gearbox rebuild issue, no selecting the lower gears

Gear selection issue, when I tighten mainshaft nut, loose ability to select lower gears (2nd, neutral and 1st) and same with outer cover off and manually moving quadrant. Gearbox has all new internals, bushes, gears, kickstart, sleeve gear (AN) , bearings etc, except selector forks, main and layshaft which have checked for strainghtness. Roller bearing on lay, layshaft endfloat 0.006. after shimming. Gears, bushes slide on shafts, only 1st main bit tight on spline but does not extend past boss. Have pulled apart a few times and all appears OK and quadrant timed, selecting all camplate tracks/gear positions with inner cover on, gears out and in. Any ideas? it must be related to the mainshaft but would expect that when tightening mainshaft nut would only pull shaft away from within sleeve gear a small amount and it still rotates OK with no stiff spots.

What "new" parts did you use?

Hi, does your shafts turn by hand , when all nuts tighten ?

I have found that tolerances can cause a new gear installed in mainshaft stack to introduce interference. In my case, I replaced 1st main only, and it caused 2nd main to be trapped when mainshaft nut was tightened. New 1st main gear measured 0.002" more than original. I lapped 0.002" off the new gear and then the mainshaft spun freely with a tight nut.

You replaced all the gears, so you may have a different manifestation. I suggest comparing the dimensions of your new gears with your original gears, if you have them.

NickZ said:

I have found that tolerances can cause a new gear installed in mainshaft stack to introduce interference. In my case, I replaced 1st main only, and it caused 2nd main to be trapped when mainshaft nut was tightened. New 1st main gear measured 0.002" more than original. I lapped 0.002" off the new gear and then the mainshaft spun freely with a tight nut.

You replaced all the gears, so you may have a different manifestation. I suggest comparing the dimensions of your new gears with your original gears, if you have them.

Click to expand...

I had the same , first gear mainshaft in a close ratio 4 speed (all gears were new) about 2 thou too long, it was a friend's of mine gearbox and I was the fitter , it took me some times , there was another gremlin with the new parts , but don't remember................

concours said:

What "new" parts did you use?

Some Andover Norton transmission parts quality escapes (2015)

I had the opportunity to repair my failed transmission (still working, but 2nd had a funny "thump, thump, thump" noise. Disassembly revealed a LOT of damage.... I set to work inspecting (the case, sleeve gear and first gear lay were the only usable moving parts) and ordering parts. When I...

www.accessnorton.com

View attachment 78911

Click to expand...

I used all andover parts, new gears, bushes, sleeve gear, kickstart shaft but used original lay and mainshaft, selectors and selector shaft. Can turn mainshaft shaft by hand, slight drag but may be as used assembly grease. Will borrow larger micrometer to check all thicknesses of MS gears compared to originals and look at lapping if different, get rid of grease to see if frees up.

Where are the shims ? ( or machined aluminium/bronze shim washer in my 'case)

bsasteve said:

I used all andover parts, new gears, bushes, sleeve gear, kickstart shaft but used original lay and mainshaft, selectors and selector shaft. Can turn mainshaft shaft by hand, slight drag but may be as used assembly grease. Will borrow larger micrometer to check all thicknesses of MS gears compared to originals and look at lapping if different, get rid of grease to see if frees up.

Click to expand...

While you are at it, maybe compare bush measurements also.

Time Warp said:

Where are the shims ? ( or machined aluminium/bronze shim washer in my 'case)

Click to expand...

Shim is steel old isolastic shim with centre enlarged to clear radius on kickstart shaft, located on lay assembly against kickstart steel bush in inner cover under stops

I just rebuilt my gearbox and had a heck of the time getting to go through the gears (on the bench). My experiences...

- The gearbox does not easily go through the gears on the bench. It needs a little backwards resistance that normally would be provided by the rear tire so you have to counter rotate or hold firm the sleeve gear when you go through the gears

- I bought a new selector fork shaft from AN and it bound the selector forks. I went back and used the original shaft. They were both bent BTW, but i guess my old one was bent better than the new one. Reading the above link on AN QC issues now has me wondering about the 2nd gear set i replaced (did that cause any of my problems?)

- Timing the quandrant is tricky. First line up the quadrant selector ratchet thingy with the window on the inner cover (equal top and bottom) than assemble everything and see if you still have full engagement of all gears. You may have to go back and readjust.

- I did not have the camplate o-ring fully seated so it was slightly pushing the campate inwards which was binding the selector forks

- I'm still not sure if i have the gearchange ratchet spring bent correctly.

I detailed my struggles in my build thread (posts 185-219)


Good luck

I keep an old clutch center just for turning the mainshaft by hand. It is a lot easier than gripping the mainshaft.

bsasteve said:

Shim is steel old isolastic shim with centre enlarged to clear radius on kickstart shaft, located on lay assembly against kickstart steel bush in inner cover under stops

Click to expand...

What I did not like about the Isolastic shim was it might drop down on the radius with an enlarged ID (so machined a sliding fit with relief for the radius shim washer from Al/Bronze)
The box ratcheted through all gears on the bench easily.

The only way I can see a problem (without giving it to much thought) with straight shift forks and the selector components in order is an over shift on layshaft Second.

As far as the main shaft First (fixed) and Second (free wheeling) are both captivated by shoulders on the shaft including at the bearing so tightening the nut should do little (unless there was bind due to dimension which would be unlikely and noticeable)

After I shimmed I checked the shift on second layshaft to make sure it would engage (swing in and out of without bind even though the first lay gear is up against a shoulder)

The box should shift nicely on the bench with the detent backed out only enough to work.

If the box selects the lower gears with the shims removed that might be the problem area.
I did not go as low as 0.006" end float deeming it to little (in my case and did two gear boxes side by side the same )

The JC video is handy also. (if you had not seen it)

Time Warp said:

As far as the main shaft First (fixed) and Second (free wheeling) are both captivated by shoulders on the shaft including at the bearing so tightening the nut should do little (unless there was bind due to dimension which would be unlikely and noticeable)

Click to expand...

What you describe as unlikely actually happened in my case, and it was not noticeable. to the naked eye.
One of the "shoulders" you refer to is the bearing (as you noted), and when the nut is tightened there can be a bind of 1st & 2nd between the bearing (1st) and mainshaft shoulder (2nd). The new 1st main that I purchased was 0.002" larger than the original, and that difference was enough to cause the bind. I am not convinced that this is 'unlikely' when replacing gears.

Awesome feedback and ideas for solving my problem, thanks, I have been working away from home and more comming up and then end of June I will start again figgering it out in a consecutive way. I do have a new selector fork shaft, so will check that too along with the gears shoulders and bushes, deten and everything sugested. I assembled gearbox without gears with outer cover first too and checked the cam plate timing, and if looking at the pictures in some assembly guides, easy to be a tooth off as angle of shot for lineup is up for interpretation. Lots to try, thanks all again

Regarding bench shifting. You do have to check that the g/b shifts up and down, better to catch an issues before you have to do another R&R.

My experience is that new shafts vary one dimension and/or another and that the splines have sharp edges and usually need to be worked such that the sliding gears move freely.

I recommend installing fully populated shafts one at a time to check for binding with full torque. If they spin free any binding, or locking, will be the result of misaligned/bent shift forks or a camplate/quadrant out of time, usually in 1st or 4th gear, one of which you, probably, won't be selectable.

Rebuilt gearboxes are going to be stiff. Keeping an old basket/hub, as mentioned earlier in this thread, makes it easier to spin the shafts, I also recommend mounting the counter shaft sprocket which will allow you to spin both shafts.

However.

Before you start adding another "what if" keep in mind that at a 1000 RPM idle the shafts spin quite a bit faster and make gear selection a non-issue if you have assembled the g/b correctly.

The real fly-in-the-ointment is that cursed pawl spring, it must have clearance on both ends to the pawl or you will get stuck in a gear. Slamming the shifter may get you going, but will cost you much to fix than new boots.

Best.

What RoadScholar said and once I think the shafts, gears, and bushings are working properly, how I verify and finish:

When I rebuild them, I leave the outer cover, clutch mechanism, and kickstart return spring off but otherwise completely build without shims including tightening the case nuts fully and the main shaft nut just tight enough to ensure that the main shaft is properly located. I also install the sprocket so I can turn the sleeve gear easily and so I know that it is properly located. Once I'm satisfied that the shifting is correct and the gears are not binding I check the end float on the kickstart. Unless it is more than 10 thou I don't shim and when a do shim, I make sure there is still end float that I can feel, probably around 5 thou. Since the kickstart return spring is not installed, you have to remember to keep it parked (fully clockwise) while testing.