RC experiments

[full-throttle]

I'd like to build a reasonably light electric trailbike. My first attempt with a hub motor didn't quite cut it as it lacked small bump sensitivity and power. Apparently, both of this issues can be solved with the use of a RC motor instead. Just need to wire it up and bolt it on.. Start tinkering!

Experiment 1 - Interfacing to an ebike throttle
[youtube]hlJR9A6VWTY[/youtube]

 

[adrian_sm]

Ooooo. This should be good.

So will this be going on the Avanti D-eight. Will the motor be mounted on the swing arm? Maybe above the shock.

Looking forward to seeing this one evolve.

 

[liveforphysics]

If that 4.2amps I saw for a moment is the steady state no-load current at 48v, you gotta back the timing down on that ESC in a big way. That's 200w of heat going into that motor just as no-load.

That's like the same mount of heat it should be making for dumping 2kw into it

 

[recumpence]

Also, that looks like a Kontronic Power-Jazz ESC. If so, I would recommend not using it for human propulsion. All 3 of the people who have used those (myself included) have blown them in VERY short order. That is an expensive controller to blow.

I would recommend a Castle HV140 or HV160 instead.

Matt

 

[liveforphysics]

Also, that looks like a Kontronic Power-Jazz ESC. If so, I would recommend not using it for human propulsion. All 3 of the people who have used those (myself included) have blown them in VERY short order. That is an expensive controller to blow.

I would recommend a Castle HV140 or HV160 instead.

Matt

+1 agreed.

That was my shortest lived controller, and the most expensive.

 

[full-throttle]

Yeah, the no-load current didn't impress me the slightest. The controller came with a programming card, I'll have a look later if it defaults to advanced timing.

Thanks for the info on Jazz, I was not planning to use it on the bike - its there to aid with Hall placement and getting the benchmark RPM and current numbers. I'd like to re-terminate the motor in star as well at some stage.

D-8 would be a nice chassis for it.

 

[full-throttle]

If that 4.2amps I saw for a moment is the steady state no-load current at 48v, you gotta back the timing down on that ESC in a big way. That's 200w of heat going into that motor just as no-load.

That's like the same mount of heat it should be making for dumping 2kw into it

The skirt bearing is generating all that extra drag. Slowing down the ESC didn't make much difference. What did help though was rewiring the motor in star (really easy with the Scorpion) 1.75A with the Kv of 87.

BTW, 80-100 HXT drew over 10A @ 50V at the start, windings cold, skirt bearing hot. After a minute the current gradually went down to 6A, bearing still hot.

Makes you really appreciate the efficiency of the Astro inrunners, much easier to cool down too and fully sealed by design. $$$ though..

 

[full-throttle]

Haven't updated this for a while..

1) Rewired Turnigy 80-100 in star. Was not as easy as the Scorpion, but still - no brain surgery either. The no-load current dropped from 6A to just under 3A @50V (both with the skirt bearing)

2) Fitted internal Hall sensors. Now the motor runs from an ebike controller with the Kv of ~ 75 and no-load current of ~3A @ 50V. Will post a vid once I get time to do it.

3) Made a external-Hall PCB in CAD, don't think there's a need for it though. Will post production files in the tech ref area.

 

[full-throttle]

Her's the vid:
[youtube]_WbRy3ZkAh8[/youtube]
12FET Infineon controller @ 50V is used.
The motor is in 'star', internal Halls placed @ 120deg in stator slots.
The CRO shows a Hall trace and a adjacent phase trace + freq and duty cycle of the Hall.
The frequency is 7x the rev/sec (7 magnetic poles).
At around 27sec I flicked 120% speed sw - note the waveform and sound change!

pardon the mess..