Compact Field Oriented Controller, ASI + Grin, limited run

[CKs]

When connected to ca, I see throttle coming in but the output of throttle is does not change (sits at 1.3v base).
With ca disconnected and throttle to controller, I get nothing as well. Tried two throttles.

There is three line from throttle, Red(+5V), Black(GND) and signal line(Generally output 0.8~4.5V);
Just connect RED & Black to you controller, use a multimeter to measure the output of signal line.

 

[jk1]

What frequency does this ASI controller run at ? 10 kHz to 20 kHz ?

Also has anyone done a full throttle no load test comparing this to a normal trapezoidal controller and what is the difference ?

 

[CKs]

What frequency does this ASI controller run at ? 10 kHz to 20 kHz ?

Also has anyone done a full throttle no load test comparing this to a normal trapezoidal controller and what is the difference ?

SPWM, about 10% gap.

 

[district9prawn]

What frequency does this ASI controller run at ? 10 kHz to 20 kHz ?

The phase runner pwm freq is 13khz. I haven't found a way to change this yet.

 

[jk1]

Just wondering if this uses more or less power on a full throttle no load test vs a trapezoidal controller ?

Also for the same power can this maintain the same speed with more or less power ?

 

[Sunder]

As I'm new to FOC controller, I'm impressed by silence and smoothness already in full sensored but don't know what to expect from sensored start+sensorless run mode.
cycborg wrote sensorless run is way to go to take full advantage of FOC, but why ?

I'm keen to know this too. Can anyone please help me understand why sensorless mode takes full advantage of FOC, but sensored mode doesn't?

 

[WhatcomRider]

Just wondering if this uses more or less power on a full throttle no load test vs a trapezoidal controller ?

Also for the same power can this maintain the same speed with more or less power ?

You will find your answers starting on page 3 of this thread.

 

[WhatcomRider]

As I'm new to FOC controller, I'm impressed by silence and smoothness already in full sensored but don't know what to expect from sensored start+sensorless run mode.
cycborg wrote sensorless run is way to go to take full advantage of FOC, but why ?

I'm keen to know this too. Can anyone please help me understand why sensorless mode takes full advantage of FOC, but sensored mode doesn't?

Three Hall sensors indicate the position of the motor rotor by providing a total of 6 signal transitions for each 360-degree electrical cycle, i.e. 60 degrees apart (The period of an electrical cycle is the time it takes a pair of rotor magnets to pass a fixed point on the stator.) When using Hall signals only, to get a close approximation (within a degree or two) of the rotor position for each PWM cycle interpolation is required, but if the motor speed is changing rapidly a significant error in position can be introduced. By measuring the currents of 2 or 3 phases at a certain point in each PWM cycle a more timely and accurate estimation of the rotor position can be obtained. However, at low RPMs the back-EMF of the motor is so weak that sensorless mode can be unreliable, so sensored mode is used until the motor reaches a certain speed.

From the current measurements the FOC algorithm can account for phase current lagging phase voltage due to the inductance of the windings and is able to place the peak of the magnetic field of the stator in relation to the rotor magnets in such a way that produces the most torque. This is called the torque angle and is normally 90 degrees. Current lag increases as the electrical frequency increases, so the algorithm needs to adjust the timing of the voltage output of the inverter accordingly. In order for all this to work correctly the FOC firmware needs to know the phase resistance and inductance of the motor which it can discover during an initial auto-tune. Also, as you know, an FOC controller needs to be manually tuned for the specific motor it is running which requires an understanding of how PI torque and speed controllers work.

 

[Sunder]

. Also, as you know, an FOC controller needs to be manually tuned for the specific motor it is running which requires an understanding of how PI torque and speed controllers work.

Thanks for the explanation. I'm guessing this is not for the faint of heart then.

I'm going to get it working on Sensored mode first, then save the flash, and maybe when I have a lot of free time (retirement in 30 years?) I'll try Sensored start with Sensorless running.

Or maybe whenever someone releases exact specs for the Q100H. Whichever comes first.

Edit: Does anyone think there would be a demand for a repository of flash files for the most common motors? E.g. Q100, Q100H, Mac, MXUS, Cromotor, Bafang SW??, etc? Not sure how much I could contribute, except appreciation, but would certainly make the PhaseRunner more accessible to newbies, and Justin might be able to sell more of them.

 

[cavallo pazzo]

Still working with 1.5.0, as neither 1.5.3 nor 1.5.4 accepts to boot.
Administrator or not, XPsp3 compatibility mode not, installed on C or on D, it immediately blocks and give an error.

What's the version of your controller firmware?

[strike]Searched in BacDoor but I didn't found a controller firmware number.
In add parameter I found "bootloader software revision" in 1.100[/strike]
Happy: there's a parameter "saved software revision" at 5.5190

 

[cavallo pazzo]

As I'm new to FOC controller, I'm impressed by silence and smoothness already in full sensored but don't know what to expect from sensored start+sensorless run mode.
cycborg wrote sensorless run is way to go to take full advantage of FOC, but why ?

I'm keen to know this too. Can anyone please help me understand why sensorless mode takes full advantage of FOC, but sensored mode doesn't?

Three Hall sensors indicate the position of the motor rotor by providing a total of 6 signal transitions for each 360-degree electrical cycle, i.e. 60 degrees apart (The period of an electrical cycle is the time it takes a pair of rotor magnets to pass a fixed point on the stator.) When using Hall signals only, to get a close approximation (within a degree or two) of the rotor position for each PWM cycle interpolation is required, but if the motor speed is changing rapidly a significant error in position can be introduced. By measuring the currents of 2 or 3 phases at a certain point in each PWM cycle a more timely and accurate estimation of the rotor position can be obtained. However, at low RPMs the back-EMF of the motor is so weak that sensorless mode can be unreliable, so sensored mode is used until the motor reaches a certain speed.

From the current measurements the FOC algorithm can account for phase current lagging phase voltage due to the inductance of the windings and is able to place the peak of the magnetic field of the stator in relation to the rotor magnets in such a way that produces the most torque. This is called the torque angle and is normally 90 degrees. Current lag increases as the electrical frequency increases, so the algorithm needs to adjust the timing of the voltage output of the inverter accordingly. In order for all this to work correctly the FOC firmware needs to know the phase resistance and inductance of the motor which it can discover during an initial auto-tune. Also, as you know, an FOC controller needs to be manually tuned for the specific motor it is running which requires an understanding of how PI torque and speed controllers work.

Excellent, thanks.
I immediately changed from sensored to sensored start, sensorless run and will evaluate differences. Obviously will try bit by bit every adjustments.

 

[mjayt]

Now that the controller is communicating with the computer and I am able to program the perimeters, I am getting a controller over voltage fault. I am running a 96v 65a nimn battery(12s) and it will not let me set the Rated system voltage above 89v. I am getting a voltage scale back fault as a result.
Help!

Alright so I am guessing the max nominal voltage is 89v for the bac2000. wish I would have know that before, or I would have just gone with the phase runner. It is a GOOD thing my 96v battery is actually two 48v packs run in parallel to make 96v. I removed one battery pack and am now running at 48v with success! After changing all the settings to accommodate the 48v setup, I was able to do a auto tune and get the throttle working in sensored and sensorless. I at first had the brake voltage set above the throttle min on the CA3 so when I pulled the brake lever the motor went to WOT. glad I tested on a stand before riding. That might have made me wet myself if i was riding the bike! LOL! The throttle and brake on off voltages are now set correctly on the CA and I thought on the bac2000, but the throttle does not kick in until half way or at 2.7v. testing the ca green wire with a volt meter it is out putting the correct voltages when applying throttle and brake but the controller does not turn on the motor until 2.7 volts. The bac door software shows the controller as being set to turn on the throttle at exactly the same voltage as the CA. it does brake correctly but both the brake and throttle are either on or off. there is no modulation relative to position. the motor spins the same speed whether the throttle is outputting 2.7v or 4.5v. The analog brake function is turned on. Is there a setting I missed. I am running the green ca wire to the brake 1 input as shown and explained earlier in this topic. I am not running a separate throttle as shown in this topic.

 

[broke]

It would be great if you had bms that would hookup like adaptto.

 

[Sunder]

Guys,

Trying to auto tune my Q100H here, When I type 2 into the autontune box, it seems to "chug" along at like maybe 3-4rpm, then stop. No errors.

I tried manually updating the number of poles to the 101/202 figures that other people have used with the Q100 motors, and it just then jerks and gets an "Instantaneous Phase Current Over".

Any ideas? I seem to have missed a step somewhere.

 

[Sunder]

Alright, so I've spent quite a few hours now on this thing. (5, in fact), doing as much diagnosis as I can. I've even looked in the old BAC500 manual to see what I could get to help me.

1. The halls appear to be working. There is continuity for all 3 sensors, and I can see 5v on the end of the line. Have tested the motor with an old controller, and it sees all the halls as far as I can tell. However, when I poll the "Digital Inputs" which is meant to tell me which hall sensors are on, I get nothing. They all remain at 0. Turning the wheel forward wouldn't do much on a freewheel, but I thought turning it backwards would show the halls moving. And being 12.6:1 geared, fast, for that matter. Nothing.

2. Fiddled around manually with the hall settings. I can get combinations that seem to be smooth with like 1 phase missing, and a less smooth one with 2 phases missing. Even then though, if I crank it full throttle hard enough, I still get "instantaneous phase current over".

I'm at my wits end. Half my weekend gone, and I can't ride my bike Getting a bit desperate for help. Bowlofsalad, or anyone else who has gotten it working with a Q100 motor, can you attach/post/email me the XML dump of your config? I'm willing to try anything at this point.

Thanks.

 

[Jozzer]

Are you following procedures from this manual? http://www.ebikes.ca/downloads/BAC500_Controller_Manual_Rev%201.0.pdf
Particularly page 42/43?
Do you have it set to run on sensors - not the default setting to start with sensors then go to sensorless once the motor spins up? Are you saving the settings correctly (save to flash from the menu)?

Also read through Robbie's notes on page 2 of this thread if you haven't already..

 

[Sunder]

Thanks, I had done all those steps. I have manually got it working using the manual set up steps in the BAC500 manual. It turns out that it was a mismatched hall setting.

Even with the motor running perfectly now, it still doesn't Autotune. No idea why.

However, I can't get the sensorless mode running to take full advantage of the FOC controller. Does anyone have any suggestions on what I need to set there in order to get it to work?

 

[Jozzer]

I'm not sure if you'll ever get sensorless mode to run on a geared hubmotor, though I confess I haven't tried very hard..

 

[r3volved]

Lol I got my sensorless running but have the same phase overcurrent problem on sensored. I'm assuming you adjusted the hall table manually?

 

[Sunder]

Lol I got my sensorless running but have the same phase overcurrent problem on sensored. I'm assuming you adjusted the hall table manually?

Yeah. I have to admit, I'm not the sharpest tool in the shed, but I thought with first class honours in a degree in computer science, I could nut out the logic of this one. It seriously bent my head. I'm still recovering.

Basically, I tried to find a combo where it sounded like only one combo was working, and then tried until I found one where two combos were working. I then interpolated a working combo and bam: First go. Didn't have to try all 36 combos.

I'm sure some electrical / mechanical engineer is going to come on here and laugh at me for that methodology, but as they say in IT, when all you have is a hammer, everything looks like a nail. If all you have is algorithm design, you use algorithm design...

 

[Sunder]

Oh. If you want, I can paste the sensored values on here either as xml or as a screenshot if you want. I can also show you my wiring from controller to q100h using the factory cable if that helps.

 

[r3volved]

I don't have the same motor. I'm getting startup but faults on accel. Also seems to ramp up full right away no matter how much throttle

 

[district9prawn]

Thanks, I had done all those steps. I have manually got it working using the manual set up steps in the BAC500 manual. It turns out that it was a mismatched hall setting.

Even with the motor running perfectly now, it still doesn't Autotune. No idea why.

However, I can't get the sensorless mode running to take full advantage of the FOC controller. Does anyone have any suggestions on what I need to set there in order to get it to work?

Does it not spin up at all in sensorless? Usually if the basic settings like poles/rpm are set correctly, the motor will spin up but be jerky or give over current errors during actual riding.

Lets see that xml!

 

[Sunder]

It jerks all over the shop, and triggers the current error even on no load.

I'm hoping to do a bit more tuning tonight. I set a limit of 800w on the bike, and my top speed has definitely dropped from the old 1000w trapezoidal controller I had, and sudden changes of throttle sometimes makes the motor give a "clack" noise and lose power for a few seconds. I'm hoping I can crank up the power and see if I can increase the transition of power a little more smoothly to avoid that noise.

I'll get that XML file up as soon as I can - hopefully in about 6-7 hours from now.

 

[Sunder]

XML File Attached.