch (possibly due to play in the cross-shaft mechanism?) so rolls back into 2nd gear (or can jump to 1st)?

Click to expand...

It's hard to know what's acceptable for the cross shaft mechanism. The rubber connector seems to be designed for a little play, but those parts are original to the bike so its definitely not out of the question that they may be worn. Is there any particular method to assess the state of the cross shaft mechanism? Again, everything seems to work fine until the bike is running.

L.A.B. said:

Both of which suggest there's clutch plate or other 'drag'? Are the (5) friction plates the original bronze type?

Click to expand...

These are my plates, I couldn't fit them all in the photo but yes 5 total. They appear to be a brake pad type material bonded to aluminum, not the bronze type. This may be getting too much into personal preference but is there a generally preferred type? As you can see in the photo they came out with a bit of oil fouling, do you think this may be enough to cause the drag problems?

I do have a CNW version of the pushrod seal which can be seen in the first photo posted. It looked pretty dry upon disassembly earlier today, so the oil on the plates is likely from the primary case (7oz 20w50). The basket did have a bit of oil pooled in the bottom grooves, but I'm not sure if this would be out of the norm as its not really a sealed assembly.

As for chain tension, I measured with rider on and came to 1/2" play downwards movement from rest and 5/8" upward movement from rest. So assuming the measurement should be 3/4" to 1" in EACH direction and not total, I may be running the chain a little tight. I will try loosening it up a bit.

Thanks,
Anthony

I found the neutral switch plunger adjustment was critical to be able to find neutral. I also found it kinda fiddly to get right.

napanorton said:

I found the neutral switch plunger adjustment was critical to be able to find neutral. I also found it kinda fiddly to get right.

Click to expand...

It’s definitely a pain. I readjusted it for good measure today while I had the gearbox open, to the bare minimum depth while still lighting up. I assume that’s the target, correct? I haven’t started the bike yet with that adjustment made so with any luck it will help.

Anthony

Anvill said:

It's hard to know what's acceptable for the cross shaft mechanism. The rubber connector seems to be designed for a little play, but those parts are original to the bike so its definitely not out of the question that they may be worn. Is there any particular method to assess the state of the cross shaft mechanism? Again, everything seems to work fine until the bike is running.

These are my plates, I couldn't fit them all in the photo but yes 5 total. They appear to be a brake pad type material bonded to aluminum, not the bronze type. This may be getting too much into personal preference but is there a generally preferred type? As you can see in the photo they came out with a bit of oil fouling, do you think this may be enough to cause the drag problems?

I do have a CNW version of the pushrod seal which can be seen in the first photo posted. It looked pretty dry upon disassembly earlier today, so the oil on the plates is likely from the primary case (7oz 20w50). The basket did have a bit of oil pooled in the bottom grooves, but I'm not sure if this would be out of the norm as its not really a sealed assembly.

As for chain tension, I measured with rider on and came to 1/2" play downwards movement from rest and 5/8" upward movement from rest. So assuming the measurement should be 3/4" to 1" in EACH direction and not total, I may be running the chain a little tight. I will try loosening it up a bit.

Thanks,
Anthony

Click to expand...

Those are older style Barnett plates.(newer ones have more segments) Good plates. Wash them in gasoline, along with the steel plates before you reinstall them.

1" is TOTAL, not each way, so your chain was only very slightly loose.

Anvill said:

The rubber connector seems to be designed for a little play,

Click to expand...

The "rubber connector" should be a (plastic) cover...

...over a splined connector and have only minimal play...

Anvill said:

These are my plates, I couldn't fit them all in the photo but yes 5 total. They appear to be a brake pad type material bonded to aluminum, not the bronze type.This may be getting too much into personal preference but is there a generally preferred type? As you can see in the photo they came out with a bit of oil fouling, do you think this may be enough to cause the drag problems?

Click to expand...

Barnett friction plates as gortnipper says and yes, clean off any fouling.

napanorton said:

I found the neutral switch plunger adjustment was critical to be able to find neutral. I also found it kinda fiddly to get right.

Click to expand...

Anvill said:

It’s definitely a pain. I readjusted it for good measure today while I had the gearbox open, to the bare minimum depth while still lighting up. I assume that’s the target, correct?

Click to expand...

Yes, that could help to improve neutral selection and be screwed in just enough for it to operate reliably.

Thanks for the help guys.

I'll try to dial in the chain tension a little better but it doesn't seem like its the cause of my symptoms.

I did take a closer look at the cross shaft--while it's not exactly a tight fit in the primary outer cover bushing, I imagine thats likely because the shaft has to install at an angle while the primary cover is removed/installed straight out. When attached at the gearbox the crossover shaft seems like a good fit, though it again has about 1/4" of wiggle when unsupported at the primary side, which I attribute to the connector allowing a bit of play. I imagine when its captured on both ends the cross shaft is fairly stable. Unless what I'm describing sounds out of norm to you guys?

The clutch parts have now been soaked in gasoline, which did loosen up quite a bit of crud.

Hopefully I was experiencing clutch drag and trouble finding neutral because of the neutral switch. Once I get everything back together I will let you know how it goes.

Anthony

My mk3 cross shaft is as you describe, with no problems. I did suffer slip at over 5,000rpm and a sticky feel from / to neutral. Cured by replacing my old clutch plates.

You checked the plates for flatness, correct?

concours said:

You checked the plates for flatness, correct?

Click to expand...

I didn’t put them each on a granite slab but I checked that they all stacked in various orders with zero gap. It didn’t seem like an obvious source of trouble but it’s a good idea, I’ll add this to my list of things to be a little more thorough with next time I have the plates out.

Anthony

Please don't use gasoline as a solvent . There are better ones out there . Use gloves of course.
The MK 111 model does not need primary chain adjustment.
I use sintered bronze plates . CNW clutch center , 20-50 oil and a main shaft clutch seal .
Your photo shows potential notching of the center.

But then you can add it back to your jerry can and just burn it in your lawn mower! ;-)

I might be a little late on this thread but have a good bit of experience with Commando clutches. One of the main service operations in the Norton shop I worked in was clutch cleaning. Probably paid a good part of my wages. OP's problem description sound like it needs a clutch cleaning, which I see from the photos he has done. He's also checked out the notching of the hub problem. Hopefully, the hardened hubs are still available. We replaced many soft hubs.

MKIII's are unique and troublesome because of all the screws holding the primary cover on and because they used the bronze plates rather than fiber ones. Not only do they drag when dirty but they also slip under full throttle and sometimes less; news to no one I'm sure. Our solutions ranged from substituting a Barnett plate for every other bronze plate to cutting radial grooves in the bronze plates to throw the oil out. Barnett plates tended to drag more and the bronze plate of course slipped. I settled on only three Barnett plates as the limit for dragging. The radial grooves were marginally effective as I remember and certainly time consuming to do; expensive, even at shop rates of $30/hr. in 1990's dollars.

Before I came, the owner/chief mechanic had settled on ATF as lube in the primary case. I did some further experimenting and found that type F fluid as used in Fords had better grip but tended to grab on take up. GM type was smoother on take up but had a greater tendency to slip. Either had longer service life between cleanings than motor oil. With MKIII units changing the primary oil frequently is needed to save the expensive sprag clutches, which stop working after the swarf wears them.

Perhaps the most effective cure for this whole problem is a seal to keep transmission oil from running into the clutch. Back in the day I made one for my own MKIII from a felt washer, a steel washer and a spring to hold the whole affair against the end of the gearbox input shaft. It has proven quite effective over the years. I understand you can buy such a seal ready-made these days. Not surprisingly my boss did not agree to adding this to the bikes in for service. Also not parking on the side stand for long periods helps.

Thomasa said:

MKIII's are unique and troublesome because of all the screws holding the primary cover on and because they used the bronze plates rather than fiber ones.

Click to expand...

Mk3s not unique in having bronze plates as they were used in all 850 models and also late 750 (from 212278).

Thomasa said:

Not only do they drag when dirty but they also slip under full throttle and sometimes less; news to no one I'm sure. Our solutions ranged from substituting a Barnett plate for every other bronze plate to cutting radial grooves in the bronze plates to throw the oil out. Barnett plates tended to drag more and the bronze plate of course slipped. I settled on only three Barnett plates as the limit for dragging.

Click to expand...

Anvill's friction plates appear to be Barnett.

Thomasa said:

Before I came, the owner/chief mechanic had settled on ATF as lube in the primary case. I did some further experimenting and found that type F fluid as used in Fords had better grip but tended to grab on take up. GM type was smoother on take up but had a greater tendency to slip. Either had longer service life between cleanings than motor oil. With MKIII units changing the primary oil frequently is needed to save the expensive sprag clutches, which stop working after the swarf wears them.

Click to expand...

However, Mk3 owners have discovered ATF is not always compatible with the Mk3's hydraulic primary chain tensioner and prefer to use (the factory recommended 20w/50) engine oil.

Thomasa said:

Perhaps the most effective cure for this whole problem is a seal to keep transmission oil from running into the clutch.

I understand you can buy such a seal ready-made these days.

Click to expand...

Yes, and is there in Anvill's first picture (dynodave seal).

If you don't entirely cure your clutch drag issue , I have found with my 750 is to give it a rev and select neutral as the revs come down , it seems to unload the box enough to let it get to neutral . Cheers .

Considering all potential problems have been checked, then redo the clutch adjustment from fresh, only turn the screw back half a turn and lock off, adjust cable free play, go for ride and then reduce the free play at the lever until it slips nicely into neutral.

Update:

I needed to do a full strip down of the primary to sort an alternator issue and while I was at it decided to separate the clutch center and backplate. Once I got them apart it became apparent that the center was indeed quite worn. Something about the way light reflected on it in the full assembly had made the wear look more like light polish marks than full on grooves.

I also noticed that the teeth on my old Barnett plates that interface with the hub looked a little worn. So, I bit the bullet and installed a hardened center from CNW along with a new set of plates. The new plates were much more difficult to get on and off on account of not having enough tolerance to really tilt. This seemed like a very good thing, as I would not be surprised if the clearance on the old set allowed them to rattle around enough to accelerate the wear to the hub.


I had the opportunity to briefly fire it up earlier this afternoon and had absolutely no trouble shifting to neutral at idle. So, if anyone reading this is experiencing similar problems and are having trouble finding the source of the issue, it may be worth pulling off the clutch center to have a closer look. Thanks for your help everyone,

Anthony