KU63 from BMSBattery on 16S LiFPO4

Will it hold? I'm hoping to try the 36V version of the controller. I can't seem to definitively prove if it can be overvolted on a 16S Headway pack (with BMS). LVC is not a concern, based on extensive Googling.

I guess if it blows I can try a spare KU123 on hand, with one of the shunt wires cut to halve the amperage to 15 max, which would be about right for a Q100.

Using a Q100 on 700C (fast model 328rpm).

Should be a fairly fast but non-torquey ride with one Cute.

I've run a KU63 on 15S, but not `16S. Seems iffy with the KU63's 63V cap.s.
Don't want to nay say, but it could be that somewhere in the low/mid 20's mph, that a 328 in a 700cc wheel will run out of power, become power-limited, as opposed to the more common rpm-limited. If you are an athlete, you should be able to pedal to 25 to 26 mph.
I'm thinking anything above 14S/15A will not l add much to the top speed. But given a long and steep enough downhill, the motor will still be trying to add power at 35 mph and above.
Any hills above 5%, you might as well just forget the motor.
A viable combo for a roadie type who lives in Flatland.

Thanks...maybe I will have to graduate to the KU93, as I thought the KU123 had two shunt wires one after cutting...unless I can substantially file it down a lot. I'd rather stick to 14S, but my fleet is standardized on 16S Ping and 16S Headways. The Headways are generally at 58 when charged and will drop to 53 pretty fast. I'm aiming for a "Russel type" build as you say, roady type, mild hills. Hopefully the hills won't be too bad. We'll see how it goes.

16s lifepo4 is under 60V charged so as long as your controller will support 60V it should work fine.

The KU63 uses 2SK4145 FETs usually, with an absolute maximum Vds rating of 60V. Running them at close to 60V (16S LiFePO4 hot off charge will be around 58.4V) will reduce reliability, as the overhead for spikes and power line ripple will have been reduced significantly. It may work, but it will reduce reliability in the long term, which may or may not be important to you.

Also, the power supply in the KU63 uses a 180 ohm power resistor to reduce the supply voltage to a safe value (less than 35V) for the voltage regulator and also to reduce heat dissipation in the regulator IC. This resistor will be too low a value for 50V + and there is a risk that you may damage the first stage voltage regulator with a higher than allowable input voltage. Again you may get away with it for a while, but it would be best to change this resistor for one of around 330 ohms. This resistor usually has a power rating of around 2 watts and does get warm in use normally.

May not be the case of your KU63 but some 36V controllers use 50V caps instead of 63V used with 48V systems. Check that first!
Also it seems you are trying too hard with the Q100. Your no load speed would be in excess of 60kph. Assuming the power would top out at 600W, unless this is a sporty road bike with tucked fit rider, you will be much better off at 36V which should assist you up to 45kph.

Thanks all. Sounds like I need more controller voltage-wise, and maybe some thicker wires on the motor itself. Not looking for torque here, just decent assist in the low to mid 20s mph range with an ultralight freewheeling motor. I'll put the KU63 on the shelf and use the chopped KU123 - I've used that a lot at 48V Ping xon a BPM - Its a good controller. Just need to limit the amps a lot. Luckily these units are cheap. This is a rather different build than I'm used to, focussing on "less is more" - maybe.

if you have the controller already, you can open it up to confirm the caps are not 63V. the battery is not always at the 60V level so that is mostly a concern IMO during the charging if the controller is connected to the batteries, the voltage usually drops off fast when the lifepo4 settles and loose charge floating in the electrolyte goes back to where it came from.

somebody else on here a while back did blow up their C'lyte mosfets which were 60V too when they used a new 48V battery. so it does happen like JH said. and you should change the input power resistor too. when he mentioned the 180 ohm input power resistor, that would be too low a value if you raised the input voltage. you can measure the voltages when you have it open and on the new battery and guess how much to change the input power resistor like he said.

I modified my order to 201 rpm and picked up a 48V KU93 as well, which I will file down a bit, and save the KU123 for a different project. Thanks everyone to put this project back to reality!
I'll run it on my small 10Ahr 48V Headway pack.

chvidgov.bc.ca said:

Thanks all. Sounds like I need more controller voltage-wise, and maybe some thicker wires on the motor itself. Not looking for torque here, just decent assist in the low to mid 20s mph range with an ultralight freewheeling motor. I'll put the KU63 on the shelf and use the chopped KU123 - I've used that a lot at 48V Ping xon a BPM - Its a good controller. Just need to limit the amps a lot. Luckily these units are cheap. This is a rather different build than I'm used to, focussing on "less is more" - maybe.

Click to expand...

Sounds like I need more controller voltage-wise

Click to expand...

No, you are not getting what I'm saying. Look at this sim. on Ebike CA;


12S(46V) X 18A = 620 Watts = 25.7 mph(no load speed)

14S(54V) X 15A = 620 Watts = 25.7 mph(no load speed)

Est.-16S(62V) X 10A = 620 Watts = 25.7 mph(no load speed)-This will come at the loss of all low RPM power

Not looking for torque here,

Click to expand...

Wrong again. Your top speed is set if you do not want to exceed the sustaiable power limits of the Q100(more on this in a moment).
This motor wheel configuration is POWER limited, NOT rpm limited! Increasing Voltage above 15S is pointless, as you do not need to turn it faster.
Increasing the current will increase the wheel/motor rotational force. Although that does not show up in this sim. it is there and in the real world, you will want as much of this as possible.
So what are the limits of the Q100?
The max. Watts it can sustain is debatable(I'm not willing to go there).
But what is NOT debatable is this;
Below and and at 600 Watt constant, the Q100 is quiet, smooth and happy.
Above this 600 Watt threshold, it becomes increasingly louder, a combo of a whine and a growl.
Above 700 Watts, it fairly well screams that it is not happy. It can be run this way, very carefully, by not maintaining that value for long.
Do you want a quiet assist motor or a screachy one?

and maybe some thicker wires on the motor itself

Click to expand...

Good luck on this one. There just isn't room to maneuver in there and it's doubtful it would make any difference. It's just a tiny motor, it is what it is and up-grading the wires isn't going to change that.
A note on the sim.'s no-load speeds.
Generaly, my experience with the sim., is that it's top speed numbers is dead on(a testiment to Justin's genius). But that is cases where the no-load comes about as a function of motor rpm. It cases where the NLS is a function of total power out-put, the number appears to be slewed upward. Unless I miss my guess(based on my Q100 328 in a 24"wheel @ 12 and 13S), that number would need to be reduced by 10%.
Back to the sim.-So what are the advantages of 12S vs. 14S.
Although not showing on the sim., 12S on 18A would have a little more pull off the line and up hills and help battle headwinds. Headwind are going to have a major impact with your proposed motor/wheel combo. Even a few extra mph will have you pedaling in ernest to maintain speeds above 20 mph. Same with inclines, it won't take much to start slowing you below your desired cruise speed.
The 14S/15A combo would allow you to pedal and contribute above the NLS(note the flat-line).


A maxium oft stated here on E.S., is; Gear down, up the Voltage.
But since your proposed motive force is already geared(by way of the wind and wheel size) very high, your mantra should be;Geared high, Volts down[up current].

P.S. Twin fast wind Q100's on your bike, running @12S/(2)16A shows a top speed of 30mph[say, real world, 27 to 28 mph)and climb most any hill.

The bottom line here is, as motomech is trying to illustrate, the ability of the motor to deliver the required power to sustain a given speed.

For example, a typical ebike will need around 850 W to sustain 30mph on the straight and level with no wind (a lot more if there's a head wind or gradient). The small Q100 won't sustain this power, it'll struggle to sustain half this power, TBH.

For this reason, increasing the voltage won't increase speed, but nevertheless there are some advantages in going down this route. If you use a controller with a selectable maximum speed switch, then selecting (and programming) a lower maximum speed, then using a high battery voltage, can result in much lower battery current for a given power and still keep the motor operating within limits. The benefits are probably a bit marginal, but would be beneficial if you chose to use a low maximum discharge current battery, as battery life could be usefully extended.

This latter effect is because all controllers work as high efficiency voltage to current converters, and there is no real efficiency hit from converting excess battery voltage to motor current.

I basically have to run it on 16S. I don't want to buy 36V batteries. Which would you prefer (201 or 328) on 700C? I understand I will have to sacrifice amperage to keep within 600watt power range, but would like some assist up to about 23mph.

Use the 201 rpm version and at 16S on LiFePO4 with a 700c wheel you should get around this speed.

Thanks for detailed information guys.

So far so good. I got my kit today from bmsbattery.com and installed it on my 700C Norco Yorkville Hybrid. The kit is very easy to install.

Its a 201RPM 36V Cute running on 16S 48V 10Ahr Headway pack (full charge 58.2V) with KU63 controller

I took it for a ride of several miles with no issues and was able to maintain about 23 mph with no problems on the flats.
As anticipated it is moderately gutless on hills, but will get me up big hills as long as I gear way down and peddle a lot, and it is still much
easier than with no motor. I like the very small KU63 controller and motor, and the Cute is VERY quiet for a gearmotor, about the same
as a 9C DD, growling a bit under heavier loads. I ran it sensorless with no problems. It freewheels well, perhaps not quite as well as an EZEE motor which I have on another bike.

I think if you keep the growling down, the combination will work ok on 16S. If I don't report back, it means the controller and motor are holding up at the
16S voltages. I also got a KU93 just in case the controller pops.

What amp shunt did you end up with? All well still? It sounds like they are slowly cheaping out all of the cute motors (all the Q128's, the 24v Q100's). You got yours last month, right? Still seem decent?

Everything is still good. I'm using the KU63 controller Q100 36V 201rpm, stock shunt, at 48 nominal (mostly around 52V). I've just had it on a fifty mile ride in moderately hilly terrain and its been excellent. I just don't unneccessarily overstress it, by keeping my gearing down on hills and pedalling. I like these motors on a nice light bike. They ooze efficiency...

chvidgov.bc.ca said:

Everything is still good. I'm using the KU63 controller Q100 36V 201rpm, stock shunt, at 48 nominal (mostly around 52V). I've just had it on a fifty mile ride in moderately hilly terrain and its been excellent. I just don't unneccessarily overstress it, by keeping my gearing down on hills and pedalling. I like these motors on a nice light bike. They ooze efficiency...

Click to expand...

Thanks for the reply. And no mods to the KU 63 to take the 52v, right? I was going to run 12s LiFePo4, but not I'm thinking 12s Lipo, and was worried it wouldn't work with the 36v KU63. If all I lose is the LVC, I guess I'm good with whatever I find for batteries.

I'm using the stock KU63 controller. I've got a spare KU93 which I could substitute if the controller popped, but it would be overkill with no additional benefit. It definitely was smoother with the Halls, but its nice to know it can switch over to sensorless if any the Halls die. I got mine back in October I think it was. I haven't checked motor temp at all on my last hilly ride, but it is cold now in Victoria, just over zero degrees. Around town, no worries about overstressing it. Losing LVC is a sacrifice I suppose, but I've standardized my fleet on 48V nominal all around. I want to use the same batteries on different bikes. You could get the pack level LVC back with a Cycleanalyst, but I don't use one for around Victoria riding.