QS 3000w midmotor

[PITMIX]

I made a mistake in my previous message, my engine probe is indeed a KTY83 not a KTY84. I have 977Ohms to 21,7°c. It is written in the characteristics, the motor works normally between 70 and 120°c with peaks at 150°c. You can set the protection to 130°c but I use it at more than 250A continuously with a transmission ratio of 5 and a 230kg motorcycle + rider without it exceeding 70°c internally.

 

[Elektrosherpa]

Just now I got a reply from QSmotors:
"Motor temp sensor 2 is correct, but it is not very precise"


So indeed the sensor in my motor is "type2" (KTY83), as I was assuming.

Now it would be interesting to know in which way the inaccuracy develops with higher temperatures.

I am an absolute electronics dummy, so it is not clear to me:

Will there always be the 10°C difference which I measured?

Or is this inaccuracy developping in a percentage of the real temperature?

20°real vs 31°displayed means a deviation of ~30% too much -
in this case it might be a solution to set the "overheat temperature" in the controller to the maximum which it allows (170°C).
If the real temperature is 30% lower, this would mean REAL 114°C

 

[Elektrosherpa]

...I use it at more than 250A continuously with a transmission ratio of 5 and a 230kg motorcycle + rider without it exceeding 70°c internally.

I am using not more than 100 A continuously (250 A max).
Transmission ratio is 11/70 (6,4).
Motorbike and rider together weigh less than 200 kg.

So all figures are "better"(easier for the motor) than yours - this means my temperature is definitely too high.

The question is:
WHY ??

 

[PITMIX]

Probably because you use your engine longer on me. I rarely use it for more than 20min. The probe may not be super precise but in no case 2x the value. Can you measure the resistance at room temperature. I have 977ohms at 21°c.

 

[larsb]

...I use it at more than 250A continuously with a transmission ratio of 5 and a 230kg motorcycle + rider without it exceeding 70°c internally.

I am using not more than 100 A continuously (250 A max).
Transmission ratio is 11/70 (6,4).
Motorbike and rider together weigh less than 200 kg.

So all figures are "better"(easier for the motor) than yours - this means my temperature is definitely too high.

The question is:
WHY ??

There are some things to check in your thread, right?

 

[Elektrosherpa]

Probably because you use your engine longer on me. I rarely use it for more than 20min. The probe may not be super precise but in no case 2x the value. Can you measure the resistance at room temperature. I have 977ohms at 21°c.

The overheat problem appears even after 10-15 minutes, if I am not very careful with the throttle...

I will try measuring the resistance.

btw- QSmotors recommended me to reduce the "current percent" and "battry limit" settings of the KLS7245N from 100% to 50% ...

 

[PITMIX]

I do not believe it. Your motor has no real overheating or then there is a problem with the phases. Maybe it also has a bearing Big problem.
I also use this engine 15 min 500A phases with Field Weakening activated 300A and regenerative braking to 200A, on a car without license 450kg and the motor does not heat above 70 ° C.
https://youtu.be/LbGlUO0zu9s

 

[PITMIX]

btw- QSmotors recommended me to reduce the "current percent" and "battry limit" settings of the KLS7245N from 100% to 50% ...

I did a little research from your controller and this is what I found. It is not possible to set your controller for a KTY83 probe since it only accepts 2 types of probe KTY84-130 and KTY84-150.

Excerpt from description:
http://www.cnqsmotor.com/en/article_read/KELLY%20KLS7245N%2030V-72V%20350A%20SINUSOIDAL%20BRUSHLESS%20MOTOR%20CONTROLLER%20for%204000-5000W%20Electric%20Motorcycle%20E-scooter/1021.html
(16)Configurable motor over-temperature detection and protection with the recommended thermistor KTY84-130 or KTY84-150.
KTY 83 xxx probes have a value of 1000 ohms at 25°C. The KTY84 xxx probes have the same ohmic value at 100°C, this is where the controller reading error comes from.
The only option you have left is to replace the motor probe with this model :
https://www.reichelt.com/de/fr/capteur-de-temp-rature-40-300-c-kty-84-130-p9604.html?CCOUNTRY=445&LANGUAGE=fr&&r=1

[youtube]SRLI41-DkzM[/youtube]

 

[Elektrosherpa]

Can you measure the resistance at room temperature. I have 977ohms at 21°c.

Sorry for the stupid question, but HOW do I measure the resistance?

The temp.sensor of the display (which I am using to monitor the outside temperature of the motor) has got 2 cables, so I can measure between them.

But I have only 1 cable with "temp.signal" coming out of the motor.

 

[larsb]

Between temp sensor wire and ground wire (black)

 

[PITMIX]

You can measure between the probe wire and the GND of the motor hall sensor plug. But if you read my previous post you will understand that your overheating problem is actually is a compatibility problem between the controller and the motor sensor.

 

[Elektrosherpa]

You can measure between the probe wire and the GND of the motor hall sensor plug. But if you read my previous post you will understand that your overheating problem is actually is a compatibility problem between the controller and the motor sensor.

Yes, I did understand that
(see my long reply in my build thread)
I put it there to get things more "bundled", to be found more easily, in case someone else will have similar problems.
There I also pointed out that information coming from QS are not consistent at all, what makes it even more difficult to find a solution.

Anyway, before changing the sensor, as you recommended, I would like to measure the resistance of the actual sensor.

 

[PITMIX]

Use a 1000 ohm resistor and plug it there between the probe wire and the gnd on the controller hall sensor plug. If you get 20°c, your controller is made for KTY83 probes, if you get 100°c, it is made for KTY 84.

 

[j bjork]

Here is a claim for 40Kw output with a qs 138 90h.
I cant get in to be possible in his setup, take a look:
https://www.facebook.com/groups/electricmotorcyclebuilds/permalink/2546056442198378/

 

[Jrbe]

Are the 2 sensors a constant ohm difference through their temperature swings? You might be able to add a resistor inline if yours is a 1k and it should be a 2k and they have a shifted 1k range difference.
There's no firmware update to work with that sensor or a way to define the temp sensor range with that controller besides the pre-defined sensors?

 

[PITMIX]

Installing a new temperature sensor on these motors is child's play. It is better to install a probe that is exactly what the controller accepts.

 

[larsb]

Here is a claim for 40Kw output with a qs 138 90h.
I cant get in to be possible in his setup, take a look:
https://www.facebook.com/groups/electricmotorcyclebuilds/permalink/2546056442198378/

Agreed. And what kind of engine dyno gives 55hp in 350ms? From noload max speed down to zero rpm in one extremely short brake?

 

[bikerpete]

Has anyone got any photos of the 70h opened up? I found one of the rotor & one of the stator, but not together and with no housing at all.
I'm mostly interested to see if there's any dead space in there that I can either remove by making a new housing/shaft, or fill with more rotating mass.
Preferred option is to make a shorter length housing.
Extra preferred would be a shorter housing & a double ended shaft so I can put a flywheel one end and drive the other. Not sure how hard it would be to replace the shaft.
In an absolutely perfect world I'd have a shorter motor, double ended shaft and higher kV windings.
Sounds like a completely different motor I know.

 

[amberwolf]

There are some in this thread on the last page, don't know if they're what you need, but you could ask for other pics if he hasn't got it back together yet.

Suzuki RMZ 250 EV Restomod

Picked up this 2007 Suzuki Rmz 250 frame and swingarm. Also came with the triple clamp and rear shock. First priority is to get this into a rolling chassis. Will restore this frame before mounting the motor and figuring out which battery to build or buy. I’ve ordered the front fork tubes...

endless-sphere.com

 

[SlowCo]

In an absolutely perfect world I'd have a shorter motor, double ended shaft and higher kV windings.
Sounds like a completely different motor I know.

What are you trying to achieve? Sounds like you might be better of with the QS165 motor which is shorter but taller/bigger in diameter. Or even a hubmotor converted to mid drive?

 

[bikerpete]

What are you trying to achieve? Sounds like you might be better of with the QS165 motor which is shorter but taller/bigger in diameter. Or even a hubmotor converted to mid drive?

I've got a few goals:

• A very significantly higher rotational inertia. I've currently got a 20mm thick steel flywheel the full diameter of the outer motor casing and it's getting close to enough. Going to 30mm thick would probably not be overkill. It's a problem putting that on the same end as the drive because of the excessive shaft overhang. I've currently got a fairly large proportion of the pulley width overhanging the end of the shaft.
• The overall length of the motor is getting too long with the flywheel added. I was kind of hoping there was some excess space inside the motor that could be recovered. The photos from the RMZ thread linked above suggest not much.
• Higher max RPM is of great benefit to increasing rotational inertia (inertia goes up by square of RPM), it also has benefits for me relating directly to a wider rev range. My QS138 on 48V is frustratingly slow. I'd like to go to 72V but finding a suitable controller becomes harder and/or expensive (I need an "idle" facility that maintains a more or less constant idle even with somewhat varying loads).

I hadn't really looked at the QS165 because it's "5kw" so I assumed bigger and heavier. Looking now it seems it's actually lighter than the QS138 and maybe even higher RPM. I'll need to hunt up the dimension drawing to see how it might fit. Thanks for the pointer.