what do you think of this sine wave controller? g-k125

izeman

1 GW
Joined
Jun 21, 2011
Messages
5,131
Location
Vienna, Austria
looking for a way to make my mid drive mac more silent i search aliexpress.com for sine wave.
this is what i found: http://www.aliexpress.com/store/product/Scooter-BLDC-60V-1000W-18-mosfet-sine-wave-controller-high-power-bike-engines-controller-G-K116/937734_1831530898.html
i need a controller for 2kw sustained power for several minutes. atm i have a modded 9fet infineon controller which not even gets warm. any idea why they need 18fet to get nearly that power? maybe they underpower those controllers?
anyone maybe got one and can give some feedback?
btw: this is the same one but more power: http://www.aliexpress.com/store/product/60V-2000W-18-mosfet-motor-brushless-sinusoidal-wave-controller-for-electric-bike-hub-motor-controller-G/937734_1873092974.html
 
do it. they have all kinds of good stuff. they even have pictures of the hub motor assembly. showing how they do the work. that is the best kinda marketing imo. they show you they do good work.
 
I just have a lot of conversation with Kevin from this seller. He confirmed that it is a sine wave controller and can be programmed with keywin software. So i assume it's more or less a reprogrammed (new firmware) xie chang type controller? What ever it is for the price I can test it and dump it if it doesn't work. They also rate them imho very conservative. He says 25a max for the 12fet. But i guess better underrate them instead of overrating. I will add braid wire to the board and maybe swap fets and then program it to 45a. atm i have my 9fet programmed with 45a/75a battery/phase w/o any issues (i set block time to zero to not overshoot those values).
Good thing is that I can easily add a programming and CA -v3 connector.
 
Very interesting shame the voltage didnt increase with the amp rating on the second http://www.aliexpress.com/store/product ... 92974.html
Let us know If you find out they are programmable , getting tech of some suppliers is a pain sometimes.

Edit You beat me to posting , good news on programing , how about higher voltages.
 
i invited kevin from Suzhou SND Zhenlong Motor Co. Ltd to this thread. maybe he'd like to join and answer our questions so others can profit and he can make more profit ;)
i must say i got excellent support so far and and confident that i can make this controller work for me. i asked to put it in a bigger enclosure, as i have a very special cabeling solution on my bike and like to add stuff inside the controller box (dc/dc converter and anti spark circuit). no problem for them to do it which is a BIG plus showing.
this is what i did to my 9fet/12fet box. all wires go through this slot and enter the same sizes slot in my frame, so it's water tight and no wires are visible http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=55215&p=873150#p873150
 
Emoto said:
Very interesting shame the voltage didnt increase with the amp rating on the second http://www.aliexpress.com/store/product ... 92974.html
Let us know If you find out they are programmable , getting tech of some suppliers is a pain sometimes.

Edit You beat me to posting , good news on programing , how about higher voltages.
they have that one: DC motor 60V~120V 9000W powerful controller for e-bike/electric scooter speed controller G-K041
what are you looking for? i have to ask about those voltage limits - i guess this is just LVC settings as i don't see any reason why there should be a technical difference between an 48v and 60v controller (fets, caps, resistors are all the same)
 
kevin just wrote to me:
But for our controller we use 100V main caps. Can hold 84V voltage. About the mosfet we use ST75N75.
used on 48V voltage,if use 60V~84V voltage we use IRFB4410.
so they caps are good for up to 100v anyway. i prefer the 4110 with 3.7mohm IR over the ST75N75 so i will go for a 12fet sine wave with 60v.
EDIT: it's 4410 and NOT 4110 :(
 
I just order some more 3077 from utsource. They are $2 per piece, so a $24 job for swapping all of them if it's needed. Only problem is falling temperatures as winter is coming. It's almost freezing. No way I'll go ebiking or quadcopter flying. Good thing : I got a lot is time to fine fune my bike.
 
Wow, they sure do have a wide selection.
I'll be curious to hear a test report on the sine wave controller.

I wish they would make an affordable FOC sine wave version. It could be very quiet and have good regenerative braking. Having synchronous rectification would also greatly reduce the heating and allow for a smaller package.
 
what i forgot to ask is the maximum e-rpm capability of these controllers. i need that for a geared MAC motor and it's e-rpm is around 40krpm. i know that the older eb2xx boards had troubles running those motors, and even regular kelly controllers are not able to run those motors - especially the low turn ones and/or the one with higher voltages then 50v.
kevin, i hope you join us and can answer that question as well ...
 
I remember hitting the RPM limitation on some of the very early controllers, and then overclocking the processor to about double it. The higher your pack voltage, the higher your motor will try to go too.

If you do a no-load, full throttle test with the motor it should be evident by the sound and current draw if you exceed the limit. Current might really spike if it loses sync, so watch that.

Kevin, do these sine wave controllers use synchronous rectification?
Do they do regenerative braking?
 
Great find on low cost sine wave controllers, will be looking forward to the field reports.
 
It will be interesting to know how these sinewave controllers compare to the Lyen/Em3ev controllers in terms of performance, functions, programmability etc
 
this is why i ordered one, and i will try out and let all of you know. i've got several "regular infineons" and have a lot of experience with them. i guess those sine wave controllers are more or less the same as our well known infineons, but have a different firmware running.
that's why you can use the same programming software to set them up. so features should be the same as with square wave controllers.
i always was quite happy with those controllers. yes. throttle is quite snappy and kelly does that much better. maybe there this is somehow different with the new controllers?! don't think so, but i have some hope (for no reason though).
 
I'm sure we can figure it out when we test one :wink:

Synchronous rectification means the low side MOSFETs are turned on during the off part of the PWM cycle. Most simple controllers just use the body diode in the MOSFETs to conduct instead of turning them on.
By turning on the low side switches, the heat dissipation is greatly reduced during partial throttle.
 
izeman said:
If you tell me how to test that w/o special equipment (other than a DMM) I will perform those test once I've got the controller.

i think you would need an oscilloscope.


what richard is talking about is the way the mosfet deals with the current that continues to flow when the pwm controlled mosfet is turned on and off rapidly during the partial throttle level. this happens thousands of times per minute and as the throttle increases then there is less time that the mosfet is switched off since it is trying to push more current to make the motor go more.

when the mosfet turns off, the current that is 'induced' to flow into the motor wires by the current going ahead of it does not stop immediately but is sucked through the body diode of the mosfet on the loside and that current flowing through the forward bias of the body diode makes heat. IV heat, where the V is the forward bias of the body diode and I is the flyback current induced to flow through the body diode.

if you turn the loside mosfet on during this period, when the hiside mosfet is being modulated by the pwm, then the current can flow through the turned on mosfet with almost no resistance so the heating of the conduction channel of the mosfet is limited.

so you would look for the gate signal on the loside mosfet and have it trigger off the hi side mosfet gate signal. for this you need an oscilloscope. imo.
 
Back
Top