new Kelly brushless regen controllers

[Malcolm]

One thing I just noticed about the Kelly controller, don't know if anyone's mentioned it already. The upper voltage limit for the 72V controller is just 82V, while a brushed 72V controller such as the Alltrax has an upper voltage limit of 90V.

This could be a problem for anyone planning a 72V LiFe pack, as it would normally be built using 24 cells in series for a working voltage of 76.8V. The pack voltage fresh off the charger (at say 3.75V cell) would then be around 90V. I know that this voltage drops very quickly to 3.2–3.4V under load but it's likely the controller would shut down before that happens.

It looks like the maximum LiFe cell count this controller will handle happily is 21, so a 60V nominal pack would give a comfortable margin.

I'm not trying to knock this controller, I'd like one for my next project, just wanted to point out the limitations.

 

[fechter]

I'm not sure if it would shut down or not. It might be fine. I suspect the components are rated for 100v.

Somebody is going to have to open one up and take some pics....

 

[curious]

fechter-
Part of the problem opening it up is that the metal part is solid aluminum D-shape profile, there is no bottom cover. The screws only hold plastic endcaps. I am not even sure how they thermally mount fets to the heatsink, possibly some kind of spring-loaded arrangement because the internals must slide in/out of that D-shaped case somehow.

drewjet-
What issues have you discovered with regen so far ?

 

[curious]

rarebear -
Very nice bike. I am also really impressed how clean your setup looks !!!

 

[rarebear]

thank you curious!

What i think about the higher voltages:

The controller will shut down if the voltage is too high and show you an error code (flashed led code 1,2), but it will not damaged the controller!

For test purposes i run fully charged 20S lipos (84V) and had no problem!

I would recommend to ask King or Steven from kelly about that.

 

[drewjet]

I asked Steven about 18 Lipo and he said it would be fine. Wow 20! That would be sweeeeeeeeeeeeeet.

Hmmmm That has me thinking now.

 

[fechter]

In the programming screen, there is a setting for overvoltage. I guess it will shut down if it exceeds the programmed value. What is the higest setting for that?

 

[oatnet]

I wasn't getting any answer to my emails, so I just went ahead and ordered one from their website. The biggest reliability problem on my prototypes has been the controllers, so I hope this is the reliable, durable product I have been hoping for!

-JD

 

[rarebear]

Hi fechter!

In the programming screen, there is a setting for overvoltage. I guess it will shut down if it exceeds the programmed value. What is the higest setting for that?

Highest setting for my KBL72101 Controller is 100V !!!!!!!!!!!!!!!!!!

 

[ratatouille]

Hi curious,

I'm interested in the Kelly controller too, so i follow this thread with lot of attention.

Can you confirm that the Kelly controller needs a 24V power supply to operate and if yes, how will you proceed ?

I'm annoying at having two DC-DC converters, one for the controller and one (12V) for accessories.

 

[Microbatman]

Ok, I received mine so they do exist. It took over 8 days to ship from China via EMS sitting few days in China and US customs.........
I still wait for the mating hall sensor 5-pin micro-xlr from ebikes. Once I get these I will be able to test no-load operation only since I've only started my hunt for a123 packs. So the progress will be a bit slow as I mentioned before but I'll keep reporting it.

Curious,
Did you hook yours up yet?

I am a little confused about why you need the 5-pin micro-xlr that you had to order extra from e-bikes. Why not just cut off the XLR attached to the motor wire and hard wire the wire ends to the connector included with the controller?

Either way thank you for your info and post. Very informative and much apprecieated!!!!!!

 

[Microbatman]

Rarebear

What method/how did you attach the wires to the terminals on the speed controller?


Very nice bike!!!!! What brand is your bike?

 

[SniperGaulois]

if you like dewalt you should have a read about the eBay lithium polymer in the battery technology section

 

[rarebear]

hi microbatman!
thank you!!

i used the wire harness (hall) from a blown crystalyte controller, and connect it to the include 14 pole round connector to keep it "service friendly"( wheel assembly)

you can use any universal 5 pole connector for the hall wires.

my (bike) frame is from a german bike factory (poison)

 

[rarebear]

Hello controller specialists!!


I have a question, but first my correspondence with steven from kelly:

I want to know how can I Increase the acceleration starting up from a
>> dead
>> stop? (More Torque)
>>
>> The current draw starting up from a dead stop is very low (20A) and
>> Increases only at higher speed (80A) (Voltage under load 37 V)
>>
>> Should i change the setting "Control Mode" from Speed to Torque?
>>
>> The Accelerate speed is setting "Factory default" at 30 . change to 100?
>>
>> Please support.
>>
>> Thank you!
>> Albert

Where you measure the current? Battery (or say bus) current?
>
> Our controller controls phase current. At low speed, the controller
> multiples battery current to get phase current. So the phase peak
> current will be much higher than battery current. So controller might
> output 100A phase current at low speed.
>
> If there's the case, no adjustment will get higher current.
>
> Otherwise you may ask King to do something.
>
> Best regards,
> Steven

Hi Albert,
>
> You can change the setting "Control Mode" to Torque. on torque mode the
> throttle will control the phase current .
> The Accelerate speed you can set it 10-200,10 is faster.
>
> King

o.k.!
now my settings are : Control Mode: change from speed to torque
Accelerate speed: change to 10

now my acceleration is much better and the current will reach 110 amps after a few seconds but when i starting up from a dead stop the current is still at 20-30 amps!


I have not yet change the setting "load type" from heavy load to light load (perhaps a difference?)

please reports your current (modified crystalyte) starting up from a dead stop (full throttle)

 

[magudaman]

Sounds like a slow ramp up? Is there a throttle ramp control? Should be only a couple milliseconds, sounds like yours might be set too high.

 

[fechter]

I think the key is they are using the motor phase current to control the limit, not the battery current like a Crystalyte. On takeoff, my modifed Xlyte controller hits 100 amps on the battery, which means it's like way over 100 amps on the motor phases for a second or so.

By limiting the dead stop current, the controller will be much less likely to blow up.
To get better acceleration, you'd need to either increase the current limit, or lower your gearing.

 

[rarebear]

Hi!

I think Fechter is right!

with a locked rear wheel (brake) the current is only at 20-30 amps

starting up from a dead stop after a few seconds (speed 30km/h) it reached 115 amps

Probably this is an fixed security establishment to protect the controller.

 

[fechter]

I think it is not a bad thing. It will also help prevent the motor from burning out.

It is not so good for drag racing, but I'm sure it's possible to bypass
In a Crystalyte controller, the battery current is measured, not the motor current, so you can get several times the battery current at low speeds.

 

[Mathurin]

Would it be possible to externally hack such a controller into thinking there's less current in the motor then there really is? Example a second set of freewheel diodes on the motor leads or something.

 

[fechter]

Would it be possible to externally hack such a controller into thinking there's less current in the motor then there really is? Example a second set of freewheel diodes on the motor leads or something.

That might do it if they had a lower forward voltage than what's inside. It would not increase the current on the stock controller either. Hmm... good idea.
I was thinking more like solder on the shunts.

 

[Microbatman]

See post below

 

[Microbatman]

now my acceleration is much better and the current will reach 110 amps after a few seconds but when i starting up from a dead stop the current is still at 20-30 amps!


I have not yet change the setting "load type" from heavy load to light load (perhaps a difference?)

please reports your current (modified crystalyte) starting up from a dead stop (full throttle)

Rarebear.
Thank you for finding this low current on start up problem.
This is a definite show stopper!!
I think it is very important that max current is available on start or people will look to other controllers.
This is a very important issue please keep us all posted about progress.

 

[markcycle]

Hi Folks!

At first please forgive me my poor english!!! (come from Austria) ( no we have no kangaroohs :lol: )

But I will try to answer all questions.

After a second failure on a Crystalyte i ordered a Kelly and I can tell you i am verry happy with it!

The programming options are very helpfull. (also the wiring schematic on the manual)

Only to find out the right phase wiring was a little tricky!

Here is the right wiring for an Crystalyte x5:

Motor: A blue / B green / C yellow

Hall: A yellow (13) / B blue (12) / C brown (11)

My input power is about 3,5kw an the Controller stay cool after a ride from 10 miles with full speed and some startings up from a
dead stop!!!!

I have not tried to change the settings for the advance angle,
What is the recommended setting for an x5?

oh yes! sorry i had not looked in the inside of it!

Thousand thanks to Steven and King from Kelly for the great support!! Very nice guys!!!!!!!!!!!!!!!!!!!!!!!!

My Conclusion: Highly recommended!!

I hope that I will not wrote too much waste
Albert

Just wanted to say thanks for posting the wiring to the X5 motor. Those are the things that make a forum great. Connected my new Kelly controller just as you stated and it just worked

Thanks again
Mark

 

[nlc]

It's normal you just see 20-30A when motor is stalled, because you read the battery current. It's in the motor's phase you have the 100A (or max current supported by the controller).

The controller works as a buck DC-DC converter. When motor is stalled, the back fem is 0V. Let say that the controller efficiency is 100% and the internal resistance of your phase motor is 0.1r. If your controller is limited to 100A, you have 100A on the 0.1r, thus you have 10V on the motor phase, and the output power of the controller is 1000W.
Thus you also have these 1000W at the input of the controller : If your battery is 48V, you will measure 21A, if you battery voltage is 72V you will measure 14A, and so on...

When motor begin to spin the controller output voltage rise because of the back fem of the motor, thus the output power rise too, thus the input current rise too. And there is a motor speed where motor phase current will be the same as battery current, it's when the controller PWM is 100% (the output voltage is the same as the battery voltage), and the motor current is naturally lower than the controller current limit).

I don't know if it's very clear (I am french and my english is not very good), but if some point are not very clear don't hesitate to ask !

On my side I work on a 400A controller for my Mars motor ! Because I have a fixed gearing thus to have good acceleration and good to speed, the only solution is to send a lot of current at low speed

Cyril