MP5 clicking problem

[Socalbiker]

Whenever I 'switch' from braking to acceleration or accel to brake I hear and feel a clicking sound. Have a MP5 with 2 torque arms. I've checked that all spokes are tight, the axle bolt is tight, the torque arms are tight and setup correctly.

To make sure I had the torque arms setup correctly I took off the axle bolts, put a wrench on the axle flat and tried to rotate CW and CCW, it did not move.

I can't replicate with bike on a stand, need the weight and momentum of riding which makes it difficult (impossible) to pinpoint sound and look for movement or flex. The click seems to resonate thru the bike frame, but I'm not 100% positive.

Anybody else have this problem or suggestions?

thanks

[Kneedeep]

What is your frame and setup? It could be a suspension pivot, shock or even a brake caliper/disc. I would sit still brakes on and buck it back and forth no power and see if you can recreate the symptom there. Thus mimicking the weight transfer  of a brake to accelerate event.

[Socalbiker]

Hi Kneedeep,
solid CroMo frame, caliper brakes. Think I tried this and no click, but I'll check again.  I really thought it was axle rotating but it seems the torque arms are doing good at holding solid (at least using a wrench, not sure how that compares to a kW of power)

[Bikemad]

Could it be movement between the axle and the stator like I experienced on a Smart Pie?

Unfortunately this would be more difficult to see on a rear Magic Pie as it is covered by the brake disc mounting/fan assembly.

Alan

[GM Canada]

Loose spokes can make a click sound. Pluck them to listen for loose ones.

[Socalbiker]

Gary: spokes are all tight.
Bikemad: how'd you check for play between stator and axles?

[Bikemad]

Bikemad: how'd you check for play between stator and axles?

I was fortunate enough to be able to see it moving on my front Smart Pie, but if you temporarily disconnect the brake lever wires so that the regen doesn't work, you should find that the noise will disappear if it was the stator moving on the axle (or the axle moving within the torque arms).

Without activating regen, the stator and axle will be twisted in the opposite direction to the wheel rotation under load, therefore any play in the drive key/slot between the axle and the stator should automatically be taken up on the first application of power. Subsequent applications of power should not result in a clonk unless the axle is so loose within the stator (or the frame dropouts) that the residual drag of the magnets is able to rotate the stator (or axle) in the opposite direction when power is stopped while the wheel is still moving.

The axles are usually a very tight press fit inside the stators, but if the hole in the centre of the stator is slightly oversize, and/or the axle outer diameter is slightly undersize, this can result in a very loose fit which would allow slight movement to occur each time the torque direction changes. Over time this can cause wear between the axle and the stator, and also in the keyway slots and drive key.

I really thought it was axle rotating but it seems the torque arms are doing good at holding solid (at least using a wrench, not sure how that compares to a kW of power)

I once carried out a very simple stall test using a set of digital luggage scales attached by a piece of rope tied to the outer diameter of my tyre as shown here:



My modified Pie was drawing 65.94 Amps @ 25.6V (1,694 Watts) and although the controller cut the power before the luggage scales had time to lock the reading, I was able to read a reading of at least 28kgs/61Lbs of force on a couple of the tests, which I worked out to be around 93Nm of torque!
If this force is exerted on a 10mm deep dropout the lateral force acting upon the end of the dropouts by the edges of the flats is probably around 950kg (475 kg on each end of the axle).
To visualise this amount of torque, if you were to place a 130mm long 10mm spanner onto the flats of the axle, you would have to apply a force of 73kg (160lbs or 11.5 stone) on the opposite end of the spanner to produce 93Nm of torque!



My modified controller would draw more than 65 Amps under load, so the motor produced more torque at low speed than a standard Magic Pie, but it should at least give you a rough idea of just how much torque is being transmitted through the axle.

Alan
 

[Socalbiker]

Wow, thanks for such a informative response. Very clever way to figure out what the torque is/was. Clonk is a better descriptive of the sound I hear.

I actually didn't need to disconnect the brake, whenever the battery is fresh off a 90 or 100% charge the regen brake doesn't work. Thanks to your input I now have problem isolated to movement between axle & stator or my wrench check didn't apply enough torque.

I suspect it's the axle & stator. Sadly this means I have two choices. 1) continue using until I break the key or cause some other failure 2) disable the regen brake, which was a big part of my choosing a hub and not a mid-drive motor.

You don't happen to have a link to drawings and instructions on how to disassemble and repair?

[Bikemad]

If you remove the right hand side cover and bearing, you should be able to closely inspect the keyway slot and drive key for signs of movement.
If you mount the axle in a vice using soft jaws or scrap pieces of aluminium etc. to protect the axle you can try to turn the stator by hand to see if anything moves where it shouldn't.

If you don't have a suitable vice, remount the wheel in the frame and see if the stator moves in relation to the frame. If there is noticeable movement between the axle and the stator, you may want to remove the axle for further inspection.

Hopefully it won't be a failed stator assembly like Andrew had with his.

To remove the axle from the stator, you will need to remove the circlip shown below, as well as removing the controller from the opposite side.



If the axle is loose enough inside the stator to clonk under acceleration and braking, it should be relatively easily to drift out using a wooden mallet or a hammer with a block of wood placed on the end of the axle to prevent damage to the threads etc.

My MPIII axle was extremely tight and did not have any movement whatsoever (unlike the Smart Pie in the video) and I had to place a steel tube over the axle up inside the controller housing to press directly against the centre of the stator, so I could drift the axle through it without damaging the stator assembly.

I ended up threading two nuts onto the exposed end of the axle and locked them together with the outer face of the outer nut flush with the end of the axle to give a larger area for the mallet to hit, as I had to whack it really hard before it eventually started to drift out.
A hydraulic press would have been the best tool to use, but I had to improvise as I don't have one of those.

Your MP5 axle should look similar to the "After" version of my MPIII axle shown in the picture below:



You may be able to simply refit your axle using some Red Loctite Threadlocker to hopefully eliminate the unwanted movement.

Alan
 

[Socalbiker]

thanks for all the detailed information. It will probably be awhile until I have time to try this. Plus I'll have to improvise some missing tools. In the meantime I plan to change the controller setting for how aggressive my regen braking is, mayb I can find a happy medium.