New Lebowski disciple!

[marcexec]

It would be easier for me to see what is going on if you post the hex (menu z option b) and remind me what version you have ?

To get the motor to start sensorless it is easiest when the motor is able to start with no mechanical load. The parameters affecting the start
are the wiggle rate and amplitude in the misc menu, some experimenting is necessary here. In a bike like yours it may be easy to
get the motor going by helping it a bit (bike on center stand, in gear, give a bit of gas and move rear wheel). Keep in mind that the motor
may want to go in reverse, then swap 2 out of 3 phase wires.

I'm back in business, drivers had to be replaced as well (one added 40mA in setup, the others prevented startup due to overcurrent).

It's version 2.30.

 save the following HEX lines in a text file, including the '*' termination character

0x0747	0x00BA	0x0BB8	0x0027	0x000D	0x0005	0x7FBC	0x00D8
0x0000	0x0036	0x8000	0x010E	0x0087	0x0200	0x0400	0x0400
0x0400	0xAAAA	0xAAAA	0xAAAA	0x03E8	0x0248	0x000C	0x02DB
0x000B	0xFFCD	0x0280	0xFFFF	0xFFFF	0xFFFF	0xFFFF	0xFFFF
0xFFFF	0x00A4	0x02C8	0xFFFF	0xFFFF	0x0439	0x0036	0x01EB
0x01EB	0x0800	0x0258	0x0064	0xFFFF	0xFFFF	0xFFFF	0xFFFF
0x0000	0x4CCD	0x000C	0x0000	0x00F0	0xFFFF	0xB333	0xFFF4
0x0000	0xFF10	0x0000	0x07AE	0x0018	0x0000	0x01E0	0xFFFF
0xF852	0xFFE8	0x0000	0xFE20	0x0003	0x0000	0x0078	0x0000
0x0000	0xFFFD	0x0000	0xFF88	0x0000	0x0000	0x003C	0x0003
0x0000	0x0000	0x0000	0xFFC4	0xFFFD	0x0000	0x0000	0x0000
0x00F0	0x000C	0x0000	0x0000	0x0000	0xFF10	0xFFF4	0x0000
0x0000	0x0000	0x36AC	0x0A00	0x061E	0x05DC	0x00F5	0x0035
0x00E4	0x02C6	0x013F	0x0007	0x02BE	0x0042	0x0010	0x0E10
0x0000	0x03E8	0x00C8	0x5027	0x03B6	0x6000	0x00D8	0x013F
0xFF00	0xFF00	0xFF00	0xFF00	0xFF00	0xFF00	0xFF00	0xFF00
0xFFFF	0xC519	0x764B	0x5482	0x41B3	0x35C3	0x2D7A	0x276B
0x22C9	0x1F1E	0x1C28	0x19B5	0x17A6	0x15E6	0x1463	0x1312
0x11EB	0x10E4	0x0FFB	0x0F28	0x0E6B	0x0DC0	0x0D23	0x0C94
0x0C10	0x0B97	0x0B27	0x0ABF	0x0A5F	0x0A05	0x09B1	0x0962
*

I'm trying sensorless at the moment, so Hall sensors aren't even connected. Motor should be fine, spins with the 12 FET Lyen.
Settings as above, 60V from LA test pack, tested the following:

Low side pulsing width 20
Wiggle range\Wiggle rate:	1	5	10	15	20	40	80Hz
10					X		X				X	
20					X	X	X	X	X	X	X	
30							X	
40							X		
60						X	X			X		

Low side pulsing width 40
Wiggle range\Wiggle rate:	1	5	10	15	20	40	80Hz
20							X			X		
40										X			

Low side pulsing width 80
Wiggle range\Wiggle rate:	1	5	10	15	20	40	80Hz
10						X				X		
20										X	
40						X				X

Setup is taking between 110 and 140mA, but the result is always slight movement, then stopping with the tone present, solidly on drive_2:
[youtube]Lssvkvrm6Ik[/youtube]

So, something broken, not enough voltage (I recall selecting components based on desired voltage) or keep messing with the settings?

 

[Lebowski]

There are some errors in the settings...

0) mode: HF tone

first off, you're in HF tone mode, you should be in 'Toneless, with recovery'. HF mode is unreliable and I have removed it starting v2.60.
Wiggle rate and amplitude don't work in HF mode (only in toneless), so unfortunately the tables you made are useless (sorry ).

a) calibrate throttle 1
b) calibrate throttle 2
c) polynomial coefficients throttle 1 (x, x^2, x^3): -0.0002, -0.0002, -0.0002
d) polynomial coefficients throttle 2 (x, x^2, x^3): -0.0002, -0.0002, -0.0002
e) use analog throttle 1: YES

this is in the throttle menu. Option e says to use throttle 1 but then option c says to multiply all throttle info with 0. Set option c to 1, 0, 0 for a linear throttle upto full phase current (which you set to 22A)

b) throttle current filter -3dB freq: 100 Hz

One of the changes in the later versions: option b of menu h is set to 10% of f_sample, so 4000 Hz in your case

The rest of the settings look good.

 

[marcexec]

There are some errors in the settings...

0) mode: HF tone

first off, you're in HF tone mode, you should be in 'Toneless, with recovery'. HF mode is unreliable and I have removed it starting v2.60.
Wiggle rate and amplitude don't work in HF mode (only in toneless), so unfortunately the tables you made are useless (sorry ).

OK:
[youtube]dO37Ql91qqM[/youtube]

Applied the settings as suggested:
[youtube]UbVjjt1jMkQ[/youtube]
:lol: 8)
Thanks so much for your help, I really appreciate what you are doing for our hobby, you are a legend!

next steps, a bit of tweaking and getting the Halls in as well (hopefully smoother startup for the cogging melon... )

 

[whereswally606]

well done marc this is pretty good to see another successful lebowski board

 

[Lebowski]

Are you sure the output stage is completely fixed ? It sounds kind of funny, like there is too much cogging going on, as if one fet is not switching on. I would remove motor and with the pwm test function look at each output on the scope, with 5, 50 and 95 % dutycycle...

 

[Animalector]

i agree.. that's what my motor sounded like with cheap crap fake fets... replaced them and got the beautiful sssiiirrr sound of FOC goodness

 

[marcexec]

i agree.. that's what my motor sounded like with cheap crap fake fets... replaced them and got the beautiful sssiiirrr sound of FOC goodness

Spot on mate, replaced one pair of FETs that was still from the ones I bought here with fresh ones and got an even nicer spin
They measured consistenly slightly lower on PWM on the DMM - don't have an oscilloscope
Unfortunately I probably bricked the μC after that with the batteries going low (have a bank of salvaged 7Ah LA, dropping out of FOC on higher RPMs), possibly browning out during me mistakenly messing around with the eRPM settings.
Now without setup mode the powersupply LEDs light up, but the board doesn't go into Drive_0.
Within, I can't connect:


Tried the reset jumper to no avail, anything beyond that I could try or will I have to order a new μC from you?

 

[Lebowski]

If it doesn't go in drive 0, something is seriously wrong with the settings or with the boards 5V. Are you able to go into the setup menu at all ? If so, go through it completely so see if all looks ok. If the 5V supply is weak (cannot deliver enough current) you can try by overriding (or place in parallel) it with 5V from a lab supply just to see what happens.

 

[marcexec]

5V measures 4.97V, can't program or even Drive_0 with 5.05V from external regulator
sent you a PM

 

[Lebowski]

5V measures 4.97V, can't program or even Drive_0 with 5.05V from external regulator
sent you a PM

In German no less, and i hardly speak German....

 

[marcexec]

5V measures 4.97V, can't program or even Drive_0 with 5.05V from external regulator
sent you a PM

In German no less, and i hardly speak German....

Well, assumptions are the mother of all frock-ups (Tom Clancy fans will recognize the quote)
It was an order mixed with praise and it's now translated

 

[marcexec]

OK, 2.60 chip has arrived, I had a small problem getting the FOC measurement to run (see main thread).
Calibrated Hall sensors as well, which got rid of a slight reluctance to spin up with instant full throttle (well that's their job after all).

FOC only and Hall respectively:
[youtube]apA652ZqGcI[/youtube][youtube]WJxQFZcfMuY[/youtube]

0x1570.0x01F9.0xFFDC.0x0738.0x4D5F.0x00FE.0xDA2A.0x0A7E
0x010C.0x10E5.0x0039.0x000F.0x0005.0x7FBC.0x0000.0x0266
0x0133.0x0400.0x0400.0x0400.0x02C9.0x034B.0x001E.0x041E
0x0008.0x4E9A.0xFFFE.0x05A0.0x1000.0x0000.0x0000.0xFFFF
0xFFFF.0xFFFF.0x00B0.0x02C0.0xFFFF.0xFFFF.0x0999.0x007A
0x01BA.0x01BA.0x058C.0x0258.0x0064.0xFFFF.0xFFFF.0xFFFF
0xFFFF.0x0000.0x999A.0x0030.0x0000.0x01E0.0xFFFF.0x6666
0xFFD0.0x0000.0xFE20.0x0000.0x07AE.0x0030.0x0000.0x01E0
0xFFFF.0xF852.0xFFD0.0x0000.0xFE20.0x0003.0x0000.0x0078
0x0000.0x0000.0xFFFD.0x0000.0xFF88.0x0000.0x0000.0x0003
0x0000.0x00C8.0xFFFD.0x0000.0xFF38.0x000C.0x0000.0x00F0
0xFFF4.0x0000.0xFF10.0x0A00.0x094D.0x0132.0x01CC.0x0093
0x004C.0x0149.0x7530.0x01CC.0x000B.0x0010.0x0E10.0x0000
0x2710.0x3797.0x03B6.0x6000.0x01EB.0x0073.0xFF00.0x0231
0xA93D.0xDC3D.0x5B3D.0x293D.0x8429.0xFF00.0x4000.0xFFFF
0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF
0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF
0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF
0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF
0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF
0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0xFFFF.0x0000
0x0000.0x0000.0x0032.0xFFFF.0xC519.0x764B.0x5482.0x41B3
0x35C3.0x2D7A.0x276B.0x22C9.0x1F1E.0x1C28.0x19B5.0x17A6
0x15E6.0x1463.0x1312.0x11EB.0x10E4.0x0FFB.0x0F28.0x0E6B
0x0DC0.0x0D23.0x0C94.0x0C10.0x0B97.0x0B27.0x0ABF.0x0A5F
0x0A05.0x09B1.0x0962.*

 

[Arlo1]

Are you running it from PSUs? I found that does not work well..... Need a battery.

 

[Lebowski]

Nice you got it going funny about needing to set the current to 3A for the FOC measurement, seems kind of low to me. Are you sure
the output stage is OK ?

One thing I noticed from the hex, did you autocomplete the hall menu ? The magnetic field angle compensation is at 99% instead of the 0% from
autocomplete. Or did you specifically set it to 99% because the halls are in pockets in the stator teeth ?

The chip also says it only spins to 48k-erpm, is this not kind of low for 90V ? Field weakening would definately get you more speed....

 

[marcexec]

Nice you got it going funny about needing to set the current to 3A for the FOC measurement, seems kind of low to me. Are you sure
the output stage is OK ?

One thing I noticed from the hex, did you autocomplete the hall menu ? The magnetic field angle compensation is at 99% instead of the 0% from
autocomplete. Or did you specifically set it to 99% because the halls are in pockets in the stator teeth ?

The chip also says it only spins to 48k-erpm, is this not kind of low for 90V ? Field weakening would definately get you more speed....

Yep, have the sensors in the stator teeth. I'll re-run when on the bike, with higher amp settings etc. and post the results.
I'm also working on some nice packaging for the "production" version which also has the option for 12 FETs.

 

[marcexec]

Some teaser on the packaging:


The waterblock is apparently used for GPUs (waterblock 122 will turn it up in a search, ~$7 shipped) and 12mm thick, so I ordered a 12x60x500 bar of aluminium and bought myself new blades for the hacksaw . 120mm will just fit 12x TO247 in the arrangement we're looking for. The box would actually have space for twice the FETs if turned by 90°.
The current sensor are on angled pieces of PE or PTFE from a cutting board, cheap and easy to work with.

I'm currently waiting for a M2.5 thread-cutter as this alloy is harder than the one from the CPU cooler that I used for tests, so no self-cutting.

 

[marcexec]

Is the FOC Motor Impedance measurement supposed to be stressful to the power stage?
I got the 2.80 version, set all the values as before and eventually went into the Motor Impedance measurement.
Tried first at 3A and 5A, then dropped from 12 to 10k RPM (and saw some spinning), all the while still getting an overflow.
Tried 6A and the power stage blew up again. FETs were barely warm.
The only difference to the dozens of others I did before were that the console wasn't crashing, so I ran them consecutively.
PWM measurements were fine at 5/50/95%.

Anything I missed?

Edit: The DC/DC is also out...
Lesson learned - put a fuse from the powerstage to the control board, not just in front of the controller. FETs are relatively simple to replace.

 

[marcexec]

Does anybody have experience troubleshooting DC/DC converters?
I have the board bench here and even with all secondary ICs off not even the 15V LED lights up.
Any idea where to start, i.e. what would be the most likely candidate?

The burned up component is R2 (lost the 2186 as well), no visible damage in the DC/DC stage.