A warning about KO controllers

[j bjork]

I saw this post on "Electric motorcycle builds" on facebook.
KO seems to be a company that sells upgrade parts for surron mainly, and I think I have seen them prazed here on the forum before.
There seem to be some sort of beef between them and another company, and some guy has opened up a supposedly hi performance controller from this KO company.

The guy thinks it is crap, and I must agree it certainly looks like he has a (or a few) point.
Especially looking at the powerstage.

A little from his post with some of his youtube videos, I dont know if I can link to the facebook post
What do you more electronics knowledgeable guys think?

Everything I said about KO Moto is dead on and I bought my own KO Pro controller to prove it!
https://www.youtube.com/watch?v=it6b8G936xI
https://www.youtube.com/watch?v=M7mJpAGCj1s
https://www.youtube.com/watch?v=d-YQlHYxKOM
https://www.youtube.com/watch?v=uey1HTMuISA
https://www.youtube.com/watch?v=DmayjD7wBkA

 

[j bjork]

I suppose this is the one:

https://www.ko-moto.com/product-page/ko-sur-ron-controller-purple

 

[liveforphysics]

The controller doesn't look too bad to me for an ebike controller. Less refined than an ASI, but above average design.

 

[larsb]

I didn’t see the videos, there might be obvious flaws on these controllers but from memory i think youtube id
”de-bodgery” is same person as electricgod - if i remember that correctly then he doesn’t really know electronics.

 

[j bjork]

i think youtube id
”de-bodgery” is same person as electricgod

It might very well be so, I got that feeling in a number of the facebook posts and one the videos that was manly an excessive rant. He could have made his point in a few sentences, but instead it took minutes.

 

[j bjork]

The controller doesn't look too bad to me for an ebike controller. Less refined than an ASI, but above average design.

Really? I thought the current passages looks quite horrible with gaps and no copper in some places on a controller that is supposed to do 800pA.
I would have liked to see the underside of the pcb also though, I think the current passages from the mosfets might be on the underside too?

A borrowed picture of a fardriver 84530:

https://endless-sphere.com/forums/download/file.php?id=317557

(sorry, I am useless on computers, I dont know how to get it to show without clicking on it)

So this is for 530pA, I suppose the current passages are on the underside here.
But it is a short pass and bolted down posts. It seems a lot better to me.
I cant really judge how good the design is, more how sturdy it looks.
Also a nice copper trace for positive or negative, I suppose it is something similar for the other polarity on the underside?

A picture from one of the youtube films:

Those loops from the copper traces seem to only connect to pcb traces between the fets with some extra solder.
I suppose there are copper on the underside, so that might not be a problem.
But still, 800pA on this design?

 

[liveforphysics]

It's tough to say how thick those copper bars are. I'm hoping they are 3+mm thickness and say 10-12mm width (between each parallel pair). If it's ~33mm^2 will heat at 800amps from 50degC to 100degC in 14seconds assuming no cooling.

In my experience with an 830Amp Sur Ron, it's very difficult to use the full phase current for more than say 5-6seconds at a time, because it's already reached top speed by then.

Lots of folks here are putting >100Amp phase current through 14awg phase leads. This would be the same current density as 1550Amps through a 33mm^2 bus bar. Certainly not stable for a continous duty cycle, but works fine for bursts of acceleration.

Also remember, it doesn't need random copper everywhere, it only needs it where current density is highest. One side of that controller has low current density as its power path comes from a single side to the FETs.

 

[mxlemming]

It's a horrible layout, not optimised for thermal, for switching or for current carrying capacity. I watched some of the KO "no hype just facts" nonsense video, the guy making the videos has no clue.

You can get 800A from that many TOLL FETs, just, but not continuous (few seconds) and it's anyone's guess how hot the jumper bar things get.

The KO motor does look like a serious upgrade over the stock surron or qs motor though. Wider rotor and a nice flux path optimised for MTPA. Probably generates a fair bit more torque.

Edit after LFPs post... The way the copper bars have been folded, there's no way they're 3mm. TOLL is 10mm wide, they're narrower than that and have lumps taken out the middle (again not likely possible on 3mm...) So I'd estimate more like 1-1.5mm with 10mm total width (5mm/jumper) so 10 to 15mm².

Then there's the single strips across the board. So like 7.5mm² but cooled by the IMS sinking the heat.

 

[larsb]

I suppose this is the one:

https://www.ko-moto.com/product-page/ko-sur-ron-controller-purple

Jeezus, i looked at the motors description on their site, what a load of marketing nonsense! It hurts my eyes to see so much fluff and so little useful information.

 

[toolman2]

The KO motor does look like a serious upgrade over the stock surron or qs motor though. Wider rotor and a nice flux path optimised for MTPA. Probably generates a fair bit more torque.

Hey mxlemming, curious about this nice flux path and assuming you mean potential for additional reluctance torque and wider rpm range from field weakening etc. -I've had a look for pics of internals without any luck, have you spotted some ?

 

[mxlemming]

The KO motor does look like a serious upgrade over the stock surron or qs motor though. Wider rotor and a nice flux path optimised for MTPA. Probably generates a fair bit more torque.

Hey mxlemming, curious about this nice flux path and assuming you mean potential for additional reluctance torque and wider rpm range from field weakening etc. -I've had a look for pics of internals without any luck, have you spotted some ?

https://youtu.be/iRstHnOCqNE

Long boring video but there are pics in there.

 

[KAUDIO]

This is a cheap Chinese product. It looks like the controller software of Nanjing FarDriver was copied, and as far as I know, the HEX code of Nanjing FarDriver was stolen by some people and used to produce controllers of other brands

 

[Dui ni shuo de dui]

I think the only question that really matters is "can it burn rubber"?
That's really all that matters, I've seen some controller with a nice layup that were absolute garbage, and some other with shit construction that performed very well.
Thing is, nowadays what matter most isn't so much the hardware than the firmware/software that runs it. This is usually the weak link on almost all controllers.
Mobipus, APT, more recently FarDriver, all have decent hardware but plagued with software issues.

So I guess we should wait for people to actually try them and then decide wether it is a good or a bad product. Doesn't seem fair to issue a warning on a product without giving it a proper test first.

 

[KAUDIO]

I think the only question that really matters is "can it burn rubber"?
That's really all that matters, I've seen some controller with a nice layup that were absolute garbage, and some other with shit construction that performed very well.
Thing is, nowadays what matter most isn't so much the hardware than the firmware/software that runs it. This is usually the weak link on almost all controllers.
Mobipus, APT, more recently FarDriver, all have decent hardware but plagued with software issues.

So I guess we should wait for people to actually try them and then decide wether it is a good or a bad product. Doesn't seem fair to issue a warning on a product without giving it a proper test first.

In China, I think the controller with the highest playability is mobipus, followed by the fardriver, their software is doing well, it is always updated, and there are more parameters to adjust.