Axle Thread and Motor Restoration

I've bought a used motor for cheap, the seller reckoned it was working fine before coming off his bike. I have some doubts and am trying to decide if it's worth working on. It is a large DD motor, I've opened the side cover and the windings look ok, but the laminations and rotor magnets appear to have seized together due to corrosion. I can only turn the axle with a wrench, and there is a grinding sound. What's the best way to clean up the corrosion to get it spinning again?

I'm currently on the road and only have limited access to tools and no Multimeter. I'll test for phase to housing resistance when I get home.

The Drive Side axle is damaged, it appears the seller tried to bash the stator out of the rotor and has damaged the axle in the process. Should I be using a die to repair this or perhaps a thread file? Should I first file the axle back to remove the "bottle neck" end?


The Non Drive side axle also has damaged threads. What's the best way to fix this up so I can use an axle nut again? 14mm thread die are hard to come by, most sets seem to end at M12?


Thanks

cloudy said:

14mm thread die are hard to come by, most sets seem to end at M12?

Click to expand...

You can purchase a M14 die via Ebay for $10-15 but you will need to know the pitch as well.
I suspect that may be the least of your problems.


(UPDATE) Found this reference but I can not guarantee its accuracy ...
Axle Nuts sizes :
... M16 X 1.5
... M14 X 1.5
... M12 X 1.25

cloudy said:

It is a large DD motor, I've opened the side cover and the windings look ok, but the laminations and rotor magnets appear to have seized together due to corrosion. I can only turn the axle with a wrench, and there is a grinding sound. What's the best way to clean up the corrosion to get it spinning again?

Click to expand...

Kingfish has a thread that may help
https://endless-sphere.com/forums/viewtopic.php?t=34135

cloudy said:

The Non Drive side axle also has damaged threads. What's the best way to fix this up so I can use an axle nut again? 14mm thread die are hard to come by, most sets seem to end at M12?
20220717_084354[1].jpg

Thanks

Click to expand...

Amazon for tap and die $9.50, cheap but worked ok for me for cleaning up the threads on the axle and bolt.
https://www.amazon.com/gp/product/B089YKPSBG/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

https://endless-sphere.com/forums/viewtopic.php?f=6&t=97945&start=175#p1697613

Thanks everyone for the replies, good thing this is a spare and I still have a working bike to ride. I'll get stuck into it and maybe learn a few things in the process.

file/Taper the bottle neck for the die to fit on it.

cloudy said:

The Drive Side axle is damaged, it appears the seller tried to bash the stator out of the rotor and has damaged the axle in the process.

Click to expand...

A non hardened steel axle?

Comrade said:

cloudy said:

The Drive Side axle is damaged, it appears the seller tried to bash the stator out of the rotor and has damaged the axle in the process.

Click to expand...

A non hardened steel axle?

Click to expand...

I don't think I've encountered a hardened axle in a hub motor yet.

Patience, a wire brush and a small triangle file would be how I would attempt to clean up threads after removing mushrooming. There is a fair bit of clearance between a nut and a bolt and it just has to go on. Damaged axles for a vehicle will involve additional factors around safety.

Before you get jiggy with a flat file, know that there are thread files that you can use to repair threads without horking up their profile too badly.

Chalo said:

Before you get jiggy with a flat file, know that there are thread files that you can use to repair threads without horking up their profile too badly.

Click to expand...

Ya but $9.49 for the tap and die so you can clean up the threads on the nut too. You just have to be really careful. A thread chasing die would be around $15, by itself. Of course, getting the dies and the file is probably the best option, since more tools is more better.

Chalo said:

I don't think I've encountered a hardened axle in a hub motor yet.

Click to expand...

What are they usually? I meant hardened as in higher carbon content than a nail. I never thought a quality axle could be mushroomed like that.

Comrade said:

Chalo said:

I don't think I've encountered a hardened axle in a hub motor yet.

Click to expand...

What are they usually? I meant hardened as in higher carbon content than a nail. I never thought a quality axle could be mushroomed like that.

Click to expand...

Oh I reckon they're some kind of alloy steel usually, like 1144 or 4140 or the Chinese equivalent. I just don't think they're heat treated to any higher hardness than they had when they were machined to size. Typically that would be "Condition N" or "normalized", which is not quite fully annealed, but close.

If I'm mistaken about this, it means that the axles are made from really really mild steel, and are heat treated until they can pass for decently strong steel in normalized condition. I doubt that, because it's probably cheaper to use alloy steel and skip the heat treatment.

Anyway, if you can lightly file an axle and the file bites in instead of skating across the surface, it's not hardened (or not hardenable).

Hard is brittle, a little give is better.

I finally got around to opening this motor up. It seems it was not turning properly due to friction caused by magnetic rust dust in the air gap between stator and rotor. It seems to be cleaning up ok. Must have had water in there at some point...

Once clean, I was planning to paint it with high temperature wood stove paint. Keen to hear what people think of this.

The phase wires are not looking great, so was going to replace those too. Ive had a lot of success running sensorless. So much so that I no longer plug the halls in on my working bike. Are there any disadvantages of running sensorless? Trying to decide if it's even worth pulling hall wires through the axle, or just use the space for fatter phase wires.

Can anyone tell me how to determine the number of turns in this winding? Looks like maybe 5, but hard to say from the bundled copper.

cloudy said:

Once clean, I was planning to paint it with high temperature wood stove paint. Keen to hear what people think of this.

Click to expand...

Check out that thread I linked in my other post in this thread.

Are there any disadvantages of running sensorless?

Click to expand...

Not if it works for you with the controller you have under the riding conditions you have.

Some sensorless controllers in some situations are terrible at starting up from a stop, some of them are better. If you're always pedalling from a stop or even pedalling first and then engaging the motor it's likely to work fine, since the problem is usually just getting started from a stop. .

If it has been water logged then you may also want to change the bearings out as well.

So I've had some success cleaning up the rust on the stator and magnets. I've coated the stator laminations with a product called penetrol which will hopefully reduce future corrosion. I will eventually coat it with electrical enamel. I've ordered a M14x1.5 Die and Tap kit.

I spent some time reading the link from Amber Wolf/Kingfisher, the main take homes from this and Justin's excellent presentation:

• Rust is the most common issue with Motor Hubs, sealing motor is nearly impossible so a better approach may be a small drain hole (or large vents). I've opted for a small vent hole at the outside perimeter which I will try point at the ground when not riding the bike.

• Coating with electrical varnish after cleaning rust off may improve corrosion resistance.

• Alignment of motor covers with magnet ring may be important and the motor may not turn if re-assembled with different alignment. So mark your motor covers before removal to maintain factory alignment.

One of the Bearings is very rough and probably cooked as predicted by LewTwo.

I've re-assembled the motor and have run it no-load in a bench vice using my 40A controller and 58.2V battery voltage. I measured no load speed at 683RPM using an old bicycle computer and sensor magnet. Using motor simulator, this seems to be a 5T motor, as the no load speed is close to "9C 2705". The magnets measure 35mm wide, can anyone tell me which motor in the standard motor simulator list is the best representation of this?

I don't really want to go much faster than 50km/h so will likely need to use a 20" wheel to make this fast 5T motor work for me. Any tips on building such a small wheel with the large hub would be great. I've built several road and MTB wheels over the years so know enough to get myself into trouble. Is radial spoke pattern the way to go for such a small wheel? I've seen quite a few radial DD wheels in the forums.

cloudy said:

I don't really want to go much faster than 50km/h so will likely need to use a 20" wheel to make this fast 5T motor work for me. Any tips on building such a small wheel with the large hub would be great. I've built several road and MTB wheels over the years so know enough to get myself into trouble. Is radial spoke pattern the way to go for such a small wheel? I've seen quite a few radial DD wheels in the forums.

Click to expand...

I made a spread sheet to calculate the theoretical speed for my Giant Revive using a mid-drive ... then I added a section to compare a hub motor drive. Originally it only had "miles per hour" but now I have added "kilometers per hour" to the hub motor section. Perhaps it may prove useful. It just happens to be setup for a 20x2.25 tire at the moment. A couple of points:

1) Speed is determined by the actual circumference of the tire and there can be a world of difference in the circumference due to different tire profiles on a given rim. (The controlling circumference parameter for this spreadsheet is in cell $A$8)

2) Your loaded speed is always going to be somewhat lower than the no-load speed (unless going down hill). The "Scratch Col" has the parameters you measured on your motor.

You could always run a larger rim at a lower voltage to limit the top speed. For example a 24 inch rim at nominal 36 volts (which actually tops out at about 42 volts) ... (the last three columns are volts):

Thanks for sharing your spreadsheet LewTwo. I'm still mulling over what wheel size to build. The smaller 20" wheel will perform best and will give the greatest acceleration for a given controller and wheel torque. I've had issues with my home made torque arms working loose when using regen with 26" wheels, so would be good to make the most of a limited wheel torque I guess. The 20 inch wheel bikes do look less like a "normal" pedal bikes and I'm not keen on this, it stands out and attracts attention. I still need to work on the axle threads and possibly phase wires so early days still...

I was able to clean up the threads using a flat file and M14x1.5 thread die. The die was started on effectively blank axle end, hence I was concerned that the die would misalign with the remaining threads and cause further damage. This did not happen and it worked perfectly. I chamfered the axle ends before starting the die. The long axle is a real luxury on this motor, I'll be able to use deeper nuts and thicker torque arms.

I didn't buy specialist thread file linked above, but this may have been a better tool for the job in hindsight. The file could potentially align with existing threads better.

The axle steel is very soft.






M14x1.5 die

Looks good!
Hard metal is brittle and don't give much. It's all a trade off. You want it hard with some give.