Conversion of a Bultaco Sherpa

[Elektrosherpa]

Today I put everything together, and went for a test ride.

In contrast to the first time (with the BMS built in by the battery seller),
it didn't last only 1,5 km (~1 mile), but at about 20 km.

Driving was OK, top speed was a little lower than last time (maybe 70-75 km/h)-
but I must admit that I was a bit more careful with the throttle this time,
because I didn't want to risk pushing the bike back home again... :?

So basically things do work now.

But still several points need to be looked at more closely:

- the motor got quite hot, in my opinion. I felt the heat radiation and when I was back home, I measured 65°C at the outside of the motor case. The battery didnt go over 30°C, and the Controller went up to 32, 32° C, what doesnt worry me at all. But the motor temperature appears quite high to me. Is this normal ? I dont remember to have read anything about the temperature of a QS138 ...

- the handle bar display from JK is a bad joke. At home, in my workshop, it looked good, but outside in bright daylight it is impossible to read it. Way too dark.

- the app (or my smartphone) lost its bluetooth connection to the BMS after the first few 100 meters. I couldn't get it connected again, until I was back home....

 

[larsb]

Motor temperature is strange. I have not gotten my motor that hot regardless of driving. Could be a false positive hall sensor combination but i think i saw your noload current at 4.5A? Then that’s likely not the issue.

Could be that that the kelly can not drive IPM motors like the QS 138 efficiently (these motors need field weakening for optimal drive). Could be worth it to try changing the switching frequency in some steps.

 

[Elektrosherpa]

Motor temperature is strange. ...
Could be worth it to try changing the switching frequency in some steps.

Are you talking about the setting in the Kelly Controller Settings, which I have labelled with a red arrow?

(these are not exactly my settings- it is a screenshot from GonZo's video about Kelly controller settings,
but my PWS setting is similar)

 

[_GonZo_]

I am happy that it is running well, 75Km/h is a hell of a lot for that little thing.
65ºC case temperature for an inrunner motor after 30m ride is quite reasonable. Problems may start to happen if case will go over 80ºC+.
Things you can check/try in order to see if you can improve that:
- Do a new hall sensors identification angle procedure, just in case that it is wrong (I do not think it is wrong, because then it will not be working properly...)
- Change to 10kHz the PWM frequency that makes this kind of motors run a bit more efficiently.
- Reduce the amount of e-brake, that will give more room to the motor to cool.

If it does not work as you want, then limit the power to the motor.
Increase the gear ratio.
Or change to a more powerful motor.

 

[larsb]

Sorry but you don’t know what you’re talking about. OK that 65deg C is nothing to worry about but also the qs138 is a 11kg motor with 5mohm phase resistance. I’ve run it at 300A battery current / 600A phase for drag races without getting it hot like elektrosherpa here, also done road driving at 110kph top speed. 75kph is possible with a 80100 motor.

 

[_GonZo_]

Sorry but you don’t know what you’re talking about. The qs138 is a 11kg motor with 5mohm phase resistance. I’ve run it at 300A battery current / 600A phase for drag races without getting it hot like elektrosherpa here, also done road driving at 110kph top speed. 75kph is possible with a 80100 motor.

I had no idea what motor he is running. I just gave the normal recommendations for a motor that heats up with a Kelly controller.
Anyway the size of it and its Ohms does not tell me what power is it able to give and at what regimens.
Is there any spreadsheet with motor characteristics characteristics.

 

[larsb]

I am happy that it is running well, 75Km/h is a hell of a lot for that little thing.
..

- Change to 10kHz the PWM frequency that makes this kind of motors run a bit more efficiently.
..
65ºC case temperature for an inrunner motor after 30m ride is quite reasonable....
Or change to a more powerful motor

It certainly seemed here that you know what kind of motor it is? There's a long thread on this motor if you search for it. Next time you give generic advice not related to the issue at hand you might want to take a step back and think about if it will help. "My 300 gram inrunner gets hot" and "my 11kg motor gets hot" aren't really deserving the same answer, are they?

 

[Elektrosherpa]

Thank you for the replies.
So you both agree that the motor should NOT reach a temperature of 65°C.

As far as I know, for the Kelly controller, the recommended setting for "motor temperature too high" is 130°C,
and the recommended setting for "resume motor temperature" setting is 110°C-
this sounds to me like my temperature is not (yet) critical.
But anyway a lower motor temperature is desireable.
So I will try lowering the PWS setting.

Changing the motor is not an option-after all I read about the QS138, it should be powerful enough for a 100kg bike.
Even with a 100kg rider on it. :wink:

E-brake is not activated at all (yet). I am living in a flat region, so I do not really need it. I am planning to try it once in a while, but this is not on top of my to-do-list.

I have done another test ride today, and the power really seems sufficient for my purposes.
The BMS App showed Amperages up to a maximum of 250 A
(what is at about the maximum my battery is supposed to deliver: 54 Ah, 3 C permanent and 5 C maximum discharge)

I also tested some slopes, and didnt have any problems driving up.

Since I am using the Bultaco mainly for walking my sighthound in the fields around my home,
a top speed of 70km/h is sufficient, too (the dog can do 60 km/h, and I just want to be able to keep up with her :wink: )
Cant upload videos here, but are 2 gifs:

 

[_GonZo_]

So you both agree that the motor should NOT reach a temperature of 65°C.

It is not certainly very high but high after a mild drive.

E-brake is not activated at all (yet). I am living in a flat region, so I do not really need it. I am planning to try it once in a while, but this is not on top of my to-do-list.

If you have some kind of freewheel, then perfect. If you don't have freewheel sistem then it will be interesting that you use some regen when braking, it is free energy even if you live on a flat area.

The BMS App showed Amperages up to a maximum of 250 A

Yep that is normal your controller is able to pull over 300A at peak.

Do you have the technical details of the motor?
What is the reduction you have installed?
What is the wheel diameter?

 

[Elektrosherpa]

Do you have the technical details of the motor?
What is the reduction you have installed?
What is the wheel diameter?

As pointed out in the beginning of this thread, the motor is the well-known QS138 (3 kW nominal, many reports available saying that it can handle up to 15 or even 20 kW).
The transmission ratio is 11:70 (= 1:6,36), the rear wheel is a 4" tyre on a 18" rim, this results in a wheel diameter of at about 2026 mm .

 

[Elektrosherpa]

Back from another test ride.
30 km, Voltage went down from 83.5 V to 74.5 V.
From observing the Amperes drawn while riding,
it looks like the battery range will be something between 60 and 120 km,
depending on the speed I am driving.
At slow speed (~30 km/h) the power consumption is quite low (10-15 A) and might allow a range of 120 km.
At higher speed (60-70km/h) the consumption is like 35-50 A, what would equal a (calculated) range of at about 60 km.

About the temperature:
It stayed lower this time ("PWD" Setting lowered to "10").
AND:
a more detailled temperature measuring showed me
that only the right cover of the motor (with the "QS motor logo") becomes so hot-
the central housing of the motor (with the cooling fins) stayed low, at 30, 31°C,
same as the rest of the components (battery, controller).

 

[_GonZo_]

As pointed out in the beginning of this thread, the motor is the well-known QS138 (3 kW nominal, many reports available saying that it can handle up to 15 or even 20 kW).

Nominal power means what power the motor can maintain indefinitely without overheating at near max revs and without any forced cooling.
QS motors are usually able to give a bit more than rated if properly cooled but 15Kw or 20Kw are only possible on peaks probably for that motor.
Do you know more details like the KV (Revolutions per volt)?

the rear wheel is a 4" tyre on a 18" rim, this results in a wheel diameter of at about 2026 mm

A 2m diameter wheel seems to me that it is not correct :lol:
Can it be something like 560mm?

The transmission ratio is 11:70 (= 1:6,36)

With 4000rpm at motor and a 560mm diameter wheel you will be getting around 66Km/h which is about right. We need to know the motor KV in order to check this.

The numbers from your last ride test looks quite right to me.
The area where the fins are located is where the motor windings are, so they are supposed to be the area that get more hot.
Is it the center area of the motor cover where it gets hot? because if so it could be a damaged bearing the one that is overheating.
Anyway 30-31ºC is low temperature for a electric motor.

 

[Elektrosherpa]

Do you know more details like the KV (Revolutions per volt)?

the rear wheel is a 4" tyre on a 18" rim, this results in a wheel diameter of at about 2026 mm

A 2m diameter wheel seems to me that it is not correct :lol:
Can it be something like 560mm?

Is it the center area of the motor cover where it gets hot? because if so it could be a damaged bearing the one that is overheating.
Anyway 30-31ºC is low temperature for a electric motor.

I don't know the KV, but I will look into that-should be possible to find that. This motor is frequently used.

"Diameter" was nonsense ... :lol: I meant circumference, of course.
And this is the value needed for calculating the rotation/distance/speed relation.

Locating the heat source is not possible since the cover is made from aluminium, which is a very good heat conductor.
But I was thinking the same direction: side cover - center - bearing.
I bought this motor new, directly from the producer (QS motor), together with the controller they said to be the best suitable for this motor... But after my recent experiences with Chinese "quality", I would not be really surprised to have received a motor with a factory damaged bearing

Anyway the recommendation from QS was not ideal:
The KLS7245N, which they recommended, turns out to be the weakest component of the whole vehicle -
maximum permanent current 100 A. The battery could do 170 A, and the JK BMS 200 A.
Regarding peak Amperes, the battery is the weakest part: 270 A. KLS7245N could do 350 A, and the JK BMS, too.

 

[j bjork]

The kv is about 60.
Cant you get a temp reading from the controller?
If you cant read it while riding it should be possible to connect the computer right away when you get back.

 

[Elektrosherpa]

Today another ~40 km test ride (still with first battery charge).

Tried some slopes-no problem.

Then went on our local motocross training ground and did a few rounds there-
not in motocross speed, of course (nor me nor the bike are crossers),
but faster than trials speed. Worked well, too.

Cant you get a temp reading from the controller?

When I arrived back home I connected the Notebook to the Kelly Controller
and had a look at the motor temperature reading:

(it was 82°C when I looked first, until I got the phone ready for taking the pic it was down to 76°)

So the motor definitely gets hot.

Here screenshots from my actual settings, maybe someone will find a mistake/reason for getting hot:





And one more thing:

The kv is about 60.

If I got that KV value thing right, this means, with the 69V battery voltage shown in the screenshots above, I should get a max. rotation speed of 4140/min.
But when I tried it with the rear wheel in the air, I got only something like 3500, 3600...
Any comments on that? (maybe in relation to the settings)

 

[larsb]

Check the values for three speed switch, if you don’t use one then set midspeed and lowspeed at 100%.

 

[_GonZo_]

Anyway the recommendation from QS was not ideal:
The KLS7245N, which they recommended, turns out to be the weakest component of the whole vehicle

I think it is the adequate one, smaller one will not be able to take all the motor potential, and bigger one will burnout the motor.

(it was 82°C when I looked first, until I got the phone ready for taking the pic it was down to 76°)

It is not extremely high but certainly it is high.

About the KV it is an important information to know, because the problem maybe there.
It is possible that your motor is "choked" meaning that is turning at too low revolutions in order to be efficient and trying to do more work that what it is capable of.
As per information given here:

• Motor KV: 60 rpm/V
  Motor revolutions at 72V: 60 rpm/V x 72V = 4320 rpm
  Reduction: 11:70 = 1:6.36
  Revolutions at Wheel: 679.2 rpm
  Wheel diameter: 2026mm / 3.14 = 645mm
  Vehicle speed at 679.2 rmp at wheel: 82.5 Km/h

So you should be able to reach around 80 Km/h.
To move a motorbike at 80Km/h you need around 4000W to 6000W
So it is possible that the motor does not have enough power (it is a 3000W motor) to be able to move your vehicle at that speed.
So if the numbers up here are right, then you are "choking" the motor and that is why it overheats.
The solution will be to increase the reduction.

 

[larsb]

Seriously.. as many (including myself) have used this motor at 20kW and above i don’t think it’s ”choked” at 4kW

Do you have some affiliation with Kelly by any chance?

 

[_GonZo_]

Do you have some affiliation with Kelly by any chance?

No I do not have any affiliation with Kellycontrollers or QSmotors.
But I know their products quite well.

Seriously.. as many (including myself) have used this motor at 20kW and above i don’t think it’s ”choked” at 4kW

Yes, it is possible to run an electric motor over its nominal power for short periods of time.
But, have you been able to maintain this 3Kw motor working for hours (NOT minutes) delivering 4Kw without any modification or added cooling?
The way that the nominal power of a motor is found is in a power bench where the power demanded to the motor is increased until it stays stable (heat produced = heat dissipated) at usually an ambient temperature of 22ºC.

 

[larsb]

I know this, and it aint what ”choked” means.
Maybe the best advice is to go to a votol em-150 since Kelly seems like a poor match.

I also know a bit about kelly kls since i commuted on five of them for some years. In the end i stopped using one since fany at Kelly concluded my ”motor was too powerful”. Now that’s a sure sign of not having designed a good enough controller!

 

[Elektrosherpa]

So you should be able to reach around 80 Km/h.
To move a motorbike at 80Km/h you need around 4000W to 6000W
So it is possible that the motor does not have enough power (it is a 3000W motor) to be able to move your vehicle at that speed.
So if the numbers up here are right, then you are "choking" the motor and that is why it overheats.
The solution will be to increase the reduction.

This would mean that it is not possible to use an electric motor in a motorbike efficiently without a gearbox,
because I do not drive at a constant speed all the time... ???

 

[_GonZo_]

This would mean that it is not possible to use an electric motor in a motorbike efficiently without a gearbox,
because I do not drive at a constant speed all the time... ???

Yes, a gearbox will increase the efficiency of the system as the vehicle is driving at different speeds.
But will increase weight, complexity, maintenance... So it is a compromise between one or the other.
Anyway brushless motors are very efficient in a variety of range. They are usually more efficient close to the maximum motor rpms.
That is why I was proposing to increase the gear ratio in your motorbike. (Assuming that my calculations before are right)

There is other ways to have something like different gears with an electric motor: like switching from star to delta connections, or switching from a series to a parallel winding connections, or like Toyota and Tesla that they use a motor that works in two different ways depending on the speed and torque demanded.
This other ways as well increase complexity but in the electric/electronic area.

 

[Elektrosherpa]

That is way I was proposing to increase the gear ratio in your motorbike.

I am afraid this is not possible:
I can not find a sprocket smaller than 11 teeth.
And the 70 teeth rear sprocket is the maximum I can fit without getting into contact with the swing arm etc.

 

[Elektrosherpa]

I modified my battery box/holder (added a mechanical locking mechanism for the opening side cover),
and at this opportunity I moved the temperature sensor of my controller display/speedometer
from the controller (where I had put it first) to the motor, to observe the motor temperature more closely.
The controller does not seem to have any temperature issues, so I think the new location makes more sense.

 

[Dui ni shuo de dui]

Good job on swapping your BMS, plus it seems to be an interesting one I haven't heard of before, which does active balancing, a relatively rare feature on cheap BMS :thumb:

But don't forget what I told you before: you need to make sure your cells can deliver that kind of current safely.
This means you need to get the specifications for your cells in order to program your BMS to actually protect them, otherwise it will use some random configuration and there will be no way to know if the BMS will actually protect in some cases.

If you can't get the specs of your cells, which is very possible because chinese suppliers aren't always the most service oriented people, then I suggest you very, very closely monitor your battery pack and check as often as possible all your cells balance values. Especially check them before charging.
Check that they remain always within 0.010v - 0.030V of each other The less difference between cells the better, and anytime you start having 0.100V or more it's time to stop using the pack and repair it. :wink:

How much voltage sag do you get when you accelerate ?
-What's your typical otal pack voltage when the bike does nothing?
-What's the minimum total pack voltage value you see when you accelerate at full power with yourself on the bike?

This can sometimes be a good indicator of a potential problem with the battery so try to measure those asap if not done already.