2000W MAC lawnmower motor

[John in CR]

Ja, that makes sense.

No difference in draw with two different hall combos is still a puzzle to me.

Yeah, I've only seen that on my 63 slot hubbie. Maybe he change the phases too, in which case we'd expect 3 correct ones for each direction, though Fechter swears there's a BMC with only a 1 in 36 valid combo, which I don't understand how it's possible.

[magudaman]

Ja, that makes sense.

No difference in draw with two different hall combos is still a puzzle to me.

Yeah, I've only seen that on my 63 slot hubbie. Maybe he change the phases too, in which case we'd expect 3 correct ones for each direction, though Fechter swears there's a BMC with only a 1 in 36 valid combo, which I don't understand how it's possible.

Yes I went through all 36, so I was changing phase wires too. I only found a total of 4 that seemed to work correctly, I believe I had 6 on the BMC too.

[torqueon]

Is this still a work in progress, is there a verdict.

Pete

[neptronix]

Is this still a work in progress, is there a verdict.

Pete

Looks like i may be settled in a new place within a week, week and a half. Am anxious to get the motor running for sure.

What are you wondering by now though? We kind of have a collective idea of what the motor is capable of. Seems to be similar to the MAC/BMC hub motors, containing what looks like a little more magnet and copper ( but not significantly more ). Power handling on this motor may be equivalent to the BMC 1500W motor, but we are not sure, heat dispersion may be lower due to the case.

I will get the butt dyno, cycle analyst, temp sensors, gopro, etc. all running in a bit & get some proper data, most particularly what is the constant wattage it can handle at various voltages, but that is the gist of this motor for now.

[neptronix]

Woot! I'm settled in a place near Salt Lake City, UT with a garage of my own.. the cargo frame that will be a test platform for this motor will be here in a few days. Sorry to drag my part of the testing on, but the holdup is over and the progress begins now

[torqueon]

In the continued search for that low kv 2000 watt motor, with a thin form factor. I just stumbled upon this doing a random search. A permenant magnet 3 phase generator, model, ME1016 sold by motenergy.com ( sorry i could not paste the site, for all to see ) This looks promising, 10 pole pairs, 8 inch O.D. perhaps lots of torque.

The price is $ 150 + shiping

This is listed as a A.C. generator, so no position sensors. My thought is install sensors, controller and you have a good motor.

Please check it, to see if its good, or if I am barking at the worng tree !!

Pete

[neptronix]

Action news update..

Settled in the new place, got the new frame, just dropped $130 on various bits and pieces for a 2 stage reduction. I'm going to run this motor at 72v with a ~7.5:1 ratio to a 24" wheel and see what it can do in the neighborhood of 5000rpm.

Why not take it all the way, right? I'll see what this thing is capable of for real

Then i want to do a 5:1 reduction at 48v to see if it can really do 2000w continuous. I really doubt it can at this point.
5:1 reduction is rather extreme but it can be done.. that was the whole point of checking out this motor. Because building a double reduction system is not for everyone, and looks goofy on a bike in addition.

If it can really do 2000w continuous at 48v, and you guys are still interested in it ( seems like interest dropped off pretty hard! ), i can coordinate a group buy.

Another possibility is getting our hands on the smaller 300-600w and 600-1000w MAC motors, if anyone wanted to use those as a mid drive setup or something like a brushless currie rear drive. KV seems similar and they are cheap and light. 3-6 pounds.

[neptronix]

BTW magudaman, what reduction ratio were you using, to what size wheel? and you were running 52v, correct

[AmpEater]

I'd still like to pick up several of these for electric mower applications. I could use 4+, 3 for upgrading the deck on my electric john deere and at least 1 for my autonomous solar mower bot.

[Miles]

WRT running these motors at higher speeds:

Eddy current losses go up as the square of the fundamental frequency (rpm/60*n.pole pairs), unfortunately.....

They also go up as the square of the lamination thickness.

So, to double the speed and maintain the same level of eddy current losses you need to halve the lamination thickness.

The hysteresis losses will still double because they are close to proportional to the fundamental frequency and are not related to lamination thickness.

If you measure how much the no load current goes up from its value at the original speed, the difference is due to the increase in eddy current losses. The hysteresis part of the no load current value will remain pretty much constant.

[wasp]

to get the kv down couldn't one rewind 1 of these motors?

[neptronix]

to get the kv down couldn't one rewind 1 of these motors?

I'd say it verges on not being worth your time to do that, really.

These are not particularly good at shedding heat ( by design ) nor wonderful in efficiency ( 86% peak is okay though i guess.. ), and they are oversized due to their case and metal plate design. And their 7/8" shaft sucks too.

The only thing these motors have going for them is that they're cheap

You might want to consider the GM trike motor on high volts if you just want a big single reduction chain drive motor.

http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=45443&start=0

I should be getting mine pretty soon & should be able to tell whether it can take high RPM operation or not.

[neptronix]

Spun it up on some good volts today to measure the no-load.

38v: 2688RPM / 3.15A x 38v = 119.7W loss
57v: 3878RPM / 3.6A X 57v = 205.2W loss
78v: 5037RPM / 4.0A x 78v = 312.0W loss

Not as bad as i thought. I'm wondering how much of that is controller loss and how much is motor loss. I've got a 12FET 4110 infineon here.

Since this motor is rated to produce 1500W continuous at 36v, it could very well produce 3000W at 72v.

While i was at it, i compared it to the used 'bad halls' GM BLT-650 ( with the magnets all chipped up ETC. )



38v: 1016RPM / 0.91A x 38v = 34.2W loss
57v: 1510RPM / 1.1A x 57v = 62.7W loss
78v: 2030RPM / 1.3A x 78v = 101.4W loss

[neptronix]

Whaddya think, miles? as bad as you thought? worse?

[Miles]

Pretty much as expected, I think.

[spinningmagnets]

For those without a lathe, there is a freewheel adapter for a 7/8-inch shaft from staton-inc.com for $15 + shipping. It may be possible to run a single reduction using a 20-inch wheel on the rear?

http://www.staton-inc.com/store/products/7_8_ID_x_1_2_wide_Adaptor_for_Freewheel_Sprocket-888-27.html

[neptronix]

Oh, if you've got a 20" wheel, then a 5:1 reduction would be awesome.

12S ( 46v nominal ) would produce about 2576RPM loaded, 5:1 would get you down to 515.2rpm, which is about 30mph. Just ramp the volts up from there to get into >40mph territory.

A 24" or 25" wheel would need more reduction though

As far as things that fit onto the 7/8 shaft, robotmarketplace sells a #35 15T sprocket, and staton-inc i think has a 11T #415 ( BMX chain ) sprocket for the 7/8 shaft. But i think running the first stage of reduction would be very loud with a 1/2 pitch chain like the BMX stuff.

[spinningmagnets]

"Cheap no machining jackshafts" edit: shaft shown is 12mm dia case-hardened / bearing is 12mm-28mm for ID-OD...
http://endless-sphere.com/forums/viewtopic.php?f=2&t=27320

[Jeremy Harris]

Spun it up on some good volts today to measure the no-load.

38v: 2688RPM / 3.15A x 38v = 119.7W loss
57v: 3878RPM / 3.6A X 57v = 205.2W loss
78v: 5037RPM / 4.0A x 78v = 312.0W loss

Not as bad as i thought. I'm wondering how much of that is controller loss and how much is motor loss. I've got a 12FET 4110 infineon here.

The controller loss at 4A and 78V will be small, around 1W in the FETs and 4W in the controller electronics, so maybe 5W total, or around 1.6%.

[neptronix]

I have made some progress on my test build. Finally figured out a way to get the chainline straight enough on this frame, after some experimentation.



I am going for a 6.28:1 ratio on a 24" wheel, which should give me 30mph on 12S with lots of torque ( on that voltage, this motor would be rated to do ~1500w continuous ). On 20S, this should produce 40mph.

Now i have to go pick up some #35 chain, #35 sprockets, and a smaller shaft for this build.

[spinningmagnets]

If anyone else likes this motor, but would like the option of a two-stage reduction using a quiet belt/pulley for the primary, McMaster-Carr has a 7/8-inch shaft adapter ($14 part# 6086K198) that allows a selection of pulleys to be bolted on, with a $45 22T pulley being the smallest I could see. If anyone finds something similar for a 17mm shaft, please PM me for the Ariens thread.

http://www.mcmaster.com/#quick-disconnect-bushings/=kay63e


http://www.mcmaster.com/#timing-belt-pulleys/=kay4gr

[jmygann]

so ... what is the latest thoughts on the best 1000-2000 non hub motor ?

[jmygann]

what kind of ratio would one try if 26" wheel and a nuvinci hub ?

http://www.organictransit.com/

[neptronix]

so ... what is the latest thoughts on the best 1000-2000 non hub motor ?

In terms of what quality? price? efficiency? weight? power output? form factor?

what kind of ratio would one try if 26" wheel and a nuvinci hub ?

Totally depends on what speed you want. divide loaded motor speed ( you can take the unloaded motor speeds in this thread and multiply them by 0.80 to get 80% ) by the ideal wheel speed and you will get the correct ratio.

http://www.recumbents.com/wisil/e-bent/rc_drive/rc_drive_calculator.asp

This can be helpful for finding that out.

[jmygann]

25 mph max on a 26" wheel