Cromotor + MethTek Monster 24FET + 20S lipo build

[hjns]

Well, that peak of >80km/h is downhill on a good road without any traffic. In reality I am really chicken

I did a writeup of my conclusions from the temperature readings.

I did a relatively low power commute today. After a day and night raining, the roads are slippery, so I can not do a nice powerride. Therefore, at the end of my commute, temperature was between 70 and 80oC.

At the last traffic light, you can see the temperature drop from 79 to 71oC in about 1 minutes. Driving again for one minute immediately brings the temperature up to 80oC. When I then stop completely, temperature in the first minute goes down from 80 to 72oC. However, once below 70oC, the shedding of heat seems to go much slower, taking 4 minutes to cool down to 60oC, 5 minutes from 60 to 50oC, and 12 minutes to go from 50 to 40oC. Ambient temperature was about 20oC, and I stopped measuring when the motor was 35oC.

Although I did not measure it, it is likely that the temperature drop between 120 and 80oC is linear, like from 130 to 120 and from 80 to 70. Once the wetter is better, I will meausure to prove or disprove that. Therefore, my summary is as follows:

From 130 to 70oC = it takes less than 1 minute or one minute for every 10oC drop.
From 70 to 60oC = 4 minutes
From 60 to 50oC = 5 minutes
From 50 to 40oC = 12 minutes

In conclusion, the higher the motor temperature, the faster the heat shedding, and above 70oC it is almost linear. If the motor winding temperature is more than 120oC, waiting (stopping) for 5 minutes will see a drop in winding temperature of up to a decrease of 50oC down to 70oC. After that it takes much more time to cool down.

Note, these temperature measurements are on a cromotor. The cooling-down profile of another hubmotor is probably very different.

 

[Alan B]

Great data. Faster than I recall with the 9C but that was finger testing. It really should be exponential, like a couple of RC circuits in series. The ambient temperature matters a lot, as does the wetness.

 

[hjns]

Exponential it is. Young Cromotor Knight Alan B the correct thing he says.

Just before I got home, I connected the CA analogger. I kept the CA analogger connected to the CAv3 until winding temperature was ambient. When I got home, ambient outside temperature was around 17oC, inside my wintergarten where I keep the bike it was around 19oC.

Analysis of the data:

• From 101 to 91oC = 2 minutes
• 91 - 81oC = 3 minutes
• 81 - 71oC = 5 minutes
• 71 - 61oC = 9 minutes
• 61 - 51oC = 13 minutes
• 51 - 41oC = 20 minutes
• 41 - 31oC = 34 minutes
• 31 - 21oC = 1.5 hour

Conclusion:
Multiple measurements are needed to get a good idea of the time - temperature relationship. Obviously, the first 5 minutes temperature went down from 101 to 81oC, which is a drop of 20oC. Not too shabby, but 1/3 less than my previous estimation. Given the exponential curve, the 30oC drop in 5 minutes may still hold for the higher temperatures.....

 

[hjns]

This morning's commute:

My conclusion after looking at these graphs and the previous graphs, is that with some hard real-life driving with an input between 2 and 5kW, the cromotor takes about 20 minutes to heat up to 130oC. This is what I expected based on the ebike.ca simulator. In real life, you never are at WOT for a long time. The sim says that at WOT the motor will overheat on a 15% hill within 10 minutes. However, backing the throttle to 80% completely removes the overheating issue. At least in the sim.

At 80% throttle up a 15% hill, the speed looks like 40km/h. Next time (next week), I will try limit my speed to 40km/h maximum, and see if that increases the time to heat up the motor.

 

[binlagin]

The cromotor... such a sexy beast.

Thanks for logging and sharing all this data. Much appreciated!

 

[hjns]

The cromotor... such a sexy beast.

Thanks for logging and sharing all this data. Much appreciated!

Anytime.

Can you ride yours again? Did you replace the axle?

 

[Alan B]

Really excellent data!

 

[binlagin]

Can you ride yours again? Did you replace the axle?

Yes I got my ride going again, but I had to buy a whole new motor. I also unfortunately I damaged my Kiwi dropouts when the axle snapped. It caused the other side of the dropout to bow open so I'm only running a 1500w controller right now It just doesn't do the Cromotor justice, but I'm having fun riding a slow bike fast(40-45km/h)!

I'll probably be making an update to my thread soon, so keep your eyes peeled for that!

I got some new rims I'm just waiting for the rear spokes to come in from JRH to lace it all up My front wheel is already complete and running

 

[hjns]

Slightly different commute, no mountains...

I think there is a nice effect of stopping for a traffic light on the winding temperature.

 

[hjns]

So here I found some pictures that I took during the last month. I only managed to edit and upload them this weekend.


Once I got the cromotor, I removed the cover, removed all the wires from the hole in the bearing and looked at it. The phase wires are thick and very inflexible. The insulation is of the type that will be easily damaged.

Here is a closeup of one of the phase wires. It is twisted with the ends of the motor coil wires and then soaked with solder. The thin insulation is very clearly visible.

I replaced the three phase wires with 10AWG flexible wire. I used some copper wire to keep the ends together, and then soaked the ends with solder.

Then added multiple layers of heatshrink.

Which were then covered by the teflon covers. Observe also the green thremistor lying against the windings.

This is a picture of the test setup. The axle is held in a vice, with the cromotor held horizontally. Phase wires and hall wires come from above, so that it is easy for me to determine the right combo. Then I color-coded the combo that I needed.

Painted the frame and front fork blue metallic.

Lacing the motor. The holes drilled in the flange had some sharp edges where the elbow of the spokes would rest.

Therefore, i took a 3.0mm drill and drilled the hole so that the elbow of the spokes would have better support.

Overview of lacing process.

All nipples are supported by a washer. The Halo SAS rim is great.

Completed wheel. All spokes at similar tension, wheel true to 0.01mm.

Where the wires exit the bearing, they need support from several selfmade C-washers around the axle and a ziptie to prevent rubbing against the brake disc.

Other angle of the C-washers and ziptie. There is now more than enough clearance.

Other angle of the C-washers and ziptie.

Programming the controller at night.

Thinking about better torque arms. I used cardboard to help me thinking. In the end, I used four steel plates of 8mm thickness, two on each side. One plate to fixate against the frame with two 6M bolts + nuts, the other plate to extend the first plate and act as axle fixation with a vertical groove of 10mm. On this second plate, the DrBass torque arms (also approximately 8mm thick) are fixated with another two 6M bolts, threaded into the steel plate. That makes for a surface of about 16mm steel on each side to fixate the axle and deal with acceleration and regen.

 

[hjns]

My current bike without battery box in the triangle (to follow)

Close-up of the frame adjustment with the steel plates. From front to rear:
two bolts allow for the fixation of a steel clamp around the swingarm to fixate the front steel plate to the swingarm.
One bolt fixates the FreeRide carrier to the frame. The front part of the FreeRide is fixated to the black plastic fixation.
Two bolts fixate the front and rear plates with each other. The bolt in the middle actually passes through the original frame, fixating both front and rear steel plates to the frame.


Update September 2013. After a fast downhill route, the original frame broke just between the Middle Bolt and the steel clamp. No accidents, and I was able to drive home slowly with rubbing wheel. Don't do this!


Completely in the rear, there is a 10mm sleeve in the rear steel plate where the axis fits. The DrBass Torque arm is fixated to the steel plate with 2x M6 bolts. A third hole in the DrBass torque arm is visible.

Close-up of the lefthand side. From left to right:
M16x1.5mm Nut - washer (not visible) - DrBass torque arm, fixated with two M6 bolts to the - 8mm steel plate with threaded holes - DIY C-washers to create enough space for the brake caliper which is also fixated to the steel plate with M6 bolts - disc brake - grin disc brake 2.5 mm spacer - hubmotor.

Close-up of the righthand side. From right to left:
M16x1.5mm Nut - washer (not visible) - DrBass torque arm, fixated with two M6 bolts to the - 8mm steel plate with threaded holes - 1x DIY C-washer - freewheel - another washer - hubmotor.

Overview of the rear swingarm. I will remove the CAv3 and the external shunt. That will take care of a LOT of wires.

 

[John in CR]

Very interesting swingarm extension. I've often thought of trying something similar, but always chickened out. Am I correct in understanding that your extension is secured to the original swingarm with a bolt through the original dropouts and a U clamp further up the swingarm? On the right side is the blue part at the motor axle part of the original frame, and if so, what's it for, the derailleur hanger?

If I'm understanding correctly I'll have to copy some of your approach.

John

 

[hillzofvalp]

Which disc is that? I assume 203mm

 

[hjns]

Which disc is that? I assume 203mm

Yes, it is the Avid 203mm that came with my Elixer 7 hydraulics.

 

[hjns]

Very interesting swingarm extension. I've often thought of trying something similar, but always chickened out. Am I correct in understanding that your extension is secured to the original swingarm with a bolt through the original dropouts and a U clamp further up the swingarm? On the right side is the blue part at the motor axle part of the original frame, and if so, what's it for, the derailleur hanger?

If I'm understanding correctly I'll have to copy some of your approach.

John

I will try and draw up a design for you, but I don't have any experience with that. However, yes, your understanding is right. The U clamp is in front, attached with two steel bolts to the steel plate. Then there is the bolt that goes through the original swingarm, near the original dropout. I did not use the original dropout because it is (1) open and (2) too large due to spinning of the axle when the DP420 gave up on me.

The blue parts at the motor axle are the DrBass torque arms. They were initially attached to the swingarm with D420. I then painted the combi blue. Then they fell off, after which I needed to come up with a better idea. Furthermore, I was not happy with widening the swingarm to the necessar 155mm that I now have with my cromotor. The right DrBass TA has indeed a hole for the derailleur.

The idea was to have an extension that would give me 155mm between dropouts, and a very large contact area of the torque arm with the axle. The steel plate is 8mm, the DrBass TA is 9.25mm, so I have a contact area of 17mm on each side.

Now, I do not have long-term experience with this config, and I do plan to add more U-clamps. But it works for now.

 

[Alan B]

U clamps concentrate forces on the tubes significantly and may lead to tube failure. Split blocks machined to the tubing contour would do a better job of distributing forces.

 

[hjns]

U clamps concentrate forces on the tubes significantly and may lead to tube failure. Split blocks machined to the tubing contour would do a better job of distributing forces.

Fully agree, and that is my fear all along. Unfortunately, I don't have the tools at this moment to do any machining. I want to spread the load by adding more U clamps.

 

[Lebowski]

Henk,looking at the very last picture in your post, I have to commend you on the
sheer beauty and elegance of your design, the clean execution, the professionalism it shows, blablabla...

 

[hjns]

... I have to commend you on the sheer beauty and elegance of your design...

Well, you sure have a different view on aesthetics than I.... 8)

This is what I find nicely looking.... unfortunately, it does not have the power that I want. Only about 90 Nm torque....for which you pay more than 4500 CHF.

 

[John in CR]

Thanks Henk, no need for a drawing. Your success gives enough confidence to extend one of my SuperV swingarms. I want to effectively lengthen the vertical support to get the shock attachment further from the pivot like I did with my from scratch longer swingarm, so that will allow me to form a second triangle with supports from the extensions to the higher shock attachment point should allow me to make it really strong and save a few kilos compared to the longer swingarm I made out of angle iron and plywood.

John

 

[hjns]

I want to effectively lengthen the vertical support to get the shock attachment further from the pivot like I did with my from scratch longer swingarm, so that will allow me to form a second triangle with supports from the extensions to the higher shock attachment point should allow me to make it really strong and save a few kilos compared to the longer swingarm I made out of angle iron and plywood.

Show us the results, I am sure my solution can be improved upon.

In the meantime, the heavy rain obliged me to buy a helmet with a visor.

Fits perfectly, wears very light, visor protects perfectly against the rain. Very expensive, though (220 CHF). I get very positive reactions from the public.

 

[hjns]

Cut a cardboard form.

Reshaped the cardboard. Covered the cardboard form with glass fiber on the outside, and very thin wood on the inside. The inside was then lined with a small layer of foam, to protect the lipos from impacts. The outside was taped with Tesa powertape, to make the box watertight. The door on the lefthandside is also taped close, to keep everything watertight. The box press fits inside the triangle, resting on the lower tube. Two velcro bands hold the thing in place, one in front over the upper tube, the other in the rear, under the controller around the seat tube.

Two lipos fit next to each other. I only very seldom touch the box with my right knee. No issues pedaling whatsoever. Not that I pedal for real, its all faux.

All wired up with 8AWG double isolated wires and EC5 5mm gold bullet connectors. Very easy to handle.

I also limited my topspeed to 55kmh, which corresponds to 66% in the controller. I now consistently use around 400-450Wh for a 15km commute (no hills), peak currents up to 80A during acceleration (0-55kmh in 4sec), average use of 25-30Wh/km.

 

[itsmedc65]

Did you take any pictures of the inside with the LiPo in place?

Is there some purpose to the empty triangle space at the top rear or some future use you have in mind for that space? Just curious about it.
I have folded up a box out of coroplast to fit in the main triangle of my old cruiser style frame but, have it touching the frame all the way around except for the space by the chaingaurd.

Also, what is the bike in your post about aesthetics? I like the look and the built in rack. The transparent housing is intersting but, it must have a really small hub motor. I guess it would have to in order to meet Euro or UK limits for retail.

 

[hjns]

Did you take any pictures of the inside with the LiPo in place?

No, not yet.
Is there some purpose to the empty triangle space at the top rear or some future use you have in mind for that space? Just curious about it.

I also intended to have a fixation for a child seat there. However, it's now only aesthetics.
I have folded up a box out of coroplast to fit in the main triangle of my old cruiser style frame but, have it touching the frame all the way around except for the space by the chaingaurd.

Also, what is the bike in your post about aesthetics? I like the look and the built in rack. The transparent housing is intersting but, it must have a really small hub motor. I guess it would have to in order to meet Euro or UK limits for retail.

Dolphin ebike. See http://www.dolphin-ebike.ch/

 

[hjns]

Going on a holiday today. Will be back 15th of October.

Still waiting for Methy to send me a beta supercontroller and for time+good weather to test lebowsky's controller. Also have to fix the wife-bike and change it into a 2WD (rear stock 250W geared FUN motor with PAS, front 9C 2810/Lyen 6FET mini monster controller with thumb throttle, both at 37V), so that she can climb 20% hills as well. Will have to wait a bit more until I'm back....