blown 36-48V controller upgraded to 150V

[Antonio Zanardo]

Why do you want to run such high voltage though? How about dropping your voltage to match your desired top speed, and upping the amps to match your desired power usage/output?

My answer:

I have 4 lithium batteries, each one rated 28.75V 10Ah at full charge.
I can match them as follows:

1) Two sets of 2*28.75=57.5V 10Ah connected in parallel to get 57.5V 20Ah (which is the present setup I am running with the recently repaired 20A controller)

2) Or a series of 3*28.75=86.25V 10Ah (but this way I have one battery left unused)

3) Or finally a series of 4*28.75=115V 10Ah.

Case 1) the full load top speed is 40 kmh (not mph!)
Case 2) the full load top speed is about 50 kmh (but battery Watts /hour available is less than case 1 and 3)
Case 3) the full load top speed is about 60 kmh

I cannot increase amps as I was told it won't be advisable to draw much of amps from lithiums, anyway the current is limited to 20 amps by the controller.

Of course 60 kmh is the top speed I prefer to count on, just in case...
However my usual cruising speed is a modest 30-35 kph.

Antonio Zanardo

P.S. Is there a way to know if my lithiuns are cobalt or polymer type?
The Amperex factory says they are polymer while Brett who sold them to me says they are cobalt. Who is right?

 

[xyster]

Is there a way to know if my lithiuns are cobalt or polymer type?
The Amperex factory says they are polymer while Brett who sold them to me says they are cobalt. Who is right?

Assuming your cells are rectangular in shape, they're both right. Lithium polymer cells use lithium and cobalt for the anodes and cathodes, just like regular cylindrical lithium-ion cells. The difference is that lithium polymer cells employ a polymeric electrolyte, which allows the cell to be shaped into a rectangle, also called a "prismatic" form. Prismatic lithium manganese cells have been recently brought to market, but this is still a very rare, more expensive breed. So if your cells are rectangular in shape, and lithium in chemistry, they are almost certainly lithium cobalt polymer.

http://batteryuniversity.com/partone-5.htm

The lithium Polymer battery

The lithium-polymer differentiates itself from conventional battery systems in the type of electrolyte used. The original design, dating back to the 1970s, uses a dry solid polymer electrolyte. This electrolyte resembles a plastic-like film that does not conduct electricity but allows ions exchange (electrically charged atoms or groups of atoms). The polymer electrolyte replaces the traditional porous separator, which is soaked with electrolyte.

 

[solarbbq2003]

put a bit of info up on lithiums
http://www.users.bigpond.com/solarbbq/lithiums1.htm
and for a bit of fun nice video of electric gokart
http://www.liveleak.com/view?i=2a0_1186251811

 

[BiGH]

very interesting read Brett hope you can try out some of those non metal cased LiPO4 batteries - should be nice and light

 

[Antonio Zanardo]

I am a little confused about lithium chemistry.

I know that some lithium batteries can catch fire or even explode.
Some media reported a chinese guy killed by the explosion of his cellular battery. He was keeping the phone in his shirt pocket, close to heart.
I prefer to ignore what could happen if my set of lithium batteries, which are located in the back of the bike close to my ass, decide to explode!

Other lithium batteries seem to be fireproof and not prone to explosion. In a word, they seem to be much safer.

Are those differences real? If so, how can we tell if the lithiums we are using are potential killers?



Antonio Zanardo

 

[bobmcree]

one possible solution you might have overlooked is stacking fets. you can just solder one on top of the other and by nature they will share the load. of course the higher one cannot dissipate heat as well, but there are always tradeooffs. ideally you would isolate the gates with a pair of 100 ohm resistors but it should work without. when a fet gets hot its resistance goes up and so the cooler one will take more power, forcing them to share equally. this is not the case with regular transistors.

i suspect your 14v regulator went, and you should check the irf2101 that drives the fets that blew, and the series gate resistors. even if it is a 400v transistor the power dissipation is limited and will increase with voltage. i have run at 100v with irf4110s and it works fine, but that is a 40A controller. i think that in the 20A controller it may not be possible to dissipate the power required to run at 40A with fets of the same on resistance. two fets in parallel cuts the r in half so the power dissipation is 1/4.

115v is just too much and i cannot imagine you need it on a bike. the higher the voltage the touchier the throttle will be. part of this is due to the mickey mouse way they use the soft start input of the pcm chip as the speed control input, and part is because the changes in duty cycle to send varying current to the motor require steps that are too small for such a high voltage.

 

[Doctorbass]

i have run at 100v with irf4110s and it works fine, but that is a 40A controller.

I wonder if it's too borderline to drive that everyday?

I've always think that fets need safety voltage gap to avoid them to short or destruct the jonction?

You use 100V mosfet at 100V?

I wonder if there is an overshoot of the voltage due to the inductive load of the motor or back emf with higher than 100V?

What do you think about that?

Doc

 

[fechter]

It would make sense to not go over the "absolute maximum rating", but there does seem to be a bit of headroom on the voltage rating. The fact that Lowell ran slightly over 100v and it didn't blow makes me think it would be OK.

 

[xyster]

It would make sense to not go over the "absolute maximum rating", but there does seem to be a bit of headroom on the voltage rating. The fact that Lowell ran slightly over 100v and it didn't blow makes me think it would be OK.

But Lowell's batteries sag about 20 volts under load. What if there was no sag? Even if they're fine at 106 volts resting, as his seem to be, couldn't these FETs fry with bursts of 106 volts, 80 amps?

 

[fechter]

Maybe that's why it hasn't blown yet. Coasting, there will be no sag, but no current. Under load, the voltage sags, the FETs stay happy.
The gate voltage is about 10v higher than the pack voltage on the high side FETs, so if the pack voltge is 106v, then the gates will see 116v under very low throttle conditions.

 

[bobmcree]

100v is the max voltage fresh off the charger on my trek with the modified kollmorgen motor, and there is quite a bit of sag under load with 80 cells at 5 milliohms each that is .4 ohms so at 20A it drops 8V, so the fets are really not seeing 100v very often. there may be spikes that go higher, but the fets have not blown. (of course i have burned out 3 motors, but they really do make a wild screamer at high voltage ).

i am going to take that bike back down to 48v and put a bmc motor on it. i am very happy with the bmc/uspd on the cruzbike at 48v

the kollmorgens are good up to about 1000w input and 30 mph on my uspd with standard gearing and a 26" wheel, but above that they over-rev and the magnet structure shatters

but back to controllers... in my case the motor cannot turn backwards, but with a hub motor i have been told people have blown the controller when running at high voltage and letting the bike roll backwards. i have not seen this myself.

 

[The7]

but back to controllers... in my case the motor cannot turn backwards, but with a hub motor i have been told people have blown the controller when running at high voltage and letting the bike roll backwards. i have not seen this myself.

My AL1020 has an geared hubmotor with internal free-wheel.

It is very little drag when pushing forward which is independent of switch ON/OFF..

With the switch OFF, it has a slight higher drag when pushing backward because the motor is turning backward.

With the switch ON, it has a much higher drag when pushing backward and I could hear some noise from the motor.
Apparently when the motor is turning backward, the motor enters the PLUGGING mode. PLUGGING mode is similar to switch it to REVERSE when moving FORWARD.

 

[Antonio Zanardo]

i suspect your 14v regulator went, and you should check the irf2101 that drives the fets that blew, and the series gate resistors. even if it is a 400v transistor the power dissipation is limited and will increase with voltage. i have run at 100v with irf4110s and it works fine, but that is a 40A controller. i think that in the 20A controller it may not be possible to dissipate the power required to run at 40A with fets of the same on resistance. two fets in parallel cuts the r in half so the power dissipation is 1/4.

115v is just too much and i cannot imagine you need it on a bike. the higher the voltage the touchier the throttle will be. part of this is due to the mickey mouse way they use the soft start input of the pcm chip as the speed control input, and part is because the changes in duty cycle to send varying current to the motor require steps that are too small for such a high voltage.

Hi Bob,
stacking mosfets is an idea I will take into consideration later on.

As far as the controller failure is concerned, just two mosfets have blown. The 2101, the low voltage regulator and the gate resistors did not.
I remind you that I am using 150V rated mosfets, and that 115V is the sum of 4 lithium batteries connected in series, each one supplies freshly charged 27.75V.
I replaced already the two blown mosfets, and the bike is running again. I suspect the failure was caused by a bad soldering.

At present I am running the repaired 20A controller at 57.5 V, but I plan to go back to full 115 voltage in the next days, just to see what happens.

The amps I will draw is naturally limited to 20A by the controller, but, if necessary, I will decrease it by means of the current adjusting method suggested by Fechter.

Thanks for your comments.

Antonio Zanardo

 

[The7]

but back to controllers... in my case the motor cannot turn backwards,

It seems to me all ebike motors can turn backward when rolling back.

 

[bobmcree]

you are correct of course that even a uspd drive will turn the motor backward when permitting the bike to roll back, and with the planetary gears the motor runs faster than a hub motor. i had been drilling fencepost holes all day and was so beat i didn't think about the freewheel in reverse. sorry for the misinformation, guess even i am not perfect.