Overvolted the "Name that Motor" today

[SamTexas]

... $600 is a major decision.
..., so $$ is not a factor.

Make up your mind, will you? Just kidding.

Hey, is the fork installed backward?

 

[catchinem]

Got the controller cut open. Now to chip away at the rubber stuff. Is that 13 FETS?

 

[NeilP]

FET's are not connected to a heat sink by the look of it, so they would not be good for too much current. the numbers we need are on the other side.

Am guessing a 12 FET controller with one FET doing something else

 

[Gödel]

Here's a few more details on predicting motor iife vs. voltage
http://ir.library.oregonstate.edu/xmlui/bitstream/handle/1957/16125/ExtendingElectricMotorLife.pdf
It seems if you run your motor at 33C above design winding temp you'll get 10% of the rated lifetime. Since motor winding temp rise above ambient is proportional to I squared you can figure the current it will take to do this.

 

[catchinem]

I'll go dig on the FETS now

 

[ohzee]

Good work getting in there catch - looks like that was not easy.

 

[NeilP]

That might not be such a good idea..you are going to have to bend the pins to see the numbers on them. May be ok, but a bit risky

Main caps are 63 volt, so running 14 series should be OK.

 

[catchinem]

Here's the numbers

 

[SamTexas]

Good job! Now you know you're good for up to 15s Lipo, 4.2 * 15 = 63V.

 

[catchinem]

I can't do Lipo Sam. I have to charge on the 2nd floor with a hardwood floor of a wooden condo with 4 other units in my building. How about that 52v 11.5 lifepo triangle battery Cell_man has? It will fit in my triangle just fine and look good!Can the controller handle it. By the way I think I can reassemble this controller. I didn't mess anything up I don't think. I'm pretty handy.

 

[SamTexas]

Just FYI: RC Lipo is the only one considered dangerous (very volatile) in my opinion. Laptop Lipo (18650) is very safe, again in my opinion. Laptop Lipo also have much higher energy density than RC Lipo. I have deliberately shorted them (for experimental purpose), discharged to zero, discharged to negative voltage (reversed), overcharged them to 4.50V, overdischarged way beyond their c-rate. I was able to kill the experimented cells, but I could not get them to explode or catch fire. Again just for experimental purpose.

Yes, cell_man 52v battery will be fine. It's a good and safe choice. All you want to be sure of is that the FULLY charge voltage of the battery pack does not exceed 60V.

Edit: Change 63V to 60V in light of NeilP's concern about the NEC part.

 

[NeilP]

I am in the same situation, second floor apartment, and charge my LiPo at home in the house

OK, so not ideal, but take precautions and there won't be a problem. Get a LiPo bag, or a lined box..or both,

Yep max voltage on the data sheet shows 60volts....so i am not quite sure why Sam is saying 63volt is OK..maybe he understands the figures n the data sheet better than I do. I see 60volt as MAx so have taken that as face value..maybe there si something else to consider that I am not aware of.

http://www.datasheetarchive.com/nec%20k4145-datasheet.html

When you try rebuilding it you could bolt the FETS to a heat sink. You will need to insulate them from the metal work of the heat sink with insulating kapton tape and individual top hat type washers so the bolts do not short the tab to the heatsink

 

[NeilP]

Here is me abusing a LiPo cell...at over 6 volts and 25 amps..for a 5 Ah Cell

had to hit it with a hammer before it finally burnt

[youtube]9FRjFxKoUYQ[/youtube]

 

[SamTexas]

Yep max voltage on the data sheet shows 60volts....so i am not quite sure why Sam is saying 63volt is OK..maybe he understands the figures n the data sheet better than I do. I see 60volt as MAx so have taken that as face value..maybe there si something else to consider that I am not aware of.

http://www.datasheetarchive.com/nec%20k4145-datasheet.html

I was only looking at the voltage of the capacitors. And I have 2 Infineon controllers with 63V caps and both of them have no problem handling 15s Lipo. But you might be right about the nec parts. I have never run into that problem so I have no hand-on experience.

 

[NeilP]

Data sheet atttached

 

[catchinem]

Thanks guys. Now....how do we determine how many amps this controller is capable of discharging? Is that controlled by the battery BMS or the contrioller? Cell_mans battery stats here: http://www.emissions-free.com/catalog/i29.html

 

[NeilP]

That will be down to the controller. nothing you can do to change that, it will be programmed in to it some where.

If you are running it at higher volts, then you will be putting more power through it, so heat sinking those FETs will be a good idea..but maybe see how hot they get first

 

[SamTexas]

Thanks guys. Now....how do we determine how many amps this controller is capable of discharging? Is that controlled by the battery BMS or the contrioller? ...

It is determined by the one with the lower capability. For example if the controller's max discharge is 20A and the Battery/BMS is 30A, then the max is 20A.

If we were to believe the sticker on your controller (33 to 44V, 500W), then the controller max rate is 15.15A (500/33).

 

[catchinem]

This case appears to have built in heat sinks along the sides. Those plates on top of the FETS rest directly on those sides.

 

[NeilP]

Ok, so there was a heatsink. That tape is the insulation between the FETS and the sink.

that tape is the insulation between the FETS and the sink

 

[catchinem]

I've got 2 more so not to worry. I guess I just need to get the bike built, strap on 4 SLA's I have laying around and go test it for speed. If Sam's assumption is correct that the controller will only put out 15 amps, then I may want to get a different one. I would like to get up around 28-30 mph out of this bike.

 

[NeilP]

Yes, I think I said about 14 amps further up this thread. I would definitely get a new controller, but no harm in trying the one you have for now. Run it up to 60 volts and se what happens

It is the voltage that gives the speed increase, so your current controller although will not accelerate quickly as it will be current limited, the higher volts will give more speed than it can at 36 volts.

 

[wesnewell]

I've got 2 more so not to worry. I guess I just need to get the bike built, strap on 4 SLA's I have laying around and go test it for speed. If Sam's assumption is correct that the controller will only put out 15 amps, then I may want to get a different one. I would like to get up around 28-30 mph out of this bike.

OK, you don't need to insulate the fet plates. Just don't short the legs. Those plates are for mounting heat sinks to them. That's why each one has a hole in it. I think most controllers use the case as a huge heatsink. put some heatsink compound on each plate when you reassemble it. Do not use any insulating tape. That will defeat the purpose of the heatsink plate. That do make heatsink tape, but I don't know where to buy it. Personally, I use auto axle grease because I've had it laying around here for over 30 years. Been using it on cpu coolers for at least 10 years. You can check the amp/watt output with any watt meter. Just get it running, hold the brake (or put biker against wall), give it power and what the meter. Most meters give both amps and watts readings, but it's easy to figure the amps with just the watts reading and source voltage. Unless that's a real fast wind motor, I expect max speed with 48v sla to be around 23mph. With 14s (58V charged) lipo I get about 25mph out of my 500W GM motor. With a $33 72V controller running 20s lipo on it, I think I got around 35mph on it. If you're looking for speed in the 30mph range You'll probably need a more voltage than you can put into that controller for that motor.

 

[catchinem]

I'm to old to be going 30mph anyway.

 

[parabellum]

OK, you don't need to insulate the fet plates. Just don't short the legs

Sure on that one? I would not give such recommendation.

Actually such "plate" is usually connected to Source or Drain of the fet, which means you will successfully short few fet groups with no insulation.