ZombieSS's power stage for Lebowski's controller video pg17

[zombiess]

If you guys make something big enough

funny you should say that.... 8)

Very funny . But it's going to take some time

 

[zombiess]

Nice video! Does it feel as smooth and silent as the video makes it look?

Yes, it is very smooth and feels completely different from what I am use to. Only major complaint I have is its missing is locked rotor recovery. I think lebowski will figure out a way eventually. If he brought the option of using hall sensors back in the latest version he could probably recover much easier. I have not run version 2.0 yet, only v 1.21

 

[Lebowski]

watch out with v2.00, w.r.t. v1.21 it goes in reverse... This has to do with the sine wave array being totally different... which was much improved in v2.10 which
does all with only one sine array (freeing up RAM for future developments). Also v2.00 can only start with the HF tone, and does not like acceleration. V2.10 is much
better with this respect as it, eh, cheats when using the HF tone, and has a completely separate and very good sensorless start (without RF).

 

[MitchJi]

I'm so stoaked to be running one of these on my roadbike. If you guys make something big enough, I will swap the Sevcon off deathbike and replace it with the beautiful sensorless future of motor control.

Hi Luke,

Maybe Zero should build a big version of these controllers, and replace the Sevcons, on their whole line of motorcycles. They could use this to gain a competitive advantage, but it might be better to start manufacturing and selling controllers. Maybe you could switch jobs and become president and CEO of their controller division !

 

[7circle]

From Arlo1 Power Stage ... http://endless-sphere.com/forums/viewtopic.php?f=30&t=35387&p=960753#p960753

Arlo1, I missed it somewhere in the conversation, but why don't you want to use the TO-220 based D44VH10/ D45VH10 transistors?

Below are two scope shots I took verifying my boost stage on my TD350E gate driver can actually deliver the claimed 15A it was designed for. The purple trace (math function) is the current in amps. You can see it delivered 14.4A in 570nS into a 1.0uF capacitor acting as my MOSFET stand in. Not sure what a 1.0uF translates to in nC. While this is not a test on an actual MOSFET that has Cgd and Cgs creating a Miller effect it does give some idea of the boost stage transistors performance. At lower current values the dI/dT increased and the over all current dropped down a bit, I think into the 12-13A range at 0.5uF hard to recall. The boost stage was fed by two 10uF XR7 caps in parallel. Gate drive power supply was 15V, 1W, so max of 68mA. It wasn't very happy supplying this current level at 19khz so I just did some single pulse tests. Only reason I have these scope shots is I decided to play around on the bench and verify some things + get some quality learning time in. That hour was VERY educational. Driver IC is a TD350E.

purple line is Amps
View attachment 2 [added attach-img-url link]

View attachment 1 [added attach-img-url link]

Here is a picture of one of the gate driver setups that was used in the test. As you can probably guess, 3 parallel IRFP4568 MOSFETs don't need a boost stage to switch at 20khz, but I included them in my design anyways so I can verify the driver design real world.

[added attach-img-url link]

Further on the Current Charge Pulse... my comments
From Arlo1 power stage thread..

CAPwm is very handy ... thanks Zombies for pointing out the phase to phase issue Arlin was getting at.
I was thinking just about the Driver Power requirement.

The CAPwm helps the big HV Cap as well spreading the Switching points across the Period instead of the one edge.
As their are some currents flowing in each phase all the time, there will be more gradual changes to the dI/dt shapes.
With all three phases Center aligned the Ripple will have a "Spread Spectrum" as the intervals between edges will be varying all the time.
So putting the phase current through the FFT on the Scope and won't show a definite 40kHz.

Zombies your test of your Booster stage D45VH10/D44VH10 TO-220's is a bit hard to relate.
I wonder what you had as the Gate resistor ON and OFF, was it 20 on and 20//20 = 10 for OFF
Plus a bit extra current from the 40ohm to the driver chip.
With no pcb circuit inductance peak current for 20ohm is 15/20 = 0.75A
With circuit inductance considered it would be peak lower later.
Where you just using the 0.104ohm current sense resistor.

C = q/V so your 1uF at 15V holds ... q = C. V = 15 * 1u = 15 uC or 15 000 nC

your 3 x IRFP4568 are just 3 x 170 nC = 510 nC at 12V so around 600 nC for 0V to 15V.
and the FUJI big block is around 5000 nC for -9V to 15V
So the 1uF is massive ... and the ESR would look very different.

I did do a bit of a hunt through your thread a few days ago to see tests on the driver board but missed these picks ... are they in your thread zombies??
Then you posted the goodies ...

I think I'll post to your thread as this a bit off topic as the test confuses what Arlin needs.

Do you have a creation date on those SCOPE pics to find in your thread what discussion was about and the Rgate values??
The KiCAD files i downloaded have 20R on 20//20 off and 10 for clamp Rbe.
Then this pic from HighHopes has other values...

Around this time [Thu Jan 30, 2014 6:38 pm] you where doing some DESAT tests.
http://endless-sphere.com/forums/viewtopic.php?p=848428#p848428

No big deal Zombies...
... no need to track it all down .... unless you feel like the efforts worth some learning for yourself...
As you have said you don't need the Boost Stage for 3 FETs but the added delay would be good to know if there is any.

7c :wink:

 

[zombiess]

Those pulse tests were done in my 18 FET TO-247 build here.
http://endless-sphere.com/forums/viewtopic.php?f=30&t=58341

This thread is my very first inverter design/build attempt. HighHopes tutored me for about a year and the result is a +20kW burst 3 phase FOC (controller chip by Lebowski, a 3rd order control algorithm) inverter in a big ugly silver "enclosure".

I didn't plan ahead for the enclosure because this was my first build so I had to "wing it" when it came to putting into an enclosure for testing. The 18 FET design is my 2nd attempt and I was flying mostly solo in it's design/layout. I have since figured out many ways to improve that design including a laminated buss setup.

I'm not sure how much this inverter can do continuously since I'm unable to load it down on the test bench and supply it with 125V DC @ 150A for an hour. It's probably thermally limited even with forced air.

Those pulse scope shots were done just as an experiment to see if my driver stage could deliver up to 15A. I don't remember how much it delivered into the 3 parallel MOSFETs but I can easily check by hooking it up again. There was no MOSFET and therefor no gate resistor int he current pulse tests, just the shunt resistor. The load was a 1 uF cap.

I need to update the KiCad files on this build and also my 18 FET build as I have several small revisions that I made which are mentioned in the threads, but not on the working schematic/layout. I did change my gate resistors when I tuned this power stage on the bench. I detailed that in this thread, I couldn't get the MOSFETs to slow down enough and I didn't want to use higher than a 20ohm gate resistor so I employed some 15nF gate to source caps. The addition of this really minimized a turn off overshoot pulse I was seeing (it was NOT ringing in the gate) and allowed my switch times to slow down. I forget which gate resistors I settled on for this build (controller is on my ebike right now so I can't look). It's all in this thread which I have saved into my lab book notes.

I welcome any comments you have on this build, let me know what details you need and I'll dig them up. I've learned a lot since this build. My 3rd controller build will be happening in a wile, only real change is the power stage design which will be slightly different than the 18 FET.