Update files Lebowski controller IC, latest: v2.A1

bobc said:
Hi bas
We just tried to update v2.80 to v2.81 & have stopped the micro working.
On windows - termite did not have a "dump a file" option so we got tera-term working.
Followed the procedure, row of dots appeared eventually ending in a *
Happy days we thought.
But no more comms to the chip.
There was a "check box" in teraterm labelled "send binary" which we left unchecked- was this our mistake?
Should we try again with this checked on another chip?
Can we send the "bad" one back to be programmed correctly again?
Thanks
Bob
Click to expand...
I don't know the program teraterm so don't know what this option does.... The update file has to be send as is, any extra inserted characters like newlines or whatever will corrupt the chip.

You can send it back to me for reprogramming, a contribution for shipping will be appreciated.
 
Thanks Lebowski/Izeman

we didn't select binary so it sent as text which appears to have been correct. I think I will get hold of a Linux box and use the terminal program you used since then we cant blame the difference in windows vs Linux on it.
 
whereswally606 said:
Thanks Lebowski/Izeman
we didn't select binary so it sent as text which appears to have been correct. I think I will get hold of a Linux box and use the terminal program you used since then we cant blame the difference in windows vs Linux on it.
Click to expand...
np. does work with windows & linux. i use both. on my home laptop i use ubuntu and like the "serial port terminal" as it's easy to use and does everything you need. pretty easy to work with with all the keyboard shortcuts as well.
under windows i prefer "coolterm" which is free as well.
 
Izeman did you flash the pic under windows?

we had no trouble using the serial port until we tried to upload the file to it.
 
whereswally606 said:
Izeman did you flash the pic under windows?
we had no trouble using the serial port until we tried to upload the file to it.
Click to expand...
yes i did. as i said, the OS doesn't matter. just the software you use can be easier to use than another. all it needs is a possibility to upload a text string.
edit: and i must say that i up/downgraded at least a few dozen times w/o the slightest issue. so as long as you do it correctly it seems to be a quite robust update mechanism.
 
Just to be certain izeman - you have successfully used "coolterm" to upissue the software from windows.
I'll get that on the laptop & use that (I have no wish to re-invent the wheel)
Thanks
Bob
EDIT coolterm + windows confirmed 100% successful operation software upgrade (in this case 2.80 all the way to 2.92) 15th jan 17
 
Tomdb said:
Changes?
Click to expand...

changes going into 2.92:

different (better) hall transition method
improved general sensorless running
will now take fieldweakening and run full throttle till max output voltage (used to conk and had to be limited by erpm limiter, now no longer necessary)

this is the version Arlo1 did his 184 kmh with...
 
Sounds good, will built a permanent setup in so I can get some real load on the controller and then I can experiment with setting up limits.
 
Hey Lebowski,

I haven't visited this site in a long time. I'm thinking about buying one of your chips.
I remember you showed a video once where you had a GUI which showed a dot spinning in a circle at the same speed of the motor. Are you still using that? Would that be a good tool to use when setting up the controller? Do you provide that? and/or different software?
 
bearing said:
Hey Lebowski,

I haven't visited this site in a long time. I'm thinking about buying one of your chips.
I remember you showed a video once where you had a GUI which showed a dot spinning in a circle at the same speed of the motor. Are you still using that? Would that be a good tool to use when setting up the controller? Do you provide that? and/or different software?
Click to expand...

The spinning circle thing was related to experiments with an HF tone to sensorless find motor position of non-turning motor. I took it out as the processor is not strong enough for this and does not have the right PWM options for this.

The general variables such as motor phase and speed, temperature, currents and voltages can be viewed using the Labview program in this post:

https://endless-sphere.com/forums/viewtopic.php?f=30&t=36602&p=1088729&hilit=labview#p1088729

I still use this as it is my main debugging tool for controller IC development... For setting up the controller with a specific motor it is not really necessary.
 
Just added v2.93 . This version has some small changes and new features.

In the erpm menu:
Code: 
a) erpm limiter (forward) rampdown start, end: 29.99, 32.98 k-erpm
b) erpm limiter (reverse) rampdown start, end: 39.99, 41.97 k-erpm
c) regen rampup start, end: 495, 1276 erpm
d) accept direction change below: 78 erpm
e) erpm boundary between dr2 & dr23: 990 erpm
f) erpm boundary between dr23 & dr3: 2475 erpm

g) use acceleration limiter: yes
h) acceleration limiter start, end: 10.10, 12.11 kerpm/sec
The erpm limiter is now set using the start (when it will auto-reduce phase current) and end (erpm by which phase current will be 0A).
For mid-drive applications, to prevent it from breaking the clutch you can now set an acceleration limit. This again works with a start and end value indicating the ramp down of the phase current.

In the battery menu:
Code: 
a) use HVC, LVC battery current limiting: yes
b) battery voltage: 64.6 V
c) HVC cutoff: 64.9 V
d) HVC start : 62.9 V
e) LVC start : 54.9 V
f) LVC cutoff: 52.9 V

The High Voltage Control and Low Voltage Control (what does the C stand for typically ?) act on the allowed battery current. When powering the motor, the controller will start reducing the allowed battery current when the battery voltage drops below the value at 'LVC start' and ramp down to 0 by the time battery voltage drops as low as 'LVC cutoff' . Same for regen and HVC. The battery current is affected, meaning the motor will still do lots of torque (high phase curent) at low erpm, but will quickly run out of steam as erpms increase.

The value under b) must be set to the current battery voltage. The way it works is that every time this menu is refreshed the controller IC measures the battery voltage (takes 0.5 sec so not really noticable) and stores the 16 bit digital code from the ADC. Then when you select b) and enter a value it calculates the ratio with regards to the 16 bit from the ADC and stores this.

The interesting consequence of this is that when you change the battery voltage and refresh the menu (press enter without selecting anything) the value under b) will be updated with to reflect the voltage change. If I lower the voltage and press enter the new menu looks like:
Code: 
a) use HVC, LVC battery current limiting: yes
b) battery voltage: 56.5 V
c) HVC cutoff: 64.9 V
d) HVC start : 62.9 V
e) LVC start : 54.9 V
f) LVC cutoff: 52.9 V
Notice how the value under b) has changed... This only works when the battery voltage measurement pin has been connected and when option a) turns on the HVC LVC limiting...
 
Is this lebowski controller only PMDC or can it be modded for ACIM also ?
 
Harold in CR said:
Is this lebowski controller only PMDC or can it be modded for ACIM also ?
Click to expand...
I don't think so.

Induction motors work on slip so the code will need to be different.
 
So, Arlo, do you think a change of code would work for ACIM. I know they need a resolver for rotor position.
 
Harold in CR said:
So, Arlo, do you think a change of code would work for ACIM. I know they need a resolver for rotor position.
Click to expand...
I don't think its a simple change. I would be interested in code for a ACIM because that would run a tesla motor.

Also the starts would be very smooth. At the moment with my setup my starts are not smooth.
 
Aha LVC - excellent! Thanks Bas
FWIW tesla use copper rotor induction motors for efficiency & these are quite hard to get hold of (though siemens sell a range of industrial motors)
Al rotor motors are so inefficient they would have to be HUGE (hint - look at the size/weight of a 10kW industrial TEFC motor....)
And they are much tougher to control than PMsync or BLDC.
Way back I started making an induction motor based drive system, this was going to rely on encoder feedback from the motor shaft and control motor torque by varying slip with open loop V/f control (not FOC). It went nowhere because the micro I went for had fatally flawed PWM timer (hitachi H8)
 
I am interested in using the e-assist from GM for motorcycle conversions. They are relatively inexpensive, and, down here in CR, there are tons of 125cc-150cc's to convert. We are still paying nearly $6.00/gallon for gasoline.

I just purchased a complete 2015 leaf battery and whole drive train and harness, so, have sufficient batteries even after selling off most of the modules and the drive train to sell. The huebner DIY controller is way out of my understanding for ACIM motors. I also have a Honda Civic controller on the way, for conversion.

I really need a AC code for this chip controller build.
 
For mid-drive applications, to prevent it from breaking the clutch you can now set an acceleration limit. This again works with a start and end value indicating the ramp down of the phase current.
Click to expand...
could you please go a bit in depth how to find the correct values?
eg my motor has a top erpm of 45k. what values would i set?
 
it is not suitable for ACIM motors, I've never had the chance to experiment with one so did not take this motor into account building the controller IC.

It is suitable however for something like the Nissan Leaf motor where the motor is part BLDC and part SR (switched reluctance)... I think it would be
doable without too much trouble to make a full sensorless SR version of the controller IC, but again I have no motor to experiment with....

Don't think many people have a SR motor anyways...
 
izeman said:
For mid-drive applications, to prevent it from breaking the clutch you can now set an acceleration limit. This again works with a start and end value indicating the ramp down of the phase current.
Click to expand...
could you please go a bit in depth how to find the correct values?
eg my motor has a top erpm of 45k. what values would i set?
Click to expand...
The acceleration limiter limits the speed with which the motor can speed up. I think you just have to think of how many seconds you want for the motor to go to full speed.
Lets say you want it to go to 45k-erpm in 3 seconds, so acceleration then is 45 / 3 = 15k-erpm / second . The limiter again works by ramping down the motor phase current,
so you have to specify kerpm/s for the start and end of the ramp. So you can set start to 14k-erpm/s and end to 16k-erpm/s for the above example.

So when clutch is not engaged and you give lots of throttle the controller will auto-throttle down and speed up the motor by around 15kerpm/s . Once the clutch engages
though acceleration drops, the controller Ic will very quickly throttle back up to the throttle given by the rider.

In Arlo1's case you can try to limit wheelspin by dialing in a 0-100 kmh time. If 100kmh is 30k-erpm and you want to set 0-100 time to 5 seconds, you set the limiter around 6k-erpm/s
(so you use 5.5 and 6.5 for ramp start and end)
 
You must log in or register to reply here.
Back
Top