Kelly Controller with hall sensored outrunner problems

[eDahon]

Ok so I went with the Kelly KBS24121L,50A,12-24V, Mini Brushless DC Controller. I have the equal zero design hall board and Turnigy SK3 6374 149kv

http://www.hobbyking.com/hobbyking/stor ... duct=18184

Lastly I am using 5s 5000mah lipo 20c

so I figured out the proper hall signal phase wire combo and the no load current is about 2.5 amps. It runs smoothly if I throttle up slowly but often cuts out at full throttle. I also can brake harshly if I throttle down too quickly. And when it is running near full throttle it hiccups violently and then will run smoothly and then hiccup again. I will then get a fault code on the Kelly controller. It is internal reset error

I was stuck on the fact that it was a problem with my timing or my sensor or board but I have now ruled that out. When I attached the motor to an amped bikes ebike controller with 36v lipo it ran extremely smooth no problems at all. I am now confident that it is a controller problem although I cant rule out the battery although it does seem unlikely particularly at the low current that the motor is pulling with no load and there is no noticeable voltage drop. It is at 19.2 volts

Do you think I have a faulty Kelly controller or is it a setting issue? Or is there any fundamental reason why this controller would not work?

Any help would be greatly appreciated!
Thanks!

 

[gwhy!1]

From the sounds of it , it sounds like the timing is not right or wrong hall/phase combination.. what hall sensor is it ? do you have a link ? on 5cells I wound have thought it should be more like 1-1.5A

 

[eDahon]

its this kit:
http://e0designs.com/documentation/hall-effect-sensor-board-adapter-installation/

No I believe the hall sensor phase wire is correct. The timing may be slightly off but nothing substantial to cause it to behave like this. At least from what I can tell. I had assumed it was what you suggested as well but when I hooked it up to MY AMPEDBIKES controller it ran flawlessly and no, this was not in sensorless mode. these controller only run sensored. Of note that was at 36 volts with a different battery so I cant rule out battery as the issue. May test that next.

 

[gwhy!1]

its this kit:
http://e0designs.com/documentation/hall-effect-sensor-board-adapter-installation/

No I believe the hall sensor phase wire is correct. The timing may be slightly off but nothing substantial to cause it to behave like this. At least from what I can tell. I had assumed it was what you suggested as well but when I hooked it up to MY AMPEDBIKES controller it ran flawlessly and no, this was not in sensorless mode. these controller only run sensored. Of note that was at 36 volts with a different battery so I cant rule out battery as the issue. May test that next.

What was the no load current with the other controller ?..
With the 17degree spacing for the hall sensors the cutting out was much more likely to happen , but it will very much depend on the controller check this thread out
http://endless-sphere.com/forums/viewtopic.php?f=2&t=16910

edit: also do the kelly support the speed that you are running the motor at ?

 

[eDahon]

Up to 40,000 electric RPM standard. Optional high speed 70,000 ERPM. (Electric RPM = mechanical RPM * motor pole pairs).

the lipo is about 22v and motor is 149 kv so that's about 3000 RPM. Motor poles are 14 or 7 pairs so it comes out well under 40,000 electric RPM.

Im not convinced anymore it has anything to do with the halls. Think it has something to do with this controller since it worked so well on my amped bikes controller. Unfortunately the amped bikes is 36v and I want to run motor with 5s lipo.

 

[eDahon]

What was the no load current with the other controller ?..
With the 17degree spacing for the hall sensors the cutting out was much more likely to happen , but it will very much depend on the controller check this thread out
http://endless-sphere.com/forums/viewtopic.php?f=2&t=16910

edit: also do the kelly support the speed that you are running the motor at ?[/quote]

I didn't check the no load but I was using 10s lipo with the amped bikes controller versus 5s with the Kelly controller. Obviously it was spinning quite a bit faster and assuming it was drawing more current.

I read somewhere that these low inductance rc outrunner motors can be a problem with the Kelly controllers as voltage spikes can trip the current limiter. That being said when I had my what meter didn't noctive any current spikes unless the what meter doesn't sample fast enough to pick it up.

 

[eDahon]

so have the no load current down to about 1.4amps so maybe im crazy..maybe it is a timing issue. No hiccupping but doesn't always want to start from dead stop..starting to seem like a sensorless setup.. :lol: but definitely making progress....

 

[gwhy!1]

so have the no load current down to about 1.4amps so maybe im crazy..maybe it is a timing issue. No hiccupping but doesn't always want to start from dead stop..starting to seem like a sensorless setup.. :lol: but definitely making progress....

I could never get the full potential from a outrunner motor with a hall setup @ 17 degrees it would either cut out,have starting problems or higher noload current, you could, with a lot of patience get it to some point where it becomes a compromise but never 100%.

 

[gwhy!1]

what you could try ( that sometimes helped ) is to move the complete hall sensors by one slot, you will have to find the correct wire combo again to get it to run the direction you want ( nothing to loose by trying it).

 

[eDahon]

Got me a Lyen controller and it works great. Super smooth and good torque. Something with the Kelly controller and the Hall Sensors on this outrunner just did not go well together.