Lebowski's motor controller IC, schematic & setup, new v2.A1

[Lebowski]

Well, having spent part of yesterday and this morning reading the threads by animalector and zombiess on their versions of this controller, I can now say that I would like to have a compact arrangement made-up of the CPU board stacked up on the power board. My ideal power board would be similar to zombiess' board, but I think that just one sextet of mosfets would be enough for the power I need. I would like that the current sensors be integrated on the power board to minimize the routing of large conductors.
So, zombiess' arrangement is quite ideal, and I also like its driver boards plugged into the main power board. I don't think that the booster is a necessity however.
I would use IRFP4368 or 4468 mosfets to have the smallest possible dissipation, hopefully eliminating the need for a heatsink.

Now, having well understood the complications inherent in the design of a good high-power board, I have the choice of either, waiting for zombiess to finalize the design of his boards in 6, 12, and 18 FET versions, or try to design my own, based on Lebowski ultra-simple and effective "pigtail" approach
My main problem is the lack of time, running my company all day long, and having my airplane in restoration/modification un-finished, and my new bike drivetrain project taking still more time !
Arrrrgh ! Oh and I forgot; Spending countless hours on this forum, too. :lol:

If PCboards meeting my needs were available, that would simplify my life. And I'm sure, hundreds of other's, too.

Maybe you can contact Animalector for his boards ?

 

[Altair]

[/quote] A new chip, v 3.00 , will be a 64 pins TQFP... double the processor speed... intended for 1 kA at 120V.[/quote]


<<Face palm !>>
That's it, he's leaving us and is going with the car guys...

 

[Lebowski]

Ah, well, you know, after getting fired from my day job (departement is closing) it'll be a fun project to keep me busy

 

[Altair]

Oh! That's why you suddenly have more time to devote to your hobby?
You shouldn't have any difficulty finding another job, heh? Or maybe you don't WANT another job...
In that case, maybe you could put an end to this PCB shortage, as a first personal enterprise 8)

 

[emmgee]

ThanX Lebowski for the useful video and the new v2.6 user manual on page 1 here. The effect of field weakening seems to be much bigger than I thought!

Questions:
1) With A_fieldweakening at 70% of "b)" "b)max motor phase current" the rpm are doubled. Assuming that the iron is not saturated: Is it possible to calculate/estimate the efficiency at 36V at full rpm with 70% field weakening current compared to 72V at the same rpm without field weakening?
I think that the Amps at 72V are fully used for torque, at 36V partially for torque and partially to counteract the magnetic field. Mechanical power demand is the same in both cases. But what would be the electric power for the same mechanical power? In the video: ~36V 12A (400W) vs. ~36V 6A_torque 30A_fieldweak (400W with half the torque but doubled rpm). So for the same mech. power and voltage we would need 12A without and 36A with field weakening. Where is my thinking error?
2) Besides much better user-friendliness: Is there a technological difference between v2.6 and v2.5? Can I use exactly the same parameters as in the v2.6 video for my v2.5 or do you have a special recommendation?

By the way, maybe also for others: Finally Mouser was able to ship the MTI 85W100GC Three phase full Bridge (100V, 120A, 3.2mOhm) which nourishes my hope to make advances!
Cheers, emmgee

 

[izeman]

I would be interested in knowing about efficiency cons with field weakening activated. Getting more speed from the same voltage must have some limits. Especially my MAC with thick 0.5mm lams may have some major disadvantages at higher rpm than normal?!

 

[Animalector]

congrats running the MAC... awesome to see.

Cool video too.. just watched it. excellent information thanks

Andy

 

[whereswally606]

sorry to here about the job Lebowski, hope you get something lined up (if you need it). Next month I should have funds for my build. Planning to get some boards sent off from china. Anyone need any? PM me if interested so I can build a list from interested parties. I'm following BobC's lead from the cheaper supplier he used on his new bms.

current break down of cost

pcbs about £70 for 10 (roughly)

parts for brain and psu £80 from digi key

chip 2.6ver is chf 20 from lebowski

power stage fets = £30 ish for Toshiba 120v ones to220

so,
all in about £200, which would be good if I can get some back from selling off the extra spare boards.

 

[Lebowski]

ThanX Lebowski for the useful video and the new v2.6 user manual on page 1 here. The effect of field weakening seems to be much bigger than I thought!

Questions:
1) With A_fieldweakening at 70% of "b)" "b)max motor phase current" the rpm are doubled. Assuming that the iron is not saturated: Is it possible to calculate/estimate the efficiency at 36V at full rpm with 70% field weakening current compared to 72V at the same rpm without field weakening?
I think that the Amps at 72V are fully used for torque, at 36V partially for torque and partially to counteract the magnetic field. Mechanical power demand is the same in both cases. But what would be the electric power for the same mechanical power? In the video: ~36V 12A (400W) vs. ~36V 6A_torque 30A_fieldweak (400W with half the torque but doubled rpm). So for the same mech. power and voltage we would need 12A without and 36A with field weakening. Where is my thinking error?
2) Besides much better user-friendliness: Is there a technological difference between v2.6 and v2.5? Can I use exactly the same parameters as in the v2.6 video for my v2.5 or do you have a special recommendation?

By the way, maybe also for others: Finally Mouser was able to ship the MTI 85W100GC Three phase full Bridge (100V, 120A, 3.2mOhm) which nourishes my hope to make advances!
Cheers, emmgee

I think the main difference with field weakening is that the total dissipation has gone from

P_loss = R*I_torque^2

to

P_loss = R*I_torque^2 + R*I_fieldweak^2

The power delivered to the mechanical load is V_backemf * I_torque, the fieldweakening current does not contribute to mechanical power.

But, fieldweakening is a means to get more torque at high speed. Without FW the motor will spin to 20k-erpm but any mechanical load will slow it down. With FW it can
still deliver a substantial amount of mechanical power at 20 k-erpm, but at the expense of extra dissipation

The differences going from 2.5x to 2.6x:

The MAC for instance, in sensorless, startup at full throttle it would toggle between drive 2 and drive 3 with the motor starting
spinning very fast but then conking out. This is now fixed, now it starts at instantaneous full (50A) throttle.

Hall startup had a gimmick in it where it would look for sensorless info while starting with halls. When it saw sensorless info
it would combine this with the hall info for a smoother startup (and for being able to push start when halls broke on the street).
I locked the MAC rotor and gave it full throttle (50A), it would sometimes be OK, sometimes it would shake violently. This was
due to the PWM deadtime causing false sensorless info. It could be fixed with different loop coefficients but this would really
slow down hall startup. So I went back to the hall startup from v1, which was much smoother.... and will take full phase current
with a locked rotor without any problems.

 

[Lebowski]

sorry to here about the job Lebowski, hope you get something lined up (if you need it). Next month I should have funds for my build. Planning to get some boards sent off from china. Anyone need any? PM me if interested so I can build a list from interested parties. I'm following BobC's lead from the cheaper supplier he used on his new bms.

current break down of cost

pcbs about £70 for 10 (roughly)

parts for brain and psu £80 from digi key

chip 2.6ver is chf 20 from lebowski

power stage fets = £30 ish for Toshiba 120v ones to220

so,
all in about £200, which would be good if I can get some back from selling off the extra spare boards.

chips are 25 chf..... (which is £18) plus some shipping costs (5 chf typical for upto how many chips will fit in 1 envelope).
The FETs look pricey, aren't they normally more in the order of £20 for 6 ?

Can you provide a link where you order the boards ?

 

[emmgee]

Your answer inspired me lebowski and this attempt based on the MAC numbers in the video might look interesting:

U_batt ____________ 38 _____ V _____ Battery voltage
U_backemf ________ 38 _____ V _____ same as U_batt w/o fieldweakening, right?
R ________________ 0.16 ____ Ohm __ double the measured R in wye-config, right?
I_torque ___________ 50 _____ A ____ as in the settings, b) b)
I_fieldweak ________ 35 _____ A ____ as in the settings, b) n)

P_mech _________ 1900 _____ W ____ P_mech = U_backemf * I_torque
P_loss_w/o _______ 400 _____ W ____ P_loss = R*I_torque^2
P_loss_fieldweak __ 596 _____ W ____ P_loss = R*I_torque^2 + R*I_fieldweak^2

eta_loss_w/o _________ 82.61% _____ P_mech / (P_mech + P_loss_w/o)
eta_loss_fieldweak ____ 76.12% _____ P_mech / (P_mech + P_loss_fieldweak)

So: this would mean that the rpm-range can be doubled at the expense of 76% instead of 83% efficiency (200 additional watts to get rid off). Not bad! At least if my calculation is right...

 

[Altair]

Can you provide a link where you order the boards ?

Check this place:
http://www.pcbway.com/

Thinking about making PCBs ?

 

[whereswally606]

chips are 25 chf..... (which is £18) plus some shipping costs (5 chf typical for upto how many chips will fit in 1 envelope).

sorry dudeski, totally misremembered what you told me in the PM

The FETs look pricey, aren't they normally more in the order of £20 for 6 ?

Is that sarcasm? they are about £2.10 each for 225w which is a lot cheaper that I can find for any others. I don't mind spending £30 on the prototype powerstage to get a 6.5kw 24fet one. I can do 247 fets when I get paid next (or the time after next)

Can you provide a link where you order the boards ?

http://www.pcbway.com/

Used PCBway for the circuit boards - $58 for 5 prototypes, arrived after about 10 days (including shipment from China) They look OK, will build up number 1 tomorrow

Hope this helps sorry again for misquoting the price (clearly not my intention)

 

[Arlo1]

Can you provide a link where you order the boards ?

Check this place:
http://www.pcbway.com/

Thinking about making PCBs ?

You guys use these guys? They have great pricing I was wondering the quality and the turn around time?

 

[Altair]

I've never tried them yet personally, I just recently found them out while reading old threads here.
They seem to do top notch work, however, and the prices are incredible.
If I had know about them earlier, it would have saved me thousands of dollars already in my business.

 

[izeman]

for smaller quantities, and depending on size i really can recommend oshpark. VERY nice products, and ww free shipping.

 

[Arlo1]

I use OSH park its top notch and very cheep and fast. Gold plated and all kinds of stuff. But always looking at other options for cost reduction. Usualy the boards are not the big cost.
The components are. Game changer for instance has $1248 of MOSFETs alone and when you add all the other components its close to $2000.00 parts on a set of boards worth ~$400.00

 

[izeman]

HI power controller design. Game Changer

fyi: your GAME CHANGER link goes nowhere. you may want to correct that.

 

[Arlo1]

HI power controller design. Game Changer

fyi: your GAME CHANGER link goes nowhere. you may want to correct that.

fixed thanks

 

[Animalector]

I've used PCB Way.. no issues with quality or errors that I know of... quick turn around. easy to deal with.

Andy

 

[zombiess]

Now, having well understood the complications inherent in the design of a good high-power board, I have the choice of either, waiting for zombiess to finalize the design of his boards in 6, 12, and 18 FET versions, or try to design my own, based on Lebowski ultra-simple and effective "pigtail" approach

If PCboards meeting my needs were available, that would simplify my life. And I'm sure, hundreds of other's, too.

Don't wait for me. I have my own FOC controller project going because I need lots of customizations to handle the high current stuff I'm working on. I have designed a PCB for the smd based brain chip, but I never built it. Its smaller than the double decker 40 pdip version I posted on here. I can supply the PCB and schematic to you if you would like to build it.

 

[marcexec]

Anybody else encountering terminal programs freezing on the 2.60 FOC impedance measurement? (just received mine from the Dude)
I tried Termite and CoolTerm, both just hang. Un- and re-plugging the USB-to-Serial only gives dots. Voltage doesn't drop much either (65->64V) during the measurement.


I didn't have a problem with the 2.30 before I messed the settings up beyond repair.

 

[Lebowski]

Looks like somehow the controller ic resets during the measurement and goes into motor mode... is your setup switch a push button ? A toggle switch would mean after a reset it would show the main menu, not that this is good but it just shows that the chip resets and that is has nothing to do with you pc. Rs232 by the way, when you type a letter in GtkTerm it doesn't show up on the screen, it is send to the chip who then sends it back to the pc, and only then you see it on the screen. Point being, if you type but nothing shows on the screen it means the chip is not responding as it should.

Anyway, the chip has a brownout detection, if the supply drops below 4.7v or so it resets. Are you sure this does not happen ? Did you put all the caps on the pcb ?

 

[Arlo1]

When I have issues with desat or other mechanical problems I find I have this problem and that is on all versions of the chip. Its always a problem on my end.

 

[marcexec]

Looks like somehow the controller ic resets during the measurement and goes into motor mode... is your setup switch a push button ? A toggle switch would mean after a reset it would show the main menu, not that this is good but it just shows that the chip resets and that is has nothing to do with you pc. Rs232 by the way, when you type a letter in GtkTerm it doesn't show up on the screen, it is send to the chip who then sends it back to the pc, and only then you see it on the screen. Point being, if you type but nothing shows on the screen it means the chip is not responding as it should.

Anyway, the chip has a brownout detection, if the supply drops below 4.7v or so it resets. Are you sure this does not happen ? Did you put all the caps on the pcb ?

Thanks for the explanation.
Setup switch is a jumper, board (v2.30 with 2.60 chip) is fully populated. I'll see if I can find one or 2 more 12V LA to pick the best under load and/or attach the charger.