Farfle
100 kW
The links don't seem to be visible on my end. The only way I have been able to share something from emetor is to screenshot it.
Farfle's Mk.5 Super badass outrunner
- Thread starter Farfle
- Start date
Farfle
100 kW
Yeah, looks like magnet losses will need to be mitigated with a spectrum so similar to that guy .
Small segments, and maybe an electrically insulating coating?
Good news is that no matter what, the magnets don't want to fly off the rotor :lol:
Small segments, and maybe an electrically insulating coating?
Good news is that no matter what, the magnets don't want to fly off the rotor :lol:
Farfle
100 kW
Miles said:
Ill see if i can find any epoxy coated magnets in this size range. Last I checked that was a special order kinda thing.
Farfle
100 kW
Hmm, epoxy coat the back iron, do one circumferential row seperated by fish paper, then mask and apply epoxy to only the exposed edge, and repeat. Would let me get away with using a stock size nickle plate magnet.
I tossed a few emails and filled out some automated quote forms on a half dozen magnet websites. Will report back results. This motor needs 160 20x20x5mm magnets, ideally in N50M grade, and epoxy coated.
I tossed a few emails and filled out some automated quote forms on a half dozen magnet websites. Will report back results. This motor needs 160 20x20x5mm magnets, ideally in N50M grade, and epoxy coated.
Farfle
100 kW
Looks like K&J magnettics was the first to respond: looks like 800-1000 dollars from them. Thats a good chunk of the 3000 dollar budget.
Please find our quote for these custom magnets below. Custom magnet orders require a minimum purchase amount of $150 per shape. Magnets are typically ready for shipment in 7-9 weeks after we receive your order.
Custom NdFeB Block Magnet, Part No. 3B8701
0.750" x 0.750" x 0.188" thick
Grade N50M, Black Epoxy (Ni-Cu-Ni + Epoxy) Coating, magnetized through 0.188 in thickness
Qty 19 @ $7.99 each (minimum order quantity)
Qty 25 @ $7.41 ea.
Qty 50 @ $6.37 ea.
Qty 100 @ $5.65 ea.
Qty 160 @ $5.42 ea.
Custom NdFeB Block Magnet, Part No. 3B8702
0.750" x 0.750" x 0.188" thick
Grade N48H, Black Epoxy (Ni-Cu-Ni + Epoxy) Coating, magnetized through 0.188 in thickness
Qty 18 @ $8.36 each (minimum order quantity)
Qty 25 @ $7.68 ea.
Qty 50 @ $6.62 ea.
Qty 100 @ $5.89 ea.
Qty 160 @ $5.65 ea.
Please find our quote for these custom magnets below. Custom magnet orders require a minimum purchase amount of $150 per shape. Magnets are typically ready for shipment in 7-9 weeks after we receive your order.
Custom NdFeB Block Magnet, Part No. 3B8701
0.750" x 0.750" x 0.188" thick
Grade N50M, Black Epoxy (Ni-Cu-Ni + Epoxy) Coating, magnetized through 0.188 in thickness
Qty 19 @ $7.99 each (minimum order quantity)
Qty 25 @ $7.41 ea.
Qty 50 @ $6.37 ea.
Qty 100 @ $5.65 ea.
Qty 160 @ $5.42 ea.
Custom NdFeB Block Magnet, Part No. 3B8702
0.750" x 0.750" x 0.188" thick
Grade N48H, Black Epoxy (Ni-Cu-Ni + Epoxy) Coating, magnetized through 0.188 in thickness
Qty 18 @ $8.36 each (minimum order quantity)
Qty 25 @ $7.68 ea.
Qty 50 @ $6.62 ea.
Qty 100 @ $5.89 ea.
Qty 160 @ $5.65 ea.
jonescg
100 MW
Hey Farfle,
Don't know how these prices compare to those of Lodestone here in Perth, but see what they have on offer:
www.lodestoneindustries.com.au
I got some prices a while back, but magnet prices go up and down all the time. Well worth a look, and if you do want to place an order I can pick them up for you - they're about a 7 minute walk from my house
Don't know how these prices compare to those of Lodestone here in Perth, but see what they have on offer:
www.lodestoneindustries.com.au
I got some prices a while back, but magnet prices go up and down all the time. Well worth a look, and if you do want to place an order I can pick them up for you - they're about a 7 minute walk from my house
Just stumbled across this Formula Student racer http://www.amzracing.ch/en/cars which is powered with a 'home built' outunner similar to what your specs call for. Not much detail given other than the weight is 4.6 kg and power output is 37kW (seems a bit high if you ask me).
Here's an older video showing some detail:
[youtube]Z6oZQUhATfE[/youtube]
Here's an older video showing some detail:
[youtube]Z6oZQUhATfE[/youtube]
Timmey
10 mW
mauimart said:
That 37 kW is a peak rating. The competition is limited to 85 kW, anything over will result in a DQ. So teams using multiple motors design their car for +/- 80 kW max in order to be safe.
85kW is the nominal limit? there is no limit of peak power?
In this case they have almost 150kW peak power. Seems unfair to the ICE competition. How do they check the nominal limit?
Anyway, looks like a great motor. But it probably runs above 400Hz, which I believe is what's limiting Farfle's motor, because of the controller.
In this case they have almost 150kW peak power. Seems unfair to the ICE competition. How do they check the nominal limit?
Anyway, looks like a great motor. But it probably runs above 400Hz, which I believe is what's limiting Farfle's motor, because of the controller.
Farfle
100 kW
bearing said:
I don't think that I will be using the sevcon this go round. I will probably use the 1000A 120v kelly, or two 600A 120v controllers (more cont power, and they seem to play nice in pairs with some fiddling).
That raises me frequency limit to 666hz, for the base model, and 1066/1666 hz for the upgraded models.
liveforphysics
100 TW
Farfle said:
Just as a datapoint, when we swapped from a size6 to the biggest Kelly controller made that claimed 1000A phase current controller with ultra-high-speed firmware, it blew up in under 2mins of gentle dyno testing, and never was able to make the power of the size6 before it exploded. It was also about 1.5-2ft long and maybe 25lbs.
Just because Kelly writes many zeros on the datasheet doesn't mean it actually works unfortunately.
Farfle
100 kW
liveforphysics said:
That really is a shame. After seeing what Ripperton had done with his twin Kellys, I was hoping the newer models were getting better. :x . Looks like low pole count and maybe twin size 4 Sevcons are the only doable option. The Scott drive controller would be the only other option in that power range that is sub $4k, but 400V scares the crap outta me, and is still expensive. I am hoping to keep the powertrain under $7k
jonescg
100 MW
Farfle said:
Dan is essentially running two motors through two controllers, so it's not quite the same, but apart from the odd overcurrent brown-out they seem to work well. I'm not sure I'd trust a one-off opinion so faithfully
. Kelly have been keen to tweak their controllers to suit the Colossus motors, so they might be willing to do the same for your motor once it's ready to spin.[/quote]Farfle said:
Pfft. 400 is just a number. It's only a hazard if you make it ghetto. With good precautions you can safely manage high voltage.
Timmey
10 mW
bearing said:
85 kW peak
Timmey said:
Ok, thank you, but their web page says 4x37kW. So they are carrying bigger motors than they are allowed to use?
EDIT: Hm, I read something about torque vectoring. So at times they are using 37kW from one motor, yet the sum is still <85kW?
Timmey
10 mW
Indeed sum is always < 85kW. Torque vectoring and weight transfer during high G (>1.5G) maneuvers 'limit' the usable power at the wheels.
Farfle
100 kW
I may have another controller option, and it is a monster. 80ish kw cont, and peaks well above that with lots of room to spin. That being said. I upped the stator configuration to 24t 16p, and its amazing how little yolk iron and back iron you need. took the active weight from 33 kg to 24kg with little reduction in power. Phase wires at 50% slot fill are 70.5mm (inbetween 00 and 000AWG) with two turns. Inductance has also taken a turn for the better (no pun intended) at 18uh.
Farfle
100 kW
Miles said:
Because I don't know any better
These are the reasons I chose it:
it is easier to wind
produced a lower torque ripple (at least in the configurations i tried)
and made more back EMF for a given turn count.
All reasonable points.
The advantage of the double layer configuration for 24t 16p is that all the sub-harmonics are eliminated - meaning, you don't have to worry too much about magnet losses. The single layer option, on the other hand, doesn't look too good.....
The advantage of the double layer configuration for 24t 16p is that all the sub-harmonics are eliminated - meaning, you don't have to worry too much about magnet losses. The single layer option, on the other hand, doesn't look too good.....
