3-phase IM DTC controller build

AC, DC, amps, volts and kilowatt. It's all discussed in here
Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Sat, 19 May 2012, 18:12

Hi Guys,
I recently posted on the DIYElectricCar forum, but since trolling this forum I released it's us Aussies/Kiwis doing all the homebrew AC motor drive development anyway.


Okay, basically I need to build a DTC-DSVM controller for a project I'm working on. I pretty much have the hardware design finalised, I have the C code working and running my motor on SimPowerSystems, and have done a CFD on the watercooled heat sink I'll be getting machined.

I know there are probably a few people from the Tumanako project plus a few other builds a saw on this forum who can provide some assistance.

The things which concern me most at the moment is correct sizing of the DC capacitor bank, and implementing a phase-phase AC voltage sensor that is isolated from the LV system

For instance, my model works fine with the 5,000uF capacitor bank with hardly any DC bus ripple switching upto 400V 600A, however it seems most home builds use more than this.

Thoughts? Keep in mind this is DTC-DSVM and not FOC
Rgds,
Stiive

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Sat, 19 May 2012, 18:31

Also, wouldn't mind some laminated bus bars if someone knows where to get them in Australia.

Or a place in Melbourne to get some cheap copper flat
Rgds,
Stiive

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

3-phase IM DTC controller build

Post by zeva » Sat, 19 May 2012, 18:46

Stiive wrote: For instance, my model works fine with the 5,000uF capacitor bank with hardly any DC bus ripple switching upto 400V 600A, however it seems most home builds use more than this.

Just a quick comment on the capacitors, it's usually not the capacitance but the rated ripple current (due to the ESR) which is the limiting factor. In most designs using electrolytics, the capacitance is an order of magnitude larger than necessary. Good quality polycaps with low ESR can offer a more appropriate ratio of capacitance and ripple current, but AFAIK are significantly more expensive. At any rate, 5000uF of electrolytics sounds reasonable for a 600A controller.

FYI I have been toying with 3ph controller design lately too, specifically sinusoidal FOC of PMSM motors (using an AT32UC3B).
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Sat, 19 May 2012, 19:36

These are the caps I was looking at;

http://au.mouser.com/ProductDetail/WIMA ... GYRYQoc%3d

5 for $1k is expensive, but i guess you pay for quality. There are a heap of cheap 450V 1000uF caps on eBay for a few dollars each, but they don't provide any specs and I'm a bit skeptical of their quality. Also don't want to run them in series to get the voltage, so i can place them straight on the DC bus to keep the inductance as low as possible.

What stage are you upto in your controller build?
Rgds,
Stiive

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

3-phase IM DTC controller build

Post by zeva » Sun, 20 May 2012, 02:07

Very nice caps there, but very expensive! You get what you pay for I guess. Most controllers (AFAIK) just use cheap electrolytics, which cost perhaps $10 per 1000uF - but have far higher ESR, hence needing more of them. (Panasonic TS-ED and TS-EE series are worth a look.)

As for my own project, I have a working prototype controller on the bench running a little sensored BLDC motor, but that's about all so far. Nowhere near building one good enough for an EV, and the project has been on the backburner lately due to other distractions! But it's still on my To Do list.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Mon, 21 May 2012, 21:31

I take it those are the caps your using, have you got any data on your DC ripple? What sort of current you pushing?

Ideally I like the thought of less caps and much closer to the bus to lower the inductance, but you pay a premium for it.

What do you use your BLDC motor for? E-bike?
Rgds,
Stiive

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

3-phase IM DTC controller build

Post by Johny » Mon, 21 May 2012, 23:46

Not sure why you are using electrolytics. What not use a Film like these? http://www.digikey.com.au/scripts/DkSea ... 0v&x=0&y=0
I'm not trying to be a smart alec. I was considering ditching the Electros in my controller (Lenze 45kW soon to become 67kW - it currently uses 2 x 5600uF x 450V Electros in series - I run a 600 VDC Bus) and using a 330uF Film. I'd love to hear some pros and cons.

Edit: Added 'why'
Last edited by Johny on Mon, 21 May 2012, 13:47, edited 1 time in total.

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 01:12

Johny wrote: Not sure why you are using electrolytics.


I assume your not talking to me? But thanks for providing a link to some alternative polyprop caps.

Just looking at some of the numbers here;

The CDE 338-1402-ND is $979 for 10 @ 800V, 600uF, 68Arms, 2mOhm.
The WIMA is $1003 for 5 @ 700V, 1170uF, 60Arms, 1.3mOhm.

Both would give roughly 6000uF, however on a volume basis, I think I would have to go for the WIMA.
Last edited by Stiive on Mon, 21 May 2012, 16:13, edited 1 time in total.
Rgds,
Stiive

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 03:00

Johny wrote: I was considering ditching the Electros in my controller (Lenze 45kW soon to become 67kW - it currently uses 2 x 5600uF x 450V Electros in series - I run a 600 VDC Bus) and using a 330uF Film.


Do you think one 330uF film would be enough? What sort of ripple are you getting currently? Are you running FOC?
Rgds,
Stiive

zeva
Senior Member
Posts: 422
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

3-phase IM DTC controller build

Post by zeva » Tue, 22 May 2012, 07:15

Most controller designs use electrolytics because they're an order of magnitude cheaper than the film caps (by capacitance), but they're certainly not as good. There are two factors in play, one is dynamic voltage sag due to capacitor ESR, and the other is voltage ripple due to the actual charging/discharging of capacitance.

I forget how to calculate these things (perhaps some electrical guru can chime in) but remember doing the sums some time ago and it appeared that electros normally had about 10x more capacity than was useful compared to their rated ripple current, i.e the ESR is the dominant factor with electrolytics.

Those film capacitors might be a much more appropriate balance of capacitance and ESR - but it'd be wise to Google the formulae and do the sums before buying. At any rate I can almost guarantee 6000uF of film caps would be much more than needed. As a guesstimate I'd say about 200uF per 100A of rated controller current should suffice.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

3-phase IM DTC controller build

Post by Johny » Tue, 22 May 2012, 17:06

Stiive wrote:Do you think one 330uF film would be enough? What sort of ripple are you getting currently? Are you running FOC?
As Ian (Zeva) says, it's ESR that you have the caps for - not so much controlling low-frequency ripple. In our case the battery pack has to contend with the low frequency ripple and the caps won't help much. ESR for ESR the Film caps win out easily. That plus longer life, smaller size (and weight) make them a good choice for EV controllers IMO.
See this article.
http://www.ecicaps.com/pdf/whitepapers/ ... _Rev_4.pdf

There are a lot of things to take into account, duty cycle, inductance of load and of course current, but the conclusion of the example case is that the Film caps can be sized 1/10th to 1/15th of an Electrolytic bank.

I'm no expert on this - I'm still deciding.
I'm using a compact industrial Vector inverter made by Lenze (EVF9329). Vector is another term for FOC. I have a second one (controller) which I am upgrading the IGBTs in - changing Gate drive, and lying to the current measurement system. This is the one Im pontificating using a single Film cap instead of the two series electros.

What are the other DIY AC controller guys thinking of using (Richo, bga)?
Last edited by Johny on Tue, 22 May 2012, 07:27, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

3-phase IM DTC controller build

Post by Johny » Tue, 22 May 2012, 17:27

Link fixed

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 17:28

Yeh i was never planning to use anything other than film caps. I just want to get some data off someone who is already using them! :o

Thinking about it just now though, perhaps using some film caps, a electrolytic and snubbers in tandem could be somewhat beneficial. The high ESR electros would also act as a kind of snubber as i imagine its higher ESR is essentially the same as a resistor in series with a LESR cap. By using all 3, you'd filter out all sorts of frequency bands - i'm just not a fan of the added inductance.

Your link doesnt seem to work. But heres another good pdf on caps
http://www.epcos.com/web/generator/Web/ ... ations.pdf
Rgds,
Stiive

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 17:29

Johny wrote: Link fixed


Ah thanks
Rgds,
Stiive

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 17:43



Great article thanks! Certainly cleared some questions up for me. My model uses 1000uF of LESR caps and preforms really well, this article has given me confidence to trust it's results. Thanks
Rgds,
Stiive

User avatar
weber
Site Admin
Posts: 2624
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

3-phase IM DTC controller build

Post by weber » Tue, 22 May 2012, 17:54

Yes. Well found Johny. Stiive, for more confidence building see the datasheet for the Tritium WaveSculptor 200.
http://www.tritium.com.au/products/TRI7 ... asheet.pdf
But then, why not just buy a Wavesculptor 200, considering the thousands of hours it typically takes to debug both hardware and software in a new power switching design?
Last edited by weber on Tue, 22 May 2012, 07:55, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 18:09

weber wrote: Yes. Well found Johny. Stiive, for more confidence building see the datasheet for the Tritium WaveSculptor 200.
http://www.tritium.com.au/products/TRI7 ... asheet.pdf
But then, why not just buy a Wavesculptor 200, considering the thousands of hours it typically takes to debug both hardware and software in a new power switching design?


Thanks Weber,
I am already familiar with the Wavesculper, we use one in our electric SAE car. Unfortunately it uses FOC and I require my own DTC variant for my project.
I already have most of the hardware I need, was just troubled by the DC decoupling caps after I got rheemed by a PM on DIY forums after stating I was planning to use 1000uF of film. It's always good to consult widely :)

Obviously my project has different hardware requirements to most other peoples projects, but its good to get an idea.

For now I can get back to programming my extensive switching table... sigh Image
Rgds,
Stiive

User avatar
weber
Site Admin
Posts: 2624
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

3-phase IM DTC controller build

Post by weber » Tue, 22 May 2012, 18:33

Stiive wrote:Thanks Weber,
I am already familiar with the Wavesculper, we use one in our electric SAE car. Unfortunately it uses FOC and I require my own DTC variant for my project.
I already have most of the hardware I need, was just troubled by the DC decoupling caps after I got rheemed by a PM on DIY forums after stating I was planning to use 1000uF of film.

If you knew the 300 amp WaveSculptor only has 800 uF, what were you worried about?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Tue, 22 May 2012, 18:59

weber wrote: If you knew the 300 amp WaveSculptor only has 800 uF, what were you worried about?


The car is currently disassembled and I didn't get a chance to check the DC ripple. Plus we are only pushing 300A with the tritium, I hope to get mine over 450A - and I wont be using the same battery pack anyway. Also, will more than likely be running a different switching frequency, or variable freq in true DTC mode....

The tritium controller is only 1 example. It has more capacitance than our previous controllers. Again, it's good to consult widely :)
Rgds,
Stiive

User avatar
Richo
Senior Member
Posts: 3737
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

3-phase IM DTC controller build

Post by Richo » Tue, 22 May 2012, 20:32

I'm using just over 1000uF low ESR.
I am using a higher voltage/lower current system.
I also have 2uF of snubber on the IGBT.
Probably all overkill.
So the short answer is NO but the long answer is YES.
Help prevent road rage - get outta my way!

bga
Senior Member
Posts: 492
Joined: Mon, 01 Sep 2008, 19:27
Real Name: Bruce Armstrong
Location: Perth WA

3-phase IM DTC controller build

Post by bga » Wed, 23 May 2012, 18:23

Hi all,

Richo and I bought some caps from EACO. They make film capswith a good spec and price. The problem is that the international shipping for small quantities negates any advantage in small orders.

In 2009 my landed price for a box of 15 x 390uF/900V + a bunch of 2uF snubbers that bolt to the bus on the IGBTs was about $50 per cap.

I note that the SHD390/900 type has been deleted and they now offer physically smaller caps for the same capabilities.

These sorts of film cap tend to have an RMS current capacity of about 50 amps (75-85mm dia x 120mm long), this is RMS and the peak is about 1500 amps. Largely this is about resistive heating of the capacitor, so short term current (drag racing) can be significantly greater than the nominal 50 amps without causing damage.

The ESR of these film capacitors is so low (3mR for the above) that almost all of the ripple voltage is the result of the capacitance and this is the determining factor in sizing the capacitor.

I was looking for something that would limit the variation of current in the battery and its long leads to a 10% of the total current.
This is also dependent on the switching frequency and eneded up at 6 of the above, but that is likely overkill and 4 will suffice.

Another good feature of the film caps is that the total capacitance is relatively low so the startup precharge process willbe quicker.
It's not the end of the world, but I can see it from here.

User avatar
Richo
Senior Member
Posts: 3737
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

3-phase IM DTC controller build

Post by Richo » Wed, 23 May 2012, 20:48

Makes me wonder how much capacitance they put in a Zilla.
Ah DC that was so last century Image
So the short answer is NO but the long answer is YES.
Help prevent road rage - get outta my way!

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

3-phase IM DTC controller build

Post by Johny » Wed, 23 May 2012, 21:02

That might be Zilla's secret - Film caps.

antiscab
Senior Member
Posts: 2489
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

3-phase IM DTC controller build

Post by antiscab » Wed, 23 May 2012, 22:25

Johny wrote: That might be Zilla's secret - Film caps.


Nope - it uses electrolytics (as well as lots of little IGBT's in parrallel)

the Soliton on the other hand does use film caps and a single large IGBT


Matt
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells

Stiive
Groupie
Posts: 295
Joined: Wed, 31 Dec 2008, 08:26
Location: Melbourne

3-phase IM DTC controller build

Post by Stiive » Thu, 24 May 2012, 01:10

Thanks for the feedback on Caps guys. Still finalising my IGBTs before I decide on what capacitor size to go for.
Also want to check my simulations to get the min/max switching freqs
Rgds,
Stiive

Post Reply