New Kelly pseudo FOC controller series

[izeman]

izeman now you have tried it with a hub motor, what is your verdict on the torque delivery compared to a infineon 18 fet .

can't really compare as i only tried to see how noisy it is. the hs3540 in a 20" rim is quite slow at 50V, and i didn't tune the phase/battery amps. so it was around 3kW battery and 120A phase. and it was very nice and smooth to ride. i would need to do a side to side comparison with same values to give a correct answer. but i was quite happy with it

 

[strichter76]

- max output fre is grayed out (500), are u sure u have installed latest software?
normally you should be able to put 1000 there.
- motor poles range is only supported up to 32 motor poles. (2-32), but in my experience it does not affect the behaviour of the engine at all.
- did u unplug brake signal? my engine did not start before unplugging the brake signal 12v.

 

[mistercrash]

Those are 51 teeth, poles are number of magnet poles(magnets in any non halbach array)/2 I believe.

I had to look up ''halbach array'' to know what it means :lol: My magnets are alternating poles and there are 46 of them. The instructions say to take one pair of poles (in my case 23) and multiply by 2. So I will try 46 in my settings and see what happens.

Thanks
Ray

 

[izeman]

yes. it's poles and not pole pairs. so it almost must be a multitude of 2.

 

[mistercrash]

Success! Thanks guys. I took it out for a spin to the end of the street and back. Lots of torque and super quiet.

 

[izeman]

np. great to hear. so what WAS the problem? so we can add it to the knowledge base?

 

[mistercrash]

np. great to hear. so what WAS the problem? so we can add it to the knowledge base?

The problem was me :lol: Is there a thread where we can post stupid stuff we do sometimes? 'cause I got a good one Here it is. There are four plugs coming out of the KLS7250D. I looked at one and saw male connectors in it so I think I have to install female connectors in the opposite plugs, which I did. The one plug I looked at had male connectors but the other two had female connectors. With female connectors in opposite plugs, there was not much going on :lol:

 

[SplinterOz]

np. great to hear. so what WAS the problem? so we can add it to the knowledge base?

The problem was me :lol: Is there a thread where we can post stupid stuff we do sometimes? 'cause I got a good one Here it is. There are four plugs coming out of the KLS7250D. I looked at one and saw male connectors in it so I think I have to install female connectors in the opposite plugs, which I did. The one plug I looked at had male connectors but the other two had female connectors. With female connectors in opposite plugs, there was not much going on :lol:

Glad it is working!! I must say I am not a fan of the new plugs .

 

[fany]

I believe the Kelly is sensored only ?

i guess so. at least it's sensored for starting. and i guess it uses halls for full rpm as well. but this is something that fany would need to confirm.

Our controller is always based on FOC control from starting.
But the starting FOC is discrete.After speed up,the FOC is continuous type.

Sorry,our KLS controller can not support sensorless BLDC motor control for now.

And KLS-8080I can also support the motor with Sin/Cosin speed sensor.The part number is KLS-8080IPS.

P.S We are on golden week now.

 

[fany]

hi, i just unplugged the brake signal cable from pin1 (motor-temp). after unplugging, the engine spinned up to 2200 "motor-speed" (much faster than before).
voltage remained at 57v all the time instead of declining sharply as it happened before.

i am not sure how to connect my external 12v brake system to the controller, i expect the kelly waits for 12v on pin1/temp-sensor and then reduces power to the engine and starts regen.
in my case, with my scooter, 12v is delivered normally. during braking, 0v is delivered. no idea how to get this adapted to my scooter.

pin1 is 12V brake switch input port or motor temp input port.
You can not use both functions at the same time.
For example,the brake will be activated if 12V of controller on pin11 is input into pin1 port by brake switch.
This brake switch can do regen brake also.You just need to release throttle completely first.Turn on the brake switch will trigger the regen function.
And the regen function can be disabled or enabled and adjusted in the APP.It is called Brk-SW Brk% item.You can adjust it to 5.
If you have an external 12V,you just need to make sure this 12V gets the same ground as the controller.Then you can use 12V for this brake switch.
We have enabled the wiring diagram for 12V brake switch/motor-temp port in the KLS-S manual.There are three kinds of drawing for this port.

 

[fany]

SplinterOz,
I tried what you described twice, maybe there's something I'm doping wrong. I'll have to try again.

izeman,
Yes, 51 poles, 17 per phase. Typical 500W Chinese scooter motor I was told.

Thanks
Ray

EDIT: I just tried again and the motor does not move three times, all I hear from the motor is ''clicks'' for six times and then stops. I get no ''reset error'' screen. I turned off the controller anyway and restarted but the 170 is still in the ''Motor Identity En'' field.

Obviously 51 poles is wrong.Because the pole pairs will be 51/2=25.5 in the motor.They can not put half magnet in the motor.
The LEDs are removed from KLS-D and KLS-8080I controller.The error codes can be showed in the APP or user program.
And we will use buzzer on these two models.People can hear the sound to distinguish the error code easily.They will not check the Red LED after the controller is covered.

 

[fany]

- max output fre is grayed out (500), are u sure u have installed latest software?
normally you should be able to put 1000 there.
- motor poles range is only supported up to 32 motor poles. (2-32), but in my experience it does not affect the behaviour of the engine at all.
- did u unplug brake signal? my engine did not start before unplugging the brake signal 12v.

If the motor rated RPM*Pole pairs is below 30000,people doesn't have to change Max Fre output from 500 to 1000.

Sorry,the 12V brake switch can not be connected to 12V.Otherwise of course motor will not start up.

 

[fany]

np. great to hear. so what WAS the problem? so we can add it to the knowledge base?

The problem was me :lol: Is there a thread where we can post stupid stuff we do sometimes? 'cause I got a good one Here it is. There are four plugs coming out of the KLS7250D. I looked at one and saw male connectors in it so I think I have to install female connectors in the opposite plugs, which I did. The one plug I looked at had male connectors but the other two had female connectors. With female connectors in opposite plugs, there was not much going on :lol:

Yes,when you hook up the signal wires,the position is opposite.
And please just use the number on every wire to find the function.

 

[izeman]

i'm starting to hate it
changed MAX_FRE to 1000, did auto_config again, set max rpm to 3000, and NOTHING. the motor behaves exactly the same as before.
so there must be something else. eg. riding it in first gear it pulls very powerful up to 32km/h in no time and then settles down to 26km/h. so it needs the power to accelerate to 32km/h (this can't be explained with an overshoot cause by intertia of speeding up a lifted wheel). and it can for sure keep that speed. but something reduces rpm again. i hope fany is joining in some time soon ....

@fany: PLEASE look into this. i wrote it in my mail as well. why does the motor accelerate to a much higher rpm than it theoretically is capable of, then falls back and starts pumping, and can't hold a steady rpm?
have you tested these controllers with small geared hub motors?
could swapping the motor for one with thinner laminations solve that problem? is there anything i can "tune"? is there maybe an engineer's software to set hidden parameters so i can make this controller work?

 

[fany]

i'm starting to hate it
changed MAX_FRE to 1000, did auto_config again, set max rpm to 3000, and NOTHING. the motor behaves exactly the same as before.
so there must be something else. eg. riding it in first gear it pulls very powerful up to 32km/h in no time and then settles down to 26km/h. so it needs the power to accelerate to 32km/h (this can't be explained with an overshoot cause by intertia of speeding up a lifted wheel). and it can for sure keep that speed. but something reduces rpm again. i hope fany is joining in some time soon ....

@fany: PLEASE look into this. i wrote it in my mail as well. why does the motor accelerate to a much higher rpm than it theoretically is capable of, then falls back and starts pumping, and can't hold a steady rpm?
have you tested these controllers with small geared hub motors?
could swapping the motor for one with thinner laminations solve that problem? is there anything i can "tune"? is there maybe an engineer's software to set hidden parameters so i can make this controller work?

1,This controller was opened and modified too much.We can not be sure if the modifications will affect the performance too much.Because we don't know how is the modification on the PCB board,Mosfets etc.But I think it could affect the controller a little bit.

2,Please confirm if you get the special App from our customer.I only sent this special APP to one customer.
Then keep the max output Fre at 1000.Please adjust the max speed to 6000.

3,What is the rated RPM of the motor?And we may need to know the gear ratio and pole pairs of the motor.

4,How many RPMs is the much higher RPM than theoretically?
It is easy to read these RPM parameters in the monitor screen of App software.

By the way,did you try to do road testing?I think there is no problem with load testing?

 

[izeman]

1,This controller was opened and modified too much.We can not be sure if the modifications will affect the performance too much.Because we don't know how is the modification on the PCB board, Mosfets etc. But I think it could affect the controller a little bit.

2,Please confirm if you get the special App from our customer. I only sent this special APP to one customer.
Then keep the max output Fre at 1000.Please adjust the max speed to 6000.

3,What is the rated RPM of the motor?And we may need to know the gear ratio and pole pairs of the motor.

4,How many RPMs is the much higher RPM than theoretically?
It is easy to read these RPM parameters in the monitor screen of App software.

By the way,did you try to do road testing?I think there is no problem with load testing?

1) c'mon fany, don't pull that card. we're talking serious here, and i'm not asking for a warranty replacement (and never will). we're trying to make that controller work. there was no electric's modification and it behaves exactly the same as in the old case.

2) i have downloaded the new software. i tried 1.000. haven't tried 6.000, but will do. what will this change? i thought going higher than 1.000 was not recommended and 1.000 is only needed for high erpms?! btw: the updated software on your website allows setting higher FRE values as well. is this different?

3) i tested two MAC motors. a 10T and an 8T. the 10T is a ~42kV motor. 32 poles. i'm running 48V. so erpm is around 32k for the 10T and 40kerpm for the 8T.

4) i posted a comparison some posts ago. the 8T motor was doing 2.000rpm (real rpm measured at the stator), and confirmed by the app. with a trapezoid controller it's doing 2.200-2.300rpm. this is ok as sine-wave is lower top speed rpm.

the REAL problem is have (once auto_identify works, which does 1 out 5) is PUMPING speed. so it's not holding it's top speed. the bike accelerates to a much higher speed than what i can hold afterwards. and once it is there it's not giving constant power, but it oscillates. i gives almost no power, full power, no power, full power, at about the frequency you're reading it. so every second maybe.

you maybe could ask your engineers:

.) what happens if the stator saturates because of high power and inductance changes?
.) could it be that the MAC has some hall issues - not MY MAC, but all MACs? there is a special "BMC hall fix: yes/no" for adaptto controllers especially for the MAC. do you know what it does, and can it be included in the kelly software? the halls itself work fine.
.) is there a special software for technicians to change all parameters by hand?

edit: i looked up the adaptto readme, and it says:
"BMC halls fix - option to help overcome the problem of starting motors in BMC / MAC by shifting the timing advance when you start back. At the start slightly reduced efficiency, but start going better. "
this indicates that for MAC motors there needs to be some timing advance so maybe this can be added to the controller? maybe it helps your engineers to explain the issues i experience?!

 

[Lebowski]

.) what happens if the stator saturates because of high power and inductance changes?
.) could it be that the MAC has some hall issues - not MY MAC, but all MACs? there is a special "BMC hall fix: yes/no" for adaptto controllers especially for the MAC. do you know what it does, and can it be included in the kelly software? the halls itself work fine.?

edit: i looked up the adaptto readme, and it says:
"BMC halls fix - option to help overcome the problem of starting motors in BMC / MAC by shifting the timing advance when you start back. At the start slightly reduced efficiency, but start going better. "
this indicates that for MAC motors there needs to be some timing advance so maybe this can be added to the controller? maybe it helps your engineers to explain the issues i experience?!

My guess on the Adaptto option: based on where the halls are they (can) also see the magnetic field from the inductors. If you run a MAC the way you do (so inductor magnetic field equal to permanent magnets magnetic field) the field observed by the halls is 45 degrees advanced. So you need to retard them in the controller. Problem is that it is phase current dependent (as with no field from the inductors you only see permanent magnet field and no retarding is necessary). Or you move the halls so they only see permanent magnets, not the inductor field (this may be difficult in practice).

 

[madin88]

edit: i looked up the adaptto readme, and it says:
"BMC halls fix - option to help overcome the problem of starting motors in BMC / MAC by shifting the timing advance when you start back. At the start slightly reduced efficiency, but start going better. "
this indicates that for MAC motors there needs to be some timing advance so maybe this can be added to the controller? maybe it helps your engineers to explain the issues i experience?!

the problem the kelly controller has seems to be different to the one adaptto has.

Aside from the MAC this controller seems to work well as izeman has prooven with his direct drive motor.
If it would have now the option for field weakening (or overspeed), this kelly controller series would turn into a nice and attractive choice.

Based on where the halls are they (can) also see the magnetic field from the inductors. If you run a MAC the way you do (so inductor magnetic field equal to permanent magnets magnetic field) the field observed by the halls is 45 degrees advanced. So you need to retard them in the controller. Problem is that it is phase current dependent (as with no field from the inductors you only see permanent magnet field and no retarding
is necessary). Sensorless is anyways better...

how can a hall see magnetic field from the stator (if it is mounted properly and not reverse)? Isn't it shielded from magnetic field coming from underneath?

edit: addditionally the problem izeman has is independent from load, which means it should not be related with saturation.

 

[Lebowski]

how can a hall see magnetic field from the stator (if it is mounted properly and not reverse)? Isn't it shielded from magnetic field coming from underneath?

I'm not familiar with MAC internals, but I've seen hub motor pictures where the halls were inserted in a pocket milled in the stator teeth...

 

[izeman]

the issues i have are with the stator mounted on the bench. so almost NO load. and therefor no saturation of the stator i guess. it's only a few amps.
the halls are mounted the way i know they are mounted in all other motors as well. how could i change that to make it work better?
this is a picture of how the halls are installed in the MAC.

 

[izeman]

how can a hall see magnetic field from the stator (if it is mounted properly and not reverse)? Isn't it shielded from magnetic field coming from underneath?

I'm not familiar with MAC internals, but I've seen hub motor pictures where the halls were inserted in a pocket milled in the stator teeth...

this is how it's done with my HS3540. well at least for two halls.

 

[strichter76]

does anyone know more details about the attributes:
- Torque Speed KI "Speed Integral Ki in Torque Mode"
- Torque Speed Kp "Speed Percent Kp in Torque Mode"

 

[SplinterOz]

I have been doing some very light research on "field weakening" so I can try and understand.

This is the best summary I could find.

The field weakening for PMSM implies imposing a
negative value for the stator current on the d-axis of the
rotating frame, which has the role of weakening the air
gap flux linkage.
In the case of an inverter, the voltage output drops on
the stator’s resistance and inductive reactance, and the
remaining voltage is used to counteract BEMF. BEMF
is proportional to the motor’s speed and the voltage
constant, ΚΦ, of the motor. Considering the inverter’s
limitation of maximum output voltage, an increase in
speed (above nominal speed) can be achieved by
decreasing the voltage constant (ΚΦ), which is
proportional with the air gap flux linkage. However, a
decrease in air gap flux linkage is synonymous to the
decrease in torque. However, for certain applications,
the motor needs to run higher than the rated speeds
and therefore, the field weakening feature is useful,
which increases the speed range of motor beyond its
nominal speed rating.

If I understand this correctly, you can get more speed (beyond it's nominal speed rating for the voltage) at the expense of torque. To do this you apply a -ve current on the d-axis.

Excellent.. now I have to research the d-axis

 

[izeman]

i shamelessly copied CKs post from here: https://endless-sphere.com/forums/viewtopic.php?f=31&t=65031&p=1106087#p1106077

this is the post. do you think that this explanation also applies to the kelly controllers? would be good to know.

Would you mind taking a moment to explain what "I-P-O-C" means?

I use acronym to take the place of Instantaneous Phase Over Current

And better still, explain what Kp and Ki do?
I googled the terms, and it explained what they mean, but not what they do, and how they affect the motor.

The controller software use PI control policy(Current regulator Kp & Ki) to control 3 phase sine-wave current output response accuracy;
The bigger the Kp, the more accuracy the current output;
While, the bigger the Kp, the more vibrating the current output;
The smaller the Ki, the more accuracy the current output;
While, the smaller the Ki, the more vibrating the current output;
Good Kp & Ki setting means silence in low speed and no I-P-O-C when load suddenly changed;
So, if u can accept motor's noise and vibrating in low speed, just try bigger Current Regulator bandwidth.

More information about PI control u can check Dave's blog from Ti's website:
http://e2e.ti.com/blogs_/b/motordri...atch=PI CONTROL&tisearch=Search-EN-Everything

 

[mistercrash]

A question for Fany about my KLS7250D controller. I konw I will get an answer only next week. I have switches on both brake levers that send 12V to the rear brake light and to pin 1 for regen. When I press on the rear brake lever, the regen works without incidents. When I pressed the front brake lever, which is on the same side as the twist throttle, the scooter didn't go when I released the front brake lever and twisted the throttle. I had to sit at the intersection and reboot the controller.

If the regen is applied while there is throttle input, does that make the controller go into an error code? And this needs a reboot to erase the error? I suspect that without realizing it, I twisted the throttle ever so slightly when I squeezed the front brake lever. So if yes to the questions, I will rewire the regen to be applied on the rear brake lever only.

Thanks Fany
Ray