I could be shooting in the dark here, but per the principle of checking the easy things first: Could it be that the shift return spring is not returning the lever to the center position? You might be able to detect this by getting the rear wheel off the floor and taking it up and down through the gears with your hand, turning the rear wheel to facilitate shifting. You can also tie the clutch lever to the grip so that you're not trying to cycle the engine.
I should say that after some advice and guidance from someone experienced in these matters I actually assembled the gearbox four times. The first time it took me 45 minutes to assemble the gears, shafts, and rollers and camplate. The fourth time it took about 10 minutes and in the interim I thought I had acheived a certain feel for the operation.
After each assembly I spent a fair amount of time rotating the rear tire and shifting through gears, as Mark Z suggested.
Ok, let me go through the list here:
<ul> <li>Mark Z - Foot shifter returns to center - check <li>DMadigan - Layshaft end float - did not check, no idea how to do so after assembly - how can I check that once the inner cover is on? <li>Four/Five speed plungers? the new plungers I put in the outer cover seemed to match the originals - I suppose if they were 5-speed the difference be apparent? <li>Excalibur - by indexing you mean the position of the inner-cover quadrant relative to the camplate during assembly, right? Quite sure this is correct as I get all four gears. Wouldn't I be missing a gear at one end or the other if the indexing was incorrect? <li>Rod - well, I can't be 100% sure until I take it apart again, but as I mentioned, I did practice assembly several times and each time I got a visual on the rollers, so I'm pretty sure they are both in place. Heavy grease seemed to stick them in place quite well. </ul>
I'm probably a fool for thinking so, but I'm pretty sure I've installed all the parts correctly.
I was hoping someone would say "Hey, that happened to me once and it was just ______."
I did notice upon reassembly of the outer cover parts that one of the new plungers "scraped" a bit in it's up and down travel in the bore of the footshift quadrant, whose casting seems quite rough. I did take some emery paper to the bore and it seemed better. Of all the lines of reasoning I gravitate towards there being some kind of problem with those plungers.
Not sure what I'll do next. The bike's been on the table since last October and only came off two days ago. I'm getting "mechanical fatigue". I'm loath to take it all apart again until I have a pretty firm clue about what to look for. I mainly ride gravel roads where there are no stop lights, very few stop signs, and no traffic. I can go practically all day in third gear and fourth gears, which seem to function ok up (always) and down (nearly always).
Thanks everyone for their ideas - I'll stay tuned here while I work up my mojo for another run at it.
I've been plying the back roads the past two days and giving the gear selection train a workout - running up and down through the gears as frequently as possible.
The problem hasn't disappeared, but it is definitely easing.
I'm guessing that the sticky plunger i found during the first assembly is breaking in to it's bore.
Can anyone confirm for me the idea that one of the plungers pushes the quadrant that operates the camplate UP through the gears and the other one pushes the quadrant DOWN through the gears? If that is indeed the case then if the 'downshift' plunger is sticking down a bit it would logically explain the situation.
Your assessment of the quadrant plungers is correct; the upper one pushes the camplate down, the lower one pushes the camplate up.
The plungers however do not HOLD the camplate in position, so a sticky plunger would not explain jumping from first gear into neutral, unless the camplate was not moved all the way into first gear.
The camplate is held in position by the round spring loaded plunger that I think protrudes up from the bottom of the tranny case on Triumphs. But it has to be resting in one of the notches in the edge of the camplate to do its job.
I would say that I'd like to scream, but actually I've been screaming for the past two days. Busted up some crap too because this is really pushing me too far.
Dammit Jim, I'm a carpenter, not a f'ing mechanic!
I researched the archives extensively once I figured out that I could sort all posts by subject. I read John Healy's post on indexing the quadrant and all the other gearbox woes I could find and then took apart the entire transmission and gearbox and started all over.
Since then I've re-assembled everything three times and I really do want to scream. Each time I take it apart I find that everything is in perfect order. I've also carefully compared all the new parts in the gearbox with the old parts and everything seems to match up pretty well. The new camplate is the later type without as many indents as the old one but otherwise matches perfectly. I did find a slight difference between the profiles of the tips of the old and new camplate plungers (new one is a bit more pointy) so I put the old one back just to be sure.
Now I finally see what's happening but I don't know why or how to fix it:
If I assemble the gearbox without tightening the large nut (45 lb torque) that holds on the kickstarter rachet and put on the outer cover I can shift up and down all day long - don't even have to move the mainshaft. Nice crisp clicking up and down and up and down through all the gears.
Then I remove the cover and tighten the nut. To get the nut on all the way I have to gently tap the mainshaft all the way in from the transmission side. Then when I put on the outer cover and try to shift I get one click up or down and that's it.
What am I doing WRONG!!! Should I be tightening the clutch nut before the nut at the gearbox end of the mainshaft?
In 1967 the length of the mainshaft, T914, was changed slightly after eng number DU48144, new number T2436. As a wild guess I'm wondering if an earlier mainshaft was installed in your bike, I don't even know if that's possibles? Just a thought.
Tore down the clutch and gearbox again yesterday afternoon and got everything back together this evening.
Now, on the shop floor, the gears downshift perfectly, but when going up through the gears it almost always ends up in neutral and then it takes a second lift of the shift lever to get it into gear.
See the pattern here? I can get it to shift up or down ok, but not both at the same time.
I checked, and both pawls in the outer cover seem to go up and down freely in their bores.
I compared all new parts (outer gearbox pawls and guide plate, camplate, camplate plunger, and camplate operating quadrant) to originals. The replacement camplate is the later version, with fewer neutral notches, but other than that identical to the old part.
The new camplate plunger has a slightly different profile than the original - the tip is is slightly more pointed than the more rounded original part. Don't think this is an issue because I've tried both plungers and there's no difference.
So if it's not raining tomorrow morning I'll take a test ride to see how it goes on the road.
Took a 35-mile test ride on gravel and asphalt back roads this morning. I found 2nd, 3rd, and 4th gears are generally ok now - I say generally ok because the bike did jump out of 3rd gear once and 2nd gear twice, but at the beginning of the ride and then they seemed to settle down.
The real problem is with 1st gear. It's behaving more or less the way it did before I started. Only worse. Before, the problem went away after four or five minutes of operation. Now it's a full-time feature.
Using the clutch to go from neutral to 1st at a stop sometimes clicks, sometimes doesn't, but either way nearly always fails to go into gear. I.e. I let the clutch out and nothing happens. Subsequent clutching and downshifting usually have no effect, but even if it should go into 1st (either the first time or subsequently) it usually pops out shortly afterwards.
When downshifting to 1st on a roll it usually goes into gear, but sometimes also winds up in the same 'faux' neutral as when at a standstill. At such times, attempting to use the clutch and footshift to get into first are usually futile, but just as at a standstill, if it DOES get into 1st from that 'faux' neutral (Dr. Z's 'waiting place'?), then it almost always jumps out of gear shortly.
Two further observations:
When I get into the 'faux' neutral on a roll, I can match up the engine speed and drop it down into first with the footshift WITHOUT the clutch even when it wouldn't go in while clutching.
When in the 'faux' neutral at a standstill, if I raise the footshift lever until I get to the 'real' neutral position and then lift up into 2nd, I can then clutch and push straight down all the way to 1st and get into gear. Not sure if it would pop out after that - I only discovered that as I returned to the Castle.
I think I'm going to look for some professional help.
(After that I'll try to figure out what's wrong with the bike)
Hermit, I know you don't want to hear this, but to me it still sounds like an indexing issue. It can be very confusing! The original problem is likely to have been caused by worn components, but as you have replaced those, the problems you're having now 'must' be caused by something else. I've been down the same road myself when I started out, I indexed as per the manual and got it wrong time and again. Then I started indexing in first gear, sliding the cover in with the quadrant down until the quadrant meet the camplate, easing back while lifting the quadrant until you feel it mesh with the first tooth of the camplate. That should do it. There are no mysteries to the Triumph gear box, it does NOT have it's own will, it's a very robust design that can tolerate huge tolerances, but indexing needs to be spot on.
Stein Roger - first of all thanks for the reassurances that I'm not dealing with some kind of dark, supernatural entity here. It's also encouraging to hear that I'm dealing with something that, theoretically at least, is within my grasp.
Re indexing: I've assembled the gear cluster into the gearbox 6 times now and put on the inner cover at LEAST a couple of dozen times. I've used both the method in the manual and the 1st gear indexing methods- I like the 1st gear method better.
Once after indexing the WS manual way and once after indexing the 1st gear way, I pulled off the outer cover, used a screwdriver to position the quadrant so that the middle lug lines up with the mainshaft as per the WS manual, and both times when I pulled off the inner cover I could see that the camplate was exactly in the position it was supposed to be in according to the manual - i.e. 'arms' straight out and 'head' straight up. For this reason I want to believe that indexing is not the problem, but nevertheless acknowledge that it could very well be.
I'm going to play a bit with the old parts and at some point - not today! not tomorrow! - i will pull of the covers and try indexing once again.
Thanks for chiming in - I appreciate all clues - they will eventually all add up to a solution.
I wonder if anyone say definitively what is the effect of being one tooth off when indexing? Or let me put it this way - does the fact that the gearbox goes into all four gears eliminate the possibility of incorrect indexing?
Bruce, firstly, I forgot to comment on your blog, it's great!
I can't argue with what you have done but when I disregard what you say and simply look at the evidence as presented, it still says "indexing" to me. Yes you can have wrong indexing and still be able to "sort of" find all gears.
There are a couple of things I'm not sure was covered in previous posts; did you actually establish whether your gear plungers are of the correct 4 speed type? The 5 speed type has a larger cut-away. Did you establish whether the quadrant is the correct way around? That would seriously mess up your indexing... Are the shift fork rollers in place? I'm sure they are, but just a thought. I glued them in place once, I was so p1$$ed, and the box changed as sweet as any. Later forks didn't have loose rollers and the 5 speed forks had solid pins.
It drives you up the wall by now, I know how you feel, but again, it worked before and you WILL have it working again!
...I rebuilt several pre unit gboxes and couple late T140 ones. If jumps out of gear (mostly is third or fourth) is the indexing plunger or a thicker gasket in the inner cover. Sounds stupid but if you turn the index plunger (the one that s under the box)many times bites better and that solves the problem. If not you ll need to shim the gears. So if you rebuild the gears the right way, put the camplate like the manual and do not have too much worn out parts; you can solve the stuff like Im saying.
SR - I hear what you're saying - when I thought about it I realized that I really do NOT have four gears - just three (2-3-4), and SOMETIMES first.
So, are "gear plungers are of the correct 4 speed type?" When I compare the old plungers in the outer cover they look identical to the new ones except that the old ones are very polished on the face that slides on the guide plate.
"quadrant is the correct way around?" Hmmmm. I think it can only go in one way and have the hole for the split pin match the one in the shaft. Or am I missing something here?
I'm as sure as I can be that the rollers are in ok. All but one of the seven times I've pulled out the gear clusters and shafts the rollers were still in place. The one time a roller wound up on the bottom of the case was after I'd filled the case with gear lube and I'm pretty sure it slipped off the layshaft gear selector's peg as I pulled everything out.
One interesting thing about the rollers. The new ones that SRM sent me wouldn't fit into the channels of the camplate they sent - they were too big. The old ones were kinda loose so I ordered another set from MAPCycle and they seemed to fit just about right.
The outside diameters went like this:
Original rollers (65,000mi).4370-.4372" SRM rollers: .464-.467" MAPCycle rollers: .4395"
So there are variations in these parts and the wishful thinking part of my brain would like to believe that all these problems will disappear after a couple hundred miles when the parts all wear in together and play nice. The logical part of my brain leans towards the indexing as you say.
As I read the posts from Triless and Reverb I get confused as to whether it's best for the camplate plunger to glide easily over the edge of the camplate or dig in. Both scenarios have some sense to them.
My old camplate plunger had a more rounded profile than the new one. (See www.hermit.cc/tmc/bb/2015/wtd_2/index.htm for photo showing both new and old). The edge of the old camplate was seriously polished and grooved.
I've tried both old and new camplate plungers and it didn't seem to make any difference.
Bruce, great pictures, everything looks new. The only thing that stuck out to me is the new camplate, appears to be larger in diameter than the original one. got out my calipers and measured the width of the plates in several different places, based on your pictures, the new one is larger. Now I'm thinking that the clearance between the new cam plate and the index plunger holder has been reduced and could there be a possibility that the camplate is rubbing on the index plunger holder and or possibly the plunger is bottoming out in the holder because the spring is being compressed completely causing shifting issues. There is an easy way to check this and that is to unscrew the plunger holder from the case add a couple washers which will create more space for the spring and index plunger to operate. Also in 1968, 1969 a heavy duty spring #57-1604,was available to help alleviate gears popping out. There were a number of changes with the camplate, plunger and spring holder during these yrs. In 1970 a washer was added to the plunger spring holder. The point on the index plunger, the rounded one made for an easier shift but was More prone to jumping out of gear then the pointed one. Since the early 60's I have always assembled the transmission in first gear, i know the manual says differently but for me it's much easier.
Interesting. But I've never experienced any problems with jumping out of gear on any four or five speed Triumph gearboxes in which I've eased the point on the indexing plunger. If the gears selected and jumped out under load,I would have thought it would be an issue with the dogs! The thought that the plunger housing could be fouling the camplate is a good point, though. That would be worth closer investigation.
Hi Sam - you're right - the new camplate looks bigger in the photos, but I think it's a trompe l'oeil due to the camera lens/angle. I compared the two very closely when the new one arrived because I had been unaware of the changes in the later part and hadn't expected the new one to look so different. I concluded they were the same part - a little different, but the same. it's a good point, though, and I've made note of it for when and if I contact the vendor.
I'm pretty sure there was no interference - it spun freely 360 degrees when I put it in, and even after I put in the camplate plunger there was always clearance - we can see that in a couple of the photos I posted yesterday.
Yes, the stronger spring is what I had previously installed and I did not change it this time, although it might have been good to have done so - I renewed all the other springs.
Concerning indexing using first gear - this afternoon I started mocking up what I hope will become my Triumph gearbox indexing training simulator.
Looking at the pictures of your mock-up the quadrant looks to be installed and positioned correctly. This transmission has been basically bullet proof for years and to me simple to work on. The engineers at Triumph started making a number of changes starting in 68 with the shifting components that went into the 70's with the leaf spring Replacing the plunger. To be exact there were 7 changes between 68 and 70 to the camplate, index plunger, spring, holder and a fiber washer was added. I've worked on a lot of these over the years and never had the problems your having. I just put back together a 67 T120R transmission that is exactly like yours and it works. I have no idea what's causing your problems. From a mechanical perspective there should'nt be any.