Axiom: a 100kW+ motor controller

[TrifiveEV]

But at the end of the day this is a top notch product!

I hope it will be. I did look into the major component costs and yes, your hardware choices certainly make up a large portion of the price. The amount of robustness you have built into it with that hardware is what sells me; the cost of component failure down the track will eat up the purchase price in no time with lower quality equipment on high voltages.

The question was about other people's expectation of the cost, not the cost to build or sell it. My somewhat limited experience of EV enthusiasts is that cost is often the main factor in their component choices and this didn't match in my mind with the anticipated price of the equipment. If the average enthusiast cannot justify the cost then what is the chance that when I come to buy one they are no longer available due to lower than expected sales?

 

[Arlo1]

But at the end of the day this is a top notch product!

If the average enthusiast cannot justify the cost then what is the chance that when I come to buy one they are no longer available due to lower than expected sales?

We don't have an expectation for sales numbers. We will grow this organically. We will build up 10 then sell them then build up 20 then sell them then build up 50 etc. Also if we get many people who want to put deposits down we will make sure we order enough parts to cover all of those deposits and a few more.

This is not going to be our last product. It is a product to showcase our skills. Our main focus at this point is consultancy work. Axiom is something we use to advertise with.

As for component choices. We have not made choices from price because as you said at these power levels we need to make sure reliability is number 1. Cost vs performance is number 2. Or something like that.

 

[ray88]

But at the end of the day this is a top notch product!

If the average enthusiast cannot justify the cost then what is the chance that when I come to buy one they are no longer available due to lower than expected sales?

We don't have an expectation for sales numbers. We will grow this organically. We will build up 10 then sell them then build up 20 then sell them then build up 50 etc. Also if we get many people who want to put deposits down we will make sure we order enough parts to cover all of those deposits and a few more.

This is not going to be our last product. It is a product to showcase our skills. Our main focus at this point is consultancy work. Axiom is something we use to advertise with.

As for component choices. We have not made choices from price because as you said at these power levels we need to make sure reliability is number 1. Cost vs performance is number 2. Or something like that.

How do i put down a deposit for one of these controllers ? All i would really need is 200V and 400A max for and e-bike build to run and outrunner in FOC mode ! And the other big questions is , how much it would cost ?

Kind regards
Ray

 

[Arlo1]

But at the end of the day this is a top notch product!

If the average enthusiast cannot justify the cost then what is the chance that when I come to buy one they are no longer available due to lower than expected sales?

We don't have an expectation for sales numbers. We will grow this organically. We will build up 10 then sell them then build up 20 then sell them then build up 50 etc. Also if we get many people who want to put deposits down we will make sure we order enough parts to cover all of those deposits and a few more.

This is not going to be our last product. It is a product to showcase our skills. Our main focus at this point is consultancy work. Axiom is something we use to advertise with.

As for component choices. We have not made choices from price because as you said at these power levels we need to make sure reliability is number 1. Cost vs performance is number 2. Or something like that.

How do i put down a deposit for one of these controllers ? All i would really need is 200V and 400A max for and e-bike build to run and outrunner in FOC mode ! And the other big questions is , how much it would cost ?

Kind regards
Ray

We are aiming at a $6500 USD price. 200V is a fail Zone for controllers try to run below 120vdc or above 350Vdc

 

[mannberg]

Anyone that has can-bus data from a axiom run and is willing to share?

 

[hmf1235789]

​where can we find the FPGA code for this project?​

 

[HighHopes]

​where can we find the FPGA code for this project?​

FPGA code is not available, neither are the PCB gerbers

 

[HighHopes]

Anyone that has can-bus data from a axiom run and is willing to share?

what do you want to know? CANbus running in AXIOM is programmed via VESC software which is open source so technically you can put any messaging you want, or change it or just learn every nouance you want.. basically all the measured data is put on CANbus

 

[nismoman240sx]

Any updates or progress on this? ETA for public use?

 

[HighHopes]

Mostly we're still designing the enclosure and also documenting performance metrics and functionality (safety features operate within designed range when temperature raises etc). We work on AXIOM when time allows and our team has been pretty busy the last few months on other pursuits.. but its still coming!

 

[Hackey]

Hey Marcos, i recently sent that palta tech vesc to a friend in germany and he is using it with is 8kw qs mid motor. That motor is very low inductance like 10-15uh and throws abs overcurrent fault with overcurrent upto 1200amps( as seen on typing faults in terminal) is there any way we can run this motor.i am attaching detection results below.

Faults he received.

Fault : FAULT_CODE_UNBALANCED_CURRENTS
Motor : 1
Current : 302.8
Current filtered : 293.4
Voltage : 102.21
Gate drv voltage : 15.00
Duty : 0.537
RPM : 14595.6
Tacho : 26774
Cycles running : 137035
TIM duty : 5637
TIM val samp : 5250
TIM current samp : 5250
TIM top : 10500
Comm step : 0
Temperature : 42.98

Fault : FAULT_CODE_ABS_OVER_CURRENT
Motor : 1
Current : -747.4
Current filtered : -513.9
Voltage : 118.44
Gate drv voltage : 15.00
Duty : 0.020
RPM : 12079.4
Tacho : 28492
Cycles running : 27
TIM duty : 213
TIM val samp : 5250
TIM current samp : 5250
TIM top : 10500
Comm step : 0
Temperature : 44.39

Sent from my Poco F1 using Tapatalk

 

[HighHopes]

for very low induction machine like that its helpful to switch at higher frequency because of formula V=L*di/dt. V is DC bus, a fixed value. L is low inductance. dt is your switching period where a higher frequency dt gets smaller ... end result di , rate of change of amps, gets bigger and this is your peak amps which can cause a fault. its an unusually situation.. but can be solved with higher frequency.

however.. the gate driver power consumption goes up linearly with switching frequency and it only has so much capacity to give. so watch the temperature of the gate chip, gate resistor. do some math and figure out what your limit is.

your power switch will also experience substantially more losses so watch the average temperate on that too.

and finally.. as you increase switching frequency, decrease switching period, the same deadtime starts to take up a higher % of the switching period. don't let this pass 5% or you'll bring in to your system new problems

you know how this engineering goes. everything is a trade off

hope that was helpful

 

[Hackey]

Thanks for the reply but the problem is that the controller i have uses same power stage as axiom and I don't think i would be possible to bump switching frequency to 100khz with those igbt. So if we can increase dc bus capacitance would it help currently i am using 110uf 600v panasonic caps x 6 total of 660uf and each can handle 28amps/rms of ripple current.

And if doing this thing is possible with axiom would igbts be able to handle this.

https://github.com/vedderb/bldc/commit/9652231edb749e325adce46d213884491c348cca#diff-454464977122c90e801d0db8942dd37b

Sent from my Poco F1 using Tapatalk

 

[Arlo1]

Thanks for the reply but the problem is that the controller i have uses same power stage as axiom and I don't think i would be possible to bump switching frequency to 100khz with those igbt. So if we can increase dc bus capacitance would it help currently i am using 110uf 600v panasonic caps x 6 total of 660uf and each can handle 28amps/rms of ripple current.

And if doing this thing is possible with axiom would igbts be able to handle this.

https://github.com/vedderb/bldc/commit/9652231edb749e325adce46d213884491c348cca#diff-454464977122c90e801d0db8942dd37b

Sent from my Poco F1 using Tapatalk

What frequency are you switching at?

 

[HighHopes]

" So if we can increase dc bus capacitance would it help currently" this will probably make no difference, if not worse.

 

[Hackey]

Hey Arlo, i am switching at 16khz. I was thinking of running pwm at 3 times the frequency of control loop.

Sent from my Poco F1 using Tapatalk

 

[Arlo1]

Hey Arlo, i am switching at 16khz. I was thinking of running pwm at 3 times the frequency of control loop.

Sent from my Poco F1 using Tapatalk

Is it programmed to 16khz? IE set in the menu at 16khz... Because that is actually 8khz. Don't ask me why its set that way...

 

[mxlemming]

Hey Arlo, i am switching at 16khz. I was thinking of running pwm at 3 times the frequency of control loop.

Sent from my Poco F1 using Tapatalk

Is it programmed to 16khz? IE set in the menu at 16khz... Because that is actually 8khz. Don't ask me why its set that way...

Because it uses center aligned pwm and runs the control loop at V0 and V7 (all FETs off and all FETs on) which results in a switching frequency twice as high as the pwm.

 

[Arlo1]

Hey Arlo, i am switching at 16khz. I was thinking of running pwm at 3 times the frequency of control loop.

Sent from my Poco F1 using Tapatalk

Is it programmed to 16khz? IE set in the menu at 16khz... Because that is actually 8khz. Don't ask me why its set that way...

Because it uses center aligned pwm and runs the control loop at V0 and V7 (all FETs off and all FETs on) which results in a switching frequency twice as high as the pwm.

Actually when you set it to 16khz its actually only 8khz true PWM so that is why I asked. You can likely run more PWM and be ok. But yes the code or the GUI is not really displaying the right value. It should say 8khz instead of 16khz. Yes its 8khz center aligned which is like 16khz but its still only 8khz when you set it to 16khz.

 

[mxlemming]

Hey Arlo, i am switching at 16khz. I was thinking of running pwm at 3 times the frequency of control loop.

Sent from my Poco F1 using Tapatalk

Is it programmed to 16khz? IE set in the menu at 16khz... Because that is actually 8khz. Don't ask me why its set that way...

Because it uses center aligned pwm and runs the control loop at V0 and V7 (all FETs off and all FETs on) which results in a switching frequency twice as high as the pwm.

Actually when you set it to 16khz its actually only 8khz true PWM so that is why I asked. You can likely run more PWM and be ok. But yes the code or the GUI is not really displaying the right value. It should say 8khz instead of 16khz. Yes its 8khz center aligned which is like 16khz but its still only 8khz when you set it to 16khz.

It says switching frequency, not pwm frequency... They are distinct. It is also meaningful since in 8khz pwm mode it can update the phase duty cycles at 16kHz. My observation with an oscilloscope is that this setting is one of the ones that does behave itself.

I'm also having trouble with a low inductance (6uH) motor tripping my home made VESC at above 250A. Bumping up to 50khz switching helped a bit but not much. I think as well as the inductance causing ripple, there's something unfriendly going on with the control loop. I'm not sure yet.

 

[Arlo1]

It says switching frequency, not pwm frequency... They are distinct. It is also meaningful since in 8khz pwm mode it can update the phase duty cycles at 16kHz. My observation with an oscilloscope is that this setting is one of the ones that does behave itself.

Actually its not switching frequency. As each switch is pulsing at 8khz when you set it to 16khz.

But I bring this up because you can likely set it higher. IE. If the switches and your drivers can handle 20khz then you can set it to 40khz (which is 20khz PWM) and the benefit is you get what is like 40khz pwm because its center aligned. This might help with your low inductance motors.

 

[mxlemming]

It says switching frequency, not pwm frequency... They are distinct. It is also meaningful since in 8khz pwm mode it can update the phase duty cycles at 16kHz. My observation with an oscilloscope is that this setting is one of the ones that does behave itself.

Actually its not switching frequency. As each switch is pulsing at 8khz when you set it to 16khz.

But I bring this up because you can likely set it higher. IE. If the switches and your drivers can handle 20khz then you can set it to 40khz (which is 20khz PWM) and the benefit is you get what is like 40khz pwm because its center aligned. This might help with your low inductance motors.

Hypothesize that we have a lightbulb. I switch it on. You switch it off. How many switching events have there been?

What's the duty cycle on the light bulb? Does it depend on when i turned it on? Or when you turned it off? Or both?

This is fundamental to understanding SVPWM, and fundamental to being able to calculate the ripple current.

 

[Arlo1]

It says switching frequency, not pwm frequency... They are distinct. It is also meaningful since in 8khz pwm mode it can update the phase duty cycles at 16kHz. My observation with an oscilloscope is that this setting is one of the ones that does behave itself.

Actually its not switching frequency. As each switch is pulsing at 8khz when you set it to 16khz.

But I bring this up because you can likely set it higher. IE. If the switches and your drivers can handle 20khz then you can set it to 40khz (which is 20khz PWM) and the benefit is you get what is like 40khz pwm because its center aligned. This might help with your low inductance motors.

Hypothesize that we have a lightbulb. I switch it on. You switch it off. How many switching events have there been?

What's the duty cycle on the light bulb? Does it depend on when i turned it on? Or when you turned it off? Or both?

This is fundamental to understanding SVPWM, and fundamental to being able to calculate the ripple current.

That's not the right way to see it. each "switch" is what you need to look at.
Your light bulb analogy would be ok for single phase but the current has multiple phases it can travel though when talking about each switch. What you need to show is 3 positive switches that and 3 negative switches that can turn off all effecting each of the 3 phases. This is why you need to talk about PWM frequency of each switch. The center aligned is nice and it can help a bit. But in all controllers I have worked with other than VESC stuff it talks about PWM frequency and if its center aligned it is better but its not truly 2x better as you would expect. But to loop back. I bought this up because most don't know this. And when they set it they don't realize they are setting at 1/2 of what they really want to set it at.

 

[mxlemming]

That's not the right way to see it. each "switch" is what you need to look at.

This doesn't even mean anything

Your light bulb analogy would be ok for single phase but the current has multiple phases it can travel though when talking about each switch.

No, it's exactly because there are 3 phase switching at the same time that this technique is so effective.

What you need to show is 3 positive switches that and 3 negative switches that can turn off all effecting each of the 3 phases. This is why you need to talk about PWM frequency of each switch. The center aligned is nice and it can help a bit. But in all controllers I have worked with other than VESC stuff it talks about PWM frequency and if its center aligned it is better but its not truly 2x better as you would expect.

Yes it is literally 2x better. Considering 2 phase, in each pwm cycle with edge aligned you get 1 power and 2 freewheel states. With centre aligned you get 2 power and 2 freewheel. The kicker with the edge aligned is that there's a useless transition between two freewheel states which does nothing other than waste energy and heat your FETs/igbts. This reduces edge aligned to 1 power and one freewheel state.

In 3 phase this is 2 different power, 2 freewheel with edge aligned and 4 power 2 freewheel with centre aligned.

But to loop back. I bought this up because most don't know this. And when they set it they don't realize they are setting at 1/2 of what they really want to set it at.

Do they have any idea what they want to set it at other than "faster"? There's probably a handful of people on this forum and a far lower proportion outside this forum that stand a chance of analysing and calculating the effect.

 

[mxlemming]

Hey Arlo, i am switching at 16khz. I was thinking of running pwm at 3 times the frequency of control loop.

Sent from my Poco F1 using Tapatalk

Hackey,

At 16khz switching, 100V, 15uH, you should expect worst case ripple at 50% duty, where the power state is~31us which with BEMF of 50% gives you ripple of:
di=V*dt/L= 50*31u/15u=103A.

10uH would give about 150A.

While this is a brutal amount of ripple, it's not the root cause of your 1200A over current.

I have had a similar problem with trying to push 300A through a VESC with a 6uH motor and can confirm that upping the switching even to 50khz didn't cure it. There's something else afoot here, VESC doesn't seem to play nicely with low inductance motors... I would love to know the fix.