Vertical Wobble in Clam Shell Hub

[2door]

The amount of hub defects shown in the video isn't enough to cause the problem you're having. I suspect your sprocket adapter is the cause. It is has either been machined incorrectly or it is the wrong size for your bub.

I'm not a fan of shimming an oversized sprocket adapter to get it to fit the hub. It can cause clamping problems whereby the adapter can slip. This can cause spoke damage when it comes in contact with them and a wobble.

If your tests determine the adapter is at fault my suggestion would be to trash it and buy a good quality adapter to replace it. Check your hub diamter with a caliper and order an adapter sized to fit. Measurements must be accurate down to the thousands of an inch.

And yes, winter came to us today too. A 35 degree drop in temperature in less than two hours. Blue skys to heavy clouds, rain that turned to snow. So much for our beautiful Indian Summer.

Tom

 

[Msiert]

Finally got some time and bought a dial gauge and a digital caliper at Harbor Freight to run some tests on the rear sprocket and clam shell hub adapter.

Today I check the run out of the rear sprocket and found it had a run out of .016 I'm thinking that's not good since car brake rotors are considered out of spec at greater then .002 run out?

I also used my zip tie mounted to the frame method to check for the front to back movement of the sprocket. That tested pretty good in that there was only a strong 1/32" of movement.

 

[Msiert]

More Tests: To check the rear wheel.

I checked the rear wheel for run out (using the rim) and got .023

After this I loosen the drive chain by unbolting and sliding the engine back. I then re-checked the wheel run out and found that it got reduced quite a bit to .018 which is pretty close to what the run out of on the sprocket was.

With the difference in run outs measurement between tension on the drive sprocket chain vs no tension I thought that maybe some of the spokes are a little loose, that when there was tension on the hub shell from clam shell adapter it was causing some flexing of the hub shell thus effecting the run out of the wheel rim via the spokes?

This made me re-test the drive sprocket for run out and found it was basically the same as before except that it increased by .001

Think the next thing to do is losen the wheel and remove the chains and re-test the run out of the sprocket...maybe see if I can get the wheel back in straight and re-test the run out of the wheel.

After that....pull the clam shell hub adapter and use the digital caliper to measure the bike's hub shell and then measure the diam of the clam shell hub adapter's bore.

While the wheel is out I'll take it to a bike shop to true it up and decide if I need to buy a new clam shell adapter and sprocket before putting it all back together.

 

[skitchfish]

Mr. Msiert, When you were checking the vertical true of the rear wheel itself did you have it off the bike with the tire off? If the sprocket and the wheel were close to equal values vertical truing of the wheel would definitely help.

I'm not sure if you have a seam in your rim as you can only see it from the inside. That seem is the culprit and challenge of balancing vertical and lateral true

 

[skitchfish]

Let me regress. If your hub is in the center of your wheel your bikes drive train will be much smoother even if the hub is bolted to the drop outs

 

[Msiert]

Mr. Msiert, When you were checking the vertical true of the rear wheel itself did you have it off the bike with the tire off? If the sprocket and the wheel were close to equal values vertical truing of the wheel would definitely help.

I'm not sure if you have a seam in your rim as you can only see it from the inside. That seem is the culprit and challenge of balancing vertical and lateral true

Today I was checking for run out of the sprocket and the wheel while they were mounted on the bike with the tire mounted. I do have a seam in the rim and you can see it from the outside.

I was thinking the same thing in that with the sprocket and wheel only being off .001 from each other in run out that maybe I did not get the wheel installed straight?

What is the proper way to install the back wheel with horizontal drop outs. Been searching for the answer and what little I found I didn't do it right. There is quite a lot of play (left, right, up and down) when the wheel is sitting loose in the horizontal drop outs.

Read that you should do it while the bike is on the ground (I can understand the need to have the bike on the ground to force the axle bolts to the top of the channel) and to snug up (not fully tighten) the peddle side first while getting the correct tension on the peddle chain then pull the left side back till the wheel is straight....never fully tighten one nut then the other. do them in increments. But doing it by eyeing though it seems like you could easily be off by a couple of hundreds of an inch.

Let me regress. If your hub is in the center of your wheel your bikes drive train will be much smoother even if the hub is bolted to the drop outs

When you true a wheel are you also centering the hub shell?

 

[Msiert]

Pulled off the drive sprocket and clam shell adapter and found the following:

Final test show that my drive sprocket and adapter need to be replaced.

Drive Sprocket:
The drive sprocket is warped and when you tighten it down to the clam shell adapter it makes it worst.

When I laid just the sprocket down on a piece of glass one end was up from the glass when I mounted the hub adapter to it....it made it much worst.

The drive sprocket is a 48 tooth with a center hole of 2.4790 inch (62.96 mm) for the coaster brake arm to fit through. The new clam shell adapter will need the bore to work with a hub shell with a diameter between 1.7120 inches (43.49 mm) and 1.7110 inches (43.39 mm)

Clam Shell Adapter:
The hole in the clam shell adapter measures 1.7275 inches (43.88 mm) in one direction (left to right) and 1.6910 inches (42.96mm) in the other direction (up and down)

Hub Shell:
The hub shell diameter is (.10 mm) larger on the right side then the left side. The max diameter is 1.7120 inches (43.49 mm) on the drive sprocket side and 1.7110 inches (43.39 mm) on the peddle sprocket side. The center diameter is 1.7105 (43.45 mm).

Wheel Rim run out:
Without the drive sprocket on I mounted the rear wheel with the bike on a repair stand. I put a jack under the wheel and put some upwards pressure on it then I pulled it back as straight as I could eye it. Then I put the dial indicator on the rim and pulled out on the wheel in the low spots and improved the run out from .017 to .012.

Funny thing is the low spot and ramp up sections remained the same. Took the wheel to the bike store to true it.

 

[Msiert]

Yesterday Afternoon:

Pick up the wheel from the bike shop (3 hrs turn around time) and the mechanic said that he improved the run out from .012 to .005 and found a number of loose spokes. So last night I reinstalled the adapter, sprocket and the wheel back on the bike and this morning got the chain re-tension and engine bolted down. I rotated the sprocket 180 degrees from what it was on the hub adapter in hopes to straighten out the warp sprocket a bit.

This Morning...Temporary Fix:

Well yesterday I found out that my hub size (1.7120 inches) not a common size for clam shell hub adapters so I will either have to order a stock size (1.5 inches) and bore it out or have a custom one made. I have left messages at Jake's and Custom Motored Bicycles to see if they can help me with a new sprocket and hub adapter to fit my hub size.


While I'm waiting I went ahead and shimmed the sprocket by installing cut off pieces of feeler gauge strips between the sprocket and the arms of hub adapter by using the dial indicator. I shimmed each of the three arms with a different size shim .010 shim, .006 shim and a .004 shim in that order clockwise.

Then end result was a reduction of run out from .012 to .005 and there is a lot less change in chain tension. Might be able to get it better with different size shims but think I will leave it for now.

 

[2door]

Good for you. We're glad you're making headway.

Tom

 

[skitchfish]

2door, My bad if it sounded like misinformation about the concentricity of the sprocket hub and wheel. I apologize for being unclear but while he had it apart thought i would help him realize having his hub centered in his rim would keep the outside of his wheel from acting like a cam.

 

[2door]

Skit,
I didn't feel it was 'disinformation'. Not in the least. I understood your reasoning.
I'm just glad the problem is known and we were able to offer some advice that will
get his bike on the road without further issues. Thank you for taking the time to offer help. That's why this forum was established.

Tom

 

[Msiert]

Yes skitchfish thank you for the help. I hope I wasn't being to obsessed with this wobble with all the run out tests and digital caliper measurements.

I talked to the supplier of the sprocket today and of course he must of thought I was nuts but he is sending me a new sprocket free of charge and will check it to make sure it's not warped. When I asked him if his hub adapter would accept any other sprocket he said no, just his sprocket.

I'm going to keep the clam shell hub adapter for now.

I'm not sure if a (new) clam shell hub adapter from him would make much difference with the wheel's hub shell diameter being not the same in the full length (.10 mm larger diameter on the right side then the left side). The existing hub adapter has been rock solid in that it clamps well and has never slipped.

I'll try the replacement sprocket before forking over a bunch of money for a custom 1.7 inch bore clam shell hub adapter and new sprocket from a different supplier.

 

[Msiert]

Got the new drive sprocket Monday and it laid perfectly flat on the glass slab. I got it installed today and tested the chain for even tension while turning the wheel.

Revised 11-27: re-check the deflection, this time sitting on the floor for a straight on view and blocking the drive sprocket and just measured the distance of deflection while pushing down. .
There was a slit difference with the chain tension going from loose to tighter at the 180 degree points of the sprocket but not much (a strong 5/8" at the loose spot 1/2" at the tight spot) I'm calling it good.

I don't think I will ever get it perfect. Between the wheel having .005 run out and getting the wheel in perfectly straight makes it dang near impossible to get a .000 run out on the sprocket even if the wheel had .000 run out.

But I do have two questions.

I read several different articles on the proper tension of the chain. Some say 1/2" to 1" others say 1/2" to 3/4" and others say 1/2" defection is too tight and not to go tighter then 3/4" ? To error on the safe side which measurements should I use to set the tight spot of the chain to?

The other question is how do I measure the defection. In the some forms they measure the deflection by finding the distance between pushing down and pulling up the chain. Other sources just measure the distance by pushing the chain down. Which way is correct?

 

[2door]

Here's how I test chain tension. With the bike at rest and clutch engaged roll it forward until you feel the resistance of the piston coming into the compression stroke. The upper or return portion of the chain will go slack at that point. The slack is measured at about the center of the chain run. I'm one who prefers 1/2" to 3/4" slack. This, like many other aspects of motorized bicycle engineering is a matter of opinion based on experience. My experience has shown the 1/2 to 3/4 to work well. Others will have differing opinions based on what has worked for them. Very few things are carved in stone.

If you're using a centrifugal clutch you can block the drive sprocket so it can't rotate in lieu of using a manual clutch. Moving the bike forward will have the same effect on the chain. That is it will go slack on the top run.

Best advice is to experiment and find what works best for YOU. Good luck.

Tom

 

[maniac57]

I'm with Tom. 1/2 to 3/4 is about right. If you need more slack then keep working on centering.
Check it at the loose point and make sure it doesn't bind at the tight point and you're good.

 

[Msiert]

How do you measure the defection?

The GruBee manual says "The chain should move about 1/2" up and down" measuring this way doubles the deflection measurement vs just pushing the chain down and measuring.

Which way is right?