BLDC Motor and Controller - How to match them?

[Chris]

Hi everyone,

being a first time poster at endless-sphere (after reading for a while) I have to say I'm very impressed what seems to be achievable if you're willing to lift your butt and get your hands dirty. You guys have some really neat projects going and I'm glad that you're willing to share your experiences.

While I was reading here and at some manufacturer's websites I stumbled across a question which I wasn't able to answer myself: Brushed DC motors and controllers are usually rated by a certain voltage and current. At full load (100% duty cyle) one could virtually eliminate the controller and find that the battery voltage equals the motor voltage (same as for the current) - purely DC and easy to understand. But how do I read brushless motor data? Let's take the MARS BLDC. The fact that this motor sometimes is referred to as Mars PMAC literally accompanies my question: Given the rated data of 48 Volts / 100 Amps - is this DC or AC?

Assuming it's DC (view from battery): 48 Volts x 100 Amps x 95 % Controller Efficiency = 4600 Watts motor input power ... or AC (view to motor): 48 Volts eff x 100 Amps eff x sqrt(3) = 8300 Watts motor apparent input power ...

Same as for the Controller: Does the Kelly KBL48201 draw 120 Amps maximum from a 48 Volts battery or does it supply the motor leads with 48 Volts / 120 Amps ...

Guess that's a typical newbie question ... but anyway ... your input would greatly be appreciated.

 

[fechter]

Most brushless controllers, like the Crystalyte, limit the current based on the battery current, so the power is DC at that point and you can simply multiply current x voltage x efficiency to get the input power.

The Kelly controller, and some ot the other high end units limit on the peak (not RMS) motor current. At full throttle and no limiting, the power rating is essentially the same as if it were DC. At lower throttle settings, or during limiting, the peak motor current is much higher than the battery current, so the input power is actually less than the ratings multiplied. The efficiency would still be around 95% or better.

Whether AC or DC, you can always measure the power at the battery, where it's DC.

Hope that makes sense...

 

[Chris]

Thanks, fechter. I guess then, the only non-confusing way of rating BLDC motors would be to refer to the DC side as well ...

If this applies to the Mars PMAC then a Kelly KBL48201 (48 Volts / 120 Amps continuous) would be the right choice, I guess.

 

[fechter]

Yes, it's sort of apples and oranges when comparing the amp ratings. A 100 amp Kelly controller will be much slower off the line than a 100 amp Crystalyte. The peak motor current to battery current ratio can be something like 3:1 - 4:1 at low speeds (maybe higher). At full speed, it's 1:1.

As long as you know this, you can predict the performance pretty closely.

 

[drewjet]

Thanks, fechter. I guess then, the only non-confusing way of rating BLDC motors would be to refer to the DC side as well ...

If this applies to the Mars PMAC then a Kelly KBL48201 (48 Volts / 120 Amps continuous) would be the right choice, I guess.

It depends on what you want to do with the Mars. I have a Mars and Kelly 300 amp on a Suzuki SP200. I think the motor is still not hitting its max potential.

 

[Chris]

I think the motor is still not hitting its max potential.

That's exactly my concern as for making a controller decision. Am I right assuming that you own a Kelly KBL72301? And still not melting down or dismantling the Mars? What battery voltage/current can you reckon as "save" operating point (certainly not wanting to hold you liable)? And what would be the the thwarting controller attribute in a Kelly/Mars combination: voltage, cont. current or peak current?

Sorry for lunging at you.

 

[fechter]

I think the Mars can handle more amps. You can only draw more amps at very low speeds when accelerating anyway. Once it speeds up, the amps drop. I don't think anybody has fried or overheated one yet, even with the Sevcon controller.

 

[drewjet]

That's exactly my concern as for making a controller decision. Am I right assuming that you own a Kelly KBL72301? And still not melting down or dismantling the Mars? What battery voltage/current can you reckon as "save" operating point (certainly not wanting to hold you liable)? And what would be the the thwarting controller attribute in a Kelly/Mars combination: voltage, cont. current or peak current?

Sorry for lunging at you.

Yes I own the kelly 300 amp w/regen. I am running 75 volts of Lithium and 32.5 Amp hours 20C capable batteries. I think the limit is both the voltage and amperage. The Kelly is rated at 72 volts and I was told by Kelly that 75 would be OK. I think Jozzer is running at 80 or more volts. I can't tell you what amps I am pulling as I just got a shunt and haven't got it hooked up yet, but after a 10 mile run at or near full throttle the motor is only about 150 degrees F. And that is that I don't have the right fan, the motor is turning the wrong direction in my application and I haven't located a correct fan yet (haven't looked to hard yet).

Like Fetcher says, the high amps only come into play on acceleration, and I am sure it could handle a bit more volts and amps for a short duration, but I think the Kelly has no more headroom.

 

[Chris]

Thanks, guys. Looks like the KBL72301 would be the right choice for me - unless ... I read you fried it twice, drewjet.

Paying 600 bucks just for sending controllers forth and back doesn't look like a perfect deal. Any news on cause study?

 

[drewjet]

Both times they covered it under warranty, only cost me for shipping it back to Chicago, about $16.00

So far this last one is working well. i have put about 40 miles on it so far and no issues.

Kelly's customer service has been outstanding. I hope it stays that way.

The first one I did not open, but the second one I did and t was the FETs that blew. There is some discussions going on as to upgrading them, but nothing solid yet. If this third one blows, I may well consider doing it myself, or gettting someone with the skills to do it for me.

 

[fechter]

If the main caps fail, it might not be visible. If they fail, the FETs will fail immedately when you apply throttle.

 

[gasax]

gidday I was just reading your posts I recently set up a me3001 and a kbl48201 on my sailboat ,once I got through the colour code dramas of the motor to the controller ,I followed the wiring diagram exectly ,the motor is accually a three phase "AC" motor And the controller is a dc-ac converter ,so when your conections and switches and the phassing is all correct the motor will run as sweet as ccw and cw with the fan running the right way for forward .
I've run the motor for some 12hrs on 36volts and only draw 15 amps full load and all the gear gets as hot as a slice of toast.
what I am looking for when I stumbled in here is tacho wiring and regen volt meter wiring

 

[fechter]

What kind of tach do you have? The hall signals will toggle once for each magnet on the rotor, so the frequency will be the rpm x the number of magnet poles. I'm not sure how many magnet poles there are on a Mars motor (wild guess: 8 pairs), but you could easily measure it by putting a meter on one of the hall signal wires and slowly rotating the shaft one revolution. It may be easier to fit a separate sensor for the tach.

Regen on a sailboat? Normally you want to measure regen current, not voltage. You can measure the battery voltage, which is always a good idea. A regen current meter would go in series with the battery connection. You can use an external shunt with a remote meter. You could have two meters, one for forward current and one for regen (or a meter with zero in the center). Both meters could share the same shunt.