Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[2020_SRS_Commuter]

On my SR/S the rear sprocket has an offset. I assume every bike is going to have this to some degree but I wondered how mine was in relation to others. Did I get a "bad" one? IDK.

Viewing the rear wheel from the left side, and imagining the air fill valve as the end of the hour hand on an analog clock, when the valve is at the 2:30 position, the tension is the lowest. At 8:30, the highest. The difference between the two is 6-7hz on the Gates Carbon App.

When I ride away from a stop, the whine of the motor as I go from about 10 to 35 mph has a little variance that coincides with the tension changing. Waaaaaaaa waaaaaaaaa waaaaaaa waaaaa waaaaaa waaaaa waaaa waaa waaa waa waa waa wawawawawawawawawwawa then at about 35 mph or so its becomes a steady pitch, at least to my ears.

Is that typical? Common? Defective? Any input is appreciated. It works, I cant complain about that. I find it annoying though... shades of OCD.

[electrictwowheeler]

My SR/S belt makes some noise at certain speeds below 25 mph but it's more like resonant vibrations. Speed up or slow down a bit and it stops. Above 25 it is a smooth low whine. When I check the belt tension I do it 3 times and turn the wheel a quarter turn each time and the readings do vary but I use the average. On bikes I have had with chains there is always a tight spot it seems. I would suggest setting the belt to the low end of the tension spec and seeing if that helps. I almost always wear silicone ear plugs for the wind and traffic noise. That will definitely make things quieter. Ha ha

[talon]

I measured my tension MANY times--compressed swingarm, not compressed, from bottom side of belt, from top side of belt (halfway in between pulleys of course), tool facing both ways, rotating wheel backwards first, forwards first, and every 90 degrees and charted how those all play into the measurement. These results are of around 50 measurements.

Which way the tool faces: ~2kg difference
Direction of rotation before measurement: ~2kg difference
Offset: 21kg minimum, 27kg average, 35kg max (fairly reliably)

I know the relative offset angle now. I can now measure both perpendiculars to the resultant vector of the two high(er) measurements to get the best indication of median tension. That may be worded incorrectly, I'm tired lol

[2020_SRS_Commuter]

Next time I service the rear wheel, probably to replace the rear tire, I'm going to try improving this.... maybe removing the rear sprocket and rotating it two bolt positions, that's a thought.  But to start, I'm going to loosen the sprocket bolts to observe how closely the bolts hold the sprocket centered.

I was surprised when I replaced my front disks recently how much play can be had with the disk bolts and the holes in the disk they go through. To get the disks as centered as possible you have to put a little art into it, look closely, and line it up precisely, not just slapping it together and torqueing down. Perhaps the sprocket has a similar situation.

[talon]

Yes I'm really surprised they didn't go with countersunk holes and fasteners to self-align/center like automotive wheels typically use. Perhaps the only reason they didn't was that it would sacrifice strength? Honestly I'd love to see a shaft drive for a Zero now, damn whatever mechanical losses.

[Mooseman]

Honestly I'd love to see a shaft drive for a Zero now, damn whatever mechanical losses.

Same here. They gave us the shaft before, might as well get a real one 

[2020_SRS_Commuter]

Needed to replace a rear tire so while I had the wheel off, I took this up again.
You can see from the screen grab I posted here what centers and aligns the pulley sprocket to the wheel is not really the bolts or holes, but a milled edge on the sprocket seat on the wheel that interfaces with a milled edge on the sprocket.

The fit on mine was beautiful. Just the slightest need to press it over these edges and its immovable off the wheel's axis. The only way I could adjust the sprocket with respect to the wheel was to rotate it a bit, and of course you want it rotated so the bolts go through.

So in short, there was no way to adjust the sprocket's alignment with respect to the wheel.

HOWEVER

You can rotate the sprocket and re-attach it in the hopes that wherever the runout exists on your particular combination, it can be mitigated by turning the sprocket to one of its five possible positions. I chose to to rotate it two positions clockwise when viewed from the sprocket side of the wheel. And... boom. Much improved. I dont hear any warbling in sound as it speeds up from a stop now. Just a steady increasing whine.

[NetPro]

I'd like to add a couple of points to this thread: I spent considerable time looking into this a couple of months ago and these are my findings.
Bear in mind that my bike was one of the first production units made, I got a dealer demo, more than a couple of years ago when they came out. Your bike might be slightly different (not likely but possible).

It turns out that the rear sprocket can be adjusted by up to 15 to 20 mils (sorry European audience, my micrometers are in Inch only) if you work hard at it.
At first, when I removed the screws and felt the sprocket's fit into the wheel, I thought it would not allow for much adjustment, because the "shoulders" machined on the sprockets -as mentioned in a previous post- and the way they fit on the wheel matching shoulders.

However, being the machinist wannabe I am, I decided to try and was pleasantly surprised at the results.

It goes without saying, the rear wheel has to be off the floor and the belt removed.

To cut to the chase, the 2 more difficult things to accomplish are:

1) Secure a micrometer to a solid point to monitor the sprocket offset. See photo below.
I have a mount that attached perfectly where the frameslider (bobbin) goes. Changed the tip on the micrometer to one with a soft round profile to work well on the sprocket's "teeth" as you rotate the wheel.

2) Build a small "expander" device to fit between the sprocket's inside edge and the wheel's hub. Will attach a picture later. This allowed me to raise the sprocket a bit next to the screws where the micrometer indicated a low spot.

On those areas where the micrometer indicated "high" I would hit the sprocket with a piece of wood (to avoid marring it) and a small hammer.

The key here is to loosen the bolts to a point where it is kind of tight but still allows the sprocket to move just a few mils up or down.
Once you find that sweet spot, torque them bolts securely, I have put about 500 miles on the bike since the adjustment and it is as good as day one.

It took a lot of time to get the sprocket centered this way but let me tell you the good part: It is amazingly quiet now from the noise described in other posts, from which I too, was suffering big time.
I cannot believe the difference this made to my ride as it now feels silky smooth. I hated that whine that I could not get rid of regardless of how much tension I adjusted the belt with.

It can be frustrating that you are almost there, you want to adjust just a tiny bit more and then the whole thing gets out of wack again and you have to start over.
I am certainly glad I was able to get my sprocket within a couple of mils, on the spot right above each bolt.

It is worth noting that even after all of this, the machining of these sprockets is so poor that the micrometer would detect high and lows points between the bolts (just a few mils but there nonetheless) but it does not appear to be picked up by the belt as you ride the bike.

I had to use a milling machine to make the "expander" I mentioned above but someone with a good imagination might be able to come up with another way to apply pressure on the inside of the sprocket to make it go up just enough to get it centered.
Will post a picture of it next with a short description.

[NetPro]

This is the tool I devised to force the sprocket up where needed.
It is made of a phenolic resin 3/4 of an inch in thickness. Picked this material to avoid scratching the wheel's hub.
Using a 2 and 1/2 inch hole saw (if I remember correctly), cut the bottom part so that it would fit on the wheel's hub without sliding in any direction.

Very simple device, really, just a hole on one side to insert a screw that when turned by the nut on it, applies upwards force to get the sprocket a bit higher if needed.
I will post another picture of the device for clarity and take a break.

[NetPro]

The top piece is machined in aluminum and it has a groove the width of the sprocket's lip, so it does not rotate when you turn the bold to apply force to the sprocket.

[2020_SRS_Commuter]

So when you are making adjustments, what is adjusted? Is the metal on the wheel casting being deformed, or on the sprocket, or both, or am I not understanding?

One of the things that bugged me about this situation besides the warbling noise was that the range of adjustment in the belt tension was narrowed.... I had to fit the variance with the range limits, if you know what I mean. 

The good news for those without your skills or time to spend with it is they get 5 tries at a decent matching. You might not even have to remove the wheel to rotate the sprocket to the other positions to try. IDK.

[NetPro]

So when you are making adjustments, what is adjusted? Is the metal on the wheel casting being deformed, or on the sprocket, or both, or am I not understanding?

One of the things that bugged me about this situation besides the warbling noise was that the range of adjustment in the belt tension was narrowed.... I had to fit the variance with the range limits, if you know what I mean. 

The good news for those without your skills or time to spend with it is they get 5 tries at a decent matching. You might not even have to remove the wheel to rotate the sprocket to the other positions to try. IDK.

No metal was deformed in my adjustments: I would never go that far and I didn't have to go that far anyways.
There is a certain amount of play that allows for a small adjustment of the sprocket relative to the wheel: you don't feel it by hand but with a small force in the right direction, it was all that my bike needed to get the sprocket nearly perfectly aligned with the wheel, at least as measured right above each of the 5 bolts holding the sprocket in place.

I think luck played a part in my success but no complaints from me when I installed the belt, set it to 67 hz and rotated the wheel one fifth of a turn and got 67 hz five times in a row, as I made a complete turn of the wheel.
I measured when each of the 5 bolts was at the 12 o'clock position and I couldn't believe it because previous to this adjustment, it would vary greatly (about 9 hz between lowest and highest tension).
And this was causing the variation in pitch, the noise and even vibration that were driving me crazy and are now gone.

You are correct in that all you need to do to rotate the sprocket one or two bolt holes at the time is to remove the bolts, move the sprocket and reattach the bolts.

I tried three different combinations before I used my technique and I did not see a reduction on the sprocket's offset: The micrometer was going up and down 15 to 20 mils in all three tests so I decided to take it one step further.
I do believe the possibility is there for an improvement just by moving the sprocket one, two or three holes but without a micrometer is just a guessing game.
I say this because the manufacturing tolerances for the parts is so poor that it just might get closer to a truly centered position just by doing this but it would be pure luck.

[talon]

I have a 0.030" test indicator (+/-0.015") and my 2016 SR sprockets' powdercoating itself makes any value I get from it unusable. Do you have a rolling test point on your dial indicator?

[Demoni]

I have a 0.030" test indicator (+/-0.015") and my 2016 SR sprockets' powdercoating itself makes any value I get from it unusable. Do you have a rolling test point on your dial indicator?

In the first photo NetPro posted you can see the dial indicator probe is behind the outer face of the pulley. He is measuring the tooth just to the right of the bolt, I suspect he took measurements at each of the bolt locations.

There is a certain amount of play that allows for a small adjustment of the sprocket relative to the wheel: you don't feel it by hand but with a small force in the right direction, it was all that my bike needed to get the sprocket nearly perfectly aligned with the wheel, at least as measured right above each of the 5 bolts holding the sprocket in place.

Big thumbs up it's probably the truest running rear pulley out there. Great technique too! Gave me flashbacks of squaring stock on a Leblond 4 jaw.

-----
Looks like it's just the stacking effect of part tolerance. Is the pulley bolt centric or are there keying features it references on the wheel?

[NetPro]

I have a 0.030" test indicator (+/-0.015") and my 2016 SR sprockets' powdercoating itself makes any value I get from it unusable. Do you have a rolling test point on your dial indicator?

I don’t have a rolling test point but I was able to get accurate readings by changing the point that comes standard with the micrometer with a “soft round” one.
Many years ago I bought on eBay an inexpensive kit of about 12 points of different shapes and one of them has a nice, soft, round tip that did the trick.

It is worth noting that the shape of this point allowed me to position the micrometer actually riding  on the sprocket’s “teeth”, not on the powdercoated thin edge of the sprocket.

The key here is to only push down on the micrometer about 0.040 or 0.050 when resting dead center on one of the teeth: so when you rotate the wheel, the micrometer goes up-and-down by this amount as it jumps from teeth to teeth.
Because of the soft curve, when the next tooth comes under, it pushes up the point smoothly and you can compare the readings at different points on the sprocket.
Not the most elegant solution but it worked for me and I was surprised to find out that the point didn’t even show a scratch on it. I believe it is made of carbide or some other extra hard alloy because it looked like it was never used after I finished.