Arlo's power stage Leaf controller runs and drives page 103

[HighHopes]

but billions of living beings will breath easier and skip cancer and genetic defect experiences due to your compassionate teaching efforts here

lol, a bit much? i only put out there the basics and it wasn't even the entire system just the power path and gate driver. but we have at least 2 full disclosure projects that have decent architecture and some very good early results.

luke, what do you think about a DIY battery construction tutorial? is this a worthwhile project you could lead? is it feasible to take a rack of LiPo AA batteries (say some 2000 of them) and DIY a couple battery banks? to me it seems like a necesity of the DIY EV because it seems that 144V to 196V would be a nice range to be in. but where to get the battery for this? the charger, it is in work as you have seen elsewhere on this forumn.. but the battery is missing.

 

[Arlo1]

Making progress just working on the sensor operation then I think its ready to be finished.

Things are getting better.
This is still at 1/2 power or less.


[youtube]HMC-pwsC0Bg[/youtube]



[youtube]TEqsplvkzvc[/youtube]

 

[HighHopes]

Just to keep u motivated, this is what it takes to get to 1500HP. At least as prototype.
The cap shown is not the actual but is ok for test.

 

[liveforphysics]

Just to keep u motivated, this is what it takes to get to 1500HP. At least as prototype.
The cap shown is not the actual but is ok for test.

Voltage at 1500hp?

 

[Arlo1]

Just to keep u motivated, this is what it takes to get to 1500HP. At least as prototype.
The cap shown is not the actual but is ok for test.

I have been looking at those big igbt's they look So hard to keep the inductance down....

I just noticed they are duals so each one will be 1 phase leg. You running 2 in parallel or splitting the phases and running 6 wires out of the motor as a dual phase motor?
http://www.pwrx.com/pwrx/docs/cm900du_24nf.pdf 900 amps is not that impressive.... they are only a little better then my 1mbi800u4b-120 modules
You just have 2 times the switching components. If you want to play that way I will go order 6 more and split the wires out of the leaf motor
And don't worry I am motivated. I am just sorting out the start-up in sensor mode now.
I just worked 16 hours strait to get all my money making work out of the way so I can spend tomorrow and the next day on my CRX

 

[Vanarian]

I'm not qualified yet to post any tech comment here so I'll just give great kudos to you guys.

Sh*t your builds are making me dream of getting a good old sport car, strip it bare (and I mean it) and just let electric madness run through. Cut the panels and put them on a tubular frame, get proper coilovers & brakes just in case, then let the game begin. Main downside we still face with electric is : high power (once you use it) equals low range, and high speed equals higher power needed, which is hard... I'm eager to see improvements on batteries!

 

[HighHopes]

these IGBTs are 1200V rated,1400Amps continuous with air cooling. they are duals per package and yes two inverters parallel.
also keep this in mind when you reading high amp datasheets, its hard to find modules of this size that can also switch 20khz in practical way

 

[Arlo1]

these IGBTs are 1200V rated,1400Amps continuous with air cooling. they are duals per package and yes two inverters parallel.
also keep this in mind when you reading high amp datasheets, its hard to find modules of this size that can also switch 20khz in practical way

So as luke asked what voltage is this going to run at? And are you building it?

 

[HighHopes]

ahh... those are good questions.

huh. seems i got distracted and never answered the questions. lol.

this is for 600V solution. 1200V rated IGBT, ya, 600V is about max i would go. such large modules will have a large geometry which creates a lot of voltage spikes that need margin to exist. higher the bus voltage, also, this contributes for need of margin. so big modules at 48V bus is not equivalent for same size module at 600V.. voltage level and geometry, as one or both go up the more margin you need on IGBT voltage rating. so, for this design intended for 600Vdc, with the large module, i need 1200V rating. which i have

yes i am building it, but you know me.. i move at a snails pace cause i do so many concurrent projects. i will start with smaller power level because i want to build up some more experience with it and i am focusing on the system elements. i have only one motor in the shed right now so that's what i'm going to use. i can prove out all the other things i need to do at this power and start on the advanced stuff never before achieved. moving to much higher power is not a challenge for me so it is not where my interests are. my interests are in the other things that makes an EV reliable and developing some IP around it.. the actual EV system and its power level is straight forward.

 

[Arlo1]

I have been making some progress. Can be seen on this thread. https://endless-sphere.com/forums/viewtopic.php?f=30&t=63982
Here is a couple videos at ~410v charge.
[youtube]mu8zUJZZdx8[/youtube]

[youtube]V9ow9vxk7FU[/youtube]
What I found is when the battery gets down to ~1/2 charge which is a voltage of around 410v it lets me lay rubber. But with a fill charge I can not run anything over about 375 phase amps. When the battery was at 410v the other day I turned the peak phase amps up to 550 and it was no problem once the car was moving and in sensorless to go full throttle and lay rubber. I usually have it set at 450 as 400 is where it starts to spin both tires. And I feel If I can get success at 450 it will let me know I am making progress.

So I did try a lot of differernt control loop settings. I tried 5 kHz 6kzh 8khz 10 kHz 12khz and 14khz pwm It seems 10khz might be a smidgen better then 6 kHz but the heat sink gets hot fast and its hard to say if it really helped as I have 1 temp sensor per IGBT and when they get to 79 deg C I have current reduction of 20 peak phase amps per deg C. I did get one igbt to 84 deg with 10khz pwm. I do not have the cooling system hooked up yet. I will soon.

Lebowski has an option to capture a screen shot of all the data by pushing the setup button so I will try to implement that and continue. I have the desat holding the reset button until a key cycle and when I try to run to much power it trips the desat. I am not sure if its a noise thing or the controller konking out then tripping the desat. Hmm as I type this I realize I should setup the double pulse tester again. I added a Schottky and a Zener to the desat pin to help protect it from false triggers but if the current and voltage are high enough it might not be finished switching when the desat is fully charged. So I will double check this.
I might have to add the couple parts suggested here to create a slightly longer delay in the desat to allow my big igbts switch completely.

There is a lot of things that can be causing this. But I am very lucky I have only killed 1 igbt sorting things out. I thought the desat was set to low so I got really aggressive with settings and with a really low voltage Zener in the desat circuit and 300pf of cap to slow the circuit. I have put about 30kwh through this car doing runs in front of the shop lol and laid rubber about 70 times so far!

 

[Arlo1]

Solved. It was just a setting for the foc menue that needed changing. Full throttle at 450 phase amps at a full charge is good now. More phase amps just makes more tire smoke. It starts to break loose at 375 peak phase amps and my plan is for 900+ with good cooling. But I need slicks first.

 

[Lebowski]

and you need to connect the battery measurement pin 'cause the option you used is a patch for not doing so...

if it burns rubber at 450A ... if you set the battery current high enough it will also push 450A at high motor speed... will it burn rubber then too ? That might be seriously dangerous, rolling burnout at 100 kmh.

 

[Arlo1]

if it burns rubber at 450A ... if you set the battery current high enough it will also push 450A at high motor speed... will it burn rubber then too ? That might be seriously dangerous, rolling burnout at 100 kmh.

Ummmm that's the kind of danger I love!!! Welcome to the wonderful world of ARLO and burning rubber at 100km/h is a dream come true!


(oh and I will add voltage sensing) I just need to design a circuit and get some boards ordered.

 

[Ohbse]

Since you need to get some new tires anyway, I think the time has come for a really, really big skid

Awesome work as always Arlo, so happy to see it coming together!

 

[liveforphysics]

That might be seriously dangerous, rolling burnout at 100 kmh.

A keen observation my friend, I think you've stumbled across the reason a few of us are so excited by this thread and Arlin's amazing feat of hard work combined with your own clever code and hard work to make tire smoke at freeway speeds as a reward.

 

[zombiess]

That might be seriously dangerous, rolling burnout at 100 kmh.

No, it's seriously fun

We are not normal car enthusiasts

 

[Arlo1]

Phase amp settings at 450 amps PEAK (~320rms) with battery amp settings at 275amps (just kept it down for testing and saving the battery)
Both of those numbers are 1/2 of what I am aiming for but I will not bother turning it up more till it has some better tires and some drive shaft loops as they are welded.

This has been so worth all the work. Its now ready for some rust repair and to install the chargers and drive it daily....

Thank you so much everyone for all your help and support!

:edit I do have this posted on 2 threads. But the plan from this point forward is to push the next power stage forward on this thread. Sorry mods but for those who don't know about the other thread its here as a final victory. Now to MORE POWERFULL BUILDS



[youtube]JYO59Lun2Es[/youtube]

 

[Hillhater]

Hmm ? There has to be a limit to the amount of torque you can put to the road with a front drive set up...unless you get into gummy tires and prep, etc.
For the next power level, Maybe you should consider a rear drive layout in a nice light vehicle like a Miata or a Smart...or just a '60s Vdub ! :lol:

 

[Arlo1]

Hmm ? There has to be a limit to the amount of torque you can put to the road with a front drive set up...unless you get into gummy tires and prep, etc.
For the next power level, Maybe you should consider a rear drive layout in a nice light vehicle like a Miata or a Smart...or just a '60s Vdub ! :lol:

For this car with possi and street legal slicks it will pull hard. I am guessing 800-900 peak phase amps without loosing traction which puts it in 2.8-3 second 0-60 times.

Next car I am building a chassis from ground up and AWD!

 

[Lebowski]

275 Amp at 430 V battery, that is 120 kW or 160 hp !

nice

 

[Arlo1]

Lol lots has happend since I posted here.... Dyno'd 302.3 hp at the wheels...


Anyways to keep the sharing and a good place for all to look here is a few gate drivers that came recommended.

2ED020I12F2XUMA1 >> https://www.digikey.ca/product-detail/en/infineon-technologies/2ED020I12F2XUMA1/2ED020I12F2XUMA1CT-ND/6134200
https://www.infineon.com/dgdl/Infineon-2ED020I12_F2-DS-v02_00-en.pdf?fileId=db3a304333227b5e013344c70fc64cc8
&
1ED020I12FA2XUMA2 >> https://www.digikey.ca/product-detail/en/infineon-technologies/1ED020I12FA2XUMA2/1ED020I12FA2XUMA2CT-ND/6564405
https://www.infineon.com/dgdl/Infineon-1ED020I12FA2-DS-v02_00-en.pdf?fileId=db3a304333227b5e013344d078be4cdf
&
http://www.analog.com/media/en/technical-documentation/data-sheets/ADuM4135.pdf


As well as a boost transistor.
http://www.ixysic.com/home/pdfs.nsf/www/IXD_609.pdf/$file/IXD_609.pdf