Electric Vehicle VFD Application 1

[bill318]

Most EV conversions use a 9" 20HP (15KW) series wound DC motor. I would like to have regenerative braking using a 20-35HP(15-26KW)3-phase electric motor(0~3,000 RPM)run by a VFD. Seimens makes AC motor/drive combinations for EVs, but the cost is a bit high for my liking ($6-10K).

Are there any other commercial VFDs on the market that run from a DC power rail of 200V or less and have dynamic braking using an external load resistor bank?

The motor I would prefer to use would be a sealed, fan-cooled unit with dual out-put shafts.

What I would like to be able to do is: instead of using the resistive load bank, charge a bank of ultra-capacitors during high current braking instead. The on-board charger will then take the power from the capacitor bank and dump it back into the battery bank at a controlled rate.

I understand that most motors of this size use sensor feed-back, but was wondering if any have been developed for sensorless feed-back?

Any help would be appreciated, Bill

 

[abcd3286]

Bill
will reply as I go

Most EV conversions use a 9" 20HP (15KW) series wound DC motor. I would like to have regenerative braking using a 20-35HP(15-26KW)3-phase electric motor(0~3,000 RPM)run by a VFD.
REPLY So would I. I first thought of using a forklift drive (they are going AC and have been for several years) I liked the lower voltage on the units (currently 36 or 48 VDC battery)
BUT while you can get the motor and controller what stopped me is they will not share the software. May be different in a couple years.
SO I looked into the electric vehicle market - predominantly DC as you said.

Try these websites and let me know what you think. They have AC drives
://

http://www.metricmind.com/index1.htm

He is in Portland Ore -- if you can get him to respond via e mail you are a better man than I.

http://www.electroauto.com/contact.shtml

Have not talked to them yet on their equipment

http://www.cameronsoftware.com/ev/Welcome.html

This guy is in BC Canada I put him in here because he has a well thought out game plan for conversions - very little tech data but good strategics.


Seimens makes AC motor/drive combinations for EVs, but the cost is a bit high for my liking ($6-10K).
REPLY Yes the Siemens is pricy.

Are there any other commercial VFDs on the market that run from a DC power rail of 200V or less
REPLY Yes see above

and have dynamic braking using an external load resistor bank?
REPLY Why use a resistor when you can do regen braking and charge back to the battery? See below

The motor I would prefer to use would be a sealed, fan-cooled unit with dual out-put shafts.
REPLY Oughta work depending on waht you are doing.

What I would like to be able to do is: instead of using the resistive load bank, charge a bank of ultra-capacitors during high current braking instead. The on-board charger will then take the power from the capacitor bank and dump it back into the battery bank at a controlled rate.
REPLY The VFD in braking mode puts power back to the VFD DC bus (uh does it not?) THEREFORE if the battery is tied to VFD DC bus you can "dump" this power to the battery fron the VFD DC bus directly and not mess around with all this other stuff.

I understand that most motors of this size use sensor feed-back, but was wondering if any have been developed for sensorless feed-back?
REPLY NOt sure on this - doubt it.

Have thought of using industrial VFD but that forces me into high battery voltage -- over 300 VDC.

Any help would be appreciated, Bill
REPLY Well I dun the best I could at this stage.

I am in Seattle come on over and we can pencil think over lunch

Have been thinking of posting the same question -- thank you for the courage and ice breaking
Dan Bentler

 

[bill318]

Thanks for the reply Dan! Yea, you never know if you are going to get beat up or just have a meaningful conversation some times.

Try these websites and let me know what you think.
I tried metric mind as well and never heard back from Victor either.

http://www.acpropulsion.com

also has AC-150 drives in the high price range as well. I'll check out the other two you mentioned.

Why use a resistor when you can do regen braking and charge back to the battery? The VFD in braking mode puts power back to the VFD DC bus (uh does it not?)
I was looking into the possibility of using Lithium Ion batteries and their higher internal resistance can lead to poor braking energy recovery. Some of the VFDs I have seen in the past have a second resistive load bank circuit that is seperate from the supply side. Using this circuit, I may be able to dump the high braking energy into a capacitive bank.

I have some ideas on a battery management solution for Lithium Ion batteries and it requires a seperate DC charging bus where the ultra capacitors would also reside. This way I can absorb large braking currents in the capacitor bank and still feed the LiOn batteries a more controlled current.

I figured I would ask here because there are numerous people exposed to drive systems and someone may know of something that could be made to work at a more reasonable price. I'm glad someone else was thinking along the same lines!

If I can't find a reasonable AC system, then I guess I will go with the series DC motor and a Zilla controller. I'd have to scrap the regen idea, but I would still like to try the LiOn batteries to get a 65-100 mile range out of a small pick-up conversion.

I'm not in Seattle very often, but I do like discussing various ideas for sure.

Thanks again, Bill

 

[bill318]

I checked out the Solectria AC24 motor and the DMOC445 controller and this is still in the $6K range. Maybe I'm asking too much, but it would be nice to be in the $3-4K range for these two items.

As a comparison a 9" Advanced DC #FB1-4001A motor along with a Zilla Z1K 1000A controller runs about $3.6K.

The Beetle Electric Vehicle site does indeed have some good information, thanks for the link.

-Bill-

 

[MikeHalloran]

Last time I looked at ultracapacitors, they were severely voltage limited, 3..5VDC max, and had a very high internal resistance, at least several hundred ohms. I.e., you can't charge 'em far, and you sure can't charge 'em fast.



Mike Halloran
Pembroke Pines, FL, USA

 

[bill318]

Hi Mike,

The ultra capacitors I am looking at are similar to the Maxwell types

http://www.maxwell.com/ultracapacitors/index.html

. You are correct about the voltage, ~2.5V /cell with an ESR of about .0006 Ohms. While connecting them in series with a balancing network does raise the total ESR, the pack can still handle large charge and discharge currents.

-Bill-

 

[itsmoked]

Are the Lithium Ion that incapable of taking a brief fast charge? They seem to charge them pretty hard.

It would seem to me that the weight of the caps, their space consumption, their cost, the extra equipment, and the hassle needed for the cap/charger aspect might not be worth the effort verses a slightly less efficient charge and possibly even a braking resistor.

 

[MikeHalloran]

I'm having difficulty with the claimed ESR, of less than a milli-Ohm, especially given the stated materials of construction. I did not find a reference to test method. Apparently it is measured over a very short sample time, immeditately after applying a low- impedance load. The FAQ response on discharging the caps is slightly illuminative.

I'm getting the impression that the internal impedance is much more complicated than just a simple resistance. I.e., they may be great for removing ripple, or for powering a drill motor for a minute or two. For something like braking a vehicle down a hill and partially powering it up the next hill, the attainable discharge rate over time may limit the vehicle performance more than you would like.

In their Applications| Transportation pulldown, they talk about using the ultracaps to remove peaking stress from hybrid vehicle batteries, not as replacements for the batteries themselves. I guess that's what you were talking about; using the ultracaps to receive braking energy at high currents, and transfer the stored energy into the batteries at a lower current. It makes for an interesting control system problem.

The module manual under Support| Manuals reveals a little more about the technology. It is of course no accident that the available modules are rated 14V and 42V.

They also provide a paper dealing with a no-battery hybrid vehicle under Support| Whitepapers. All it takes is two modules, a motor, and three alternators in addition to the engine. Of course you'd use a truck for the test mule.

The marketing paper "Ultracapacitors and HEVs" under Support| Whitepapers has a nice table comparing the application range of various energy sources/stores. It clearly positions ultracapacitors as well suited for applications with discharge times between 0.3 and 30 seconds, full to empty, which places them between regular capacitors and batteries.

You probably knew all that before; I didn't. Thanks.







Mike Halloran
Pembroke Pines, FL, USA

 

[bill318]

From what I have read on the large LiIon batteries, they seem to have quite a large internal resistance which might make them heat up during regenerative braking. But you guys are right, if the battery bank was sized correctly, then they should be able to absorb the current surges from the braking without the need for the expensive capacitor bank.

Paralleling the ultra capacitors and LiIon batteries may be the best of both worlds, however, it would be nice to be able to discharge the ultra capacitors all the way down to make full use of their capability. When paralleled with a LiIon battery, you can only drain the battery/capacitor pair to 2.8V. This saves the battery, but does not make full use of the capacitor.

My idea of the battery management system would be to use the separate DC charging bus. Each LiIon cell would be connected to this bus via individual isolated off-line power supply/chargers. The bus would vary in voltage anywhere from say 90 to 300 VDC. Each charger would take care of its own battery and includes a microcontroller to handle the charging profile and report status to the main system.

This also makes the charging system for the vehicle very easy. Basically you would need a rectifier/ filter on your mains AC supply and place the unregulated DC on the charging bus. The individual chargers would take it from there and properly charge every battery in the series string. The opto-isolated serial communications between all the chargers would allow full monitoring of temperature, charge currents and voltages. If you tell the system you are plugged into a 15A AC supply, then the chargers would be set for a lower charge so as not to exceed the supply. Likewise, if you had a large AC supply available, the chargers could go at full rate which would be about 20A @4.2V (84 Watts). So a series string of 36 batteries could charge up to about the 3KW level. I figure each charger could be made for around $25-$30 in parts. To expand the battery bank, just add a charger along with each additional battery.

While driving, any time there is regen power on the bus, then the chargers could go to work and charge the batteries and continue to monitor each cells health.

Anyways, this is why I asked the question about a VFD with the separate braking load port. Still no luck finding a commercial unit that may work, but I'm still looking.

The Marathon Motor/Durapulse controller combinations looked interesting and seem reasonably priced. Still can't find A HV DC input controller in this line-up though.

Keep in mind these are just some ideas and they are certainly open for more input.

-Bill-

 

[GregLocock]

You can't really use supercaps just paralleled to the cells, they do need their own controller. Our solar car team tried them out for regen braking (they work). I have seen estimates of the viable pack size for normal vehicles, and it gets ugly quickly. However, if you are not being a purist then it is worth bearing in mind that most braking applications are around 0.3g or less, so you don't necessarily need to be able to stop the vehicle on a sixpence electrically.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[itsmoked]

GregLocock; Nice bit of info there. [0.3g]

What's a reasonable acceleration? If it's a lot more that 0.3g then it would seem the Li-ion batteries could absorb typical deaccleration loads.

bill318; Interesting scheme there... What kind of batteries are in the Prions? NiMH? They suck down the regen well. How about a small amount of them to receive[buffer] the regen?

 

[GregLocock]

Normal traffic? 0.2g (but it varies a lot with speed).

Perhaps a better way to look at it is power to weight ratio, 40 hp/ton is reasonable.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[LionelHutz]

I'll just point out that almost every commercial VFD uses (or has provisions for) a seperate braking resistor which is PWM controlled across the DC buss. Most VFD's are built to run off voltages greater than 3-phase 208VAC recified though so they don't meet your <200VDC requirement.

I think you'd find that the AC motors being used in these applications are usually called brushless DC motors or maybe a better description would be AC permanent magnet synchronous motors. They're not usually AC induction motors.

Your idea of the chargers and buss is interesting.

 

[bill318]

itsmoked, (great screen name by the way...) "What kind of batteries are in the Prions? NiMH?" If you are referring to the Toyota Prius Hybred, as far as I could find, the 1st gen had a 288V NiMH pack and the 2nd gen has a 500V NiMH pack. Don't quote me, but I think I read somewhere that these may have even been 'D' size cells at one point in time. At 1.2V/cell, that is a lot of batteries!

Greg, good point "you don't necessarily need to be able to stop the vehicle on a sixpence electrically." 40hp/ton would make a great sports car even!

Lionel, do you know of any commercial DC input VFDs even if the voltage is higher? I guess I would have no problem going to a higher voltage, power is power and we are already talking lethal voltages and currents anyways. Might even reduce some of the losses in the system by going to a higher voltage. Tack on a few more chargers... let's see, 70 LiIon batteries in series would be about 196V discharged, 297V charged. The total Lbs (or KW Hrs)of batteries would still be the same, it is just the management of so many that makes it fun. I have seen both PM and induction type motors in EVs. I guess I define an AC motor as one that does not provide its own commutation.

I think you guys have convinced me to skip the ultra capacitors for now and just beef up the battery bank. So for now I need to get the drive and motor rounded up, but I would still like regen as I live in a place that has hills and stop-n-go traffic in town.

Thanks to everyone for their input! -Bill-

 

[itsmoked]

bill318; That's probably the best deal.. Adding more LiIons to reduce the regen charge rate into them also has the advantage of increasing your range both thru bulk cell volume and thru the lower battery source resistance.

Where abouts are you located?

 

[LionelHutz]

Most commercial VFD's have a diode rectifier in the front end. So, apply DC and it feeds to the caps but this doesn't allow 2-way power. Fortunately, many also have DC buss terminals where you can directly connect to the DC buss so you need to find one of these. Check that you can turn off the 3-phase power monitoring in the VFD. I'd also suggest going about 1.5 times the motor rating with the VFD and finding a VFD with some sort of torque control with torque limiting to avoid nuisance tripping.

Your 297V charged is good for a typical 208V VFD but I doupt you'll reach the discharged voltage before the VFD goes off on a low line fault.

 

[cswilson]

A few notes here:

* I know the Prius does use ultracapacitors in addition to its NiMH battery pack (from IEEE Spectrum, I believe). The only reason I can think for this is quick absorption of regen power, then bleed more slowly to the battery. I don't know anything about the connection topology or control system for this.

* You may still want to consider a regen resistor as a load of last resort. Otherwise, what happens if you are going downhill with a fully charged battery -- you would have no "engine braking". (I debated this ten years ago with a colleague who lived right under the Hollywood sign. He didn't see the need for a regen resistor, but I pointed out that if he charged a car overnight, he would be riding his brake pedal all the way downhill each morning.)

* I recently talked to aerospace engineers who are using Li-Ion batteries in the unmanned aerial vehicles (UAVs) they're using so many of in Iraq. While they are using these because of their superior energy/weight and energy/volume ratios, they are very leery of them due to the safety issues involved. While they can get away with it because the military tolerates much higher risks than the civilian sector does, they definitely consider large Li-Ion banks too dangerous for consumer use. Last time I checked, USDOT had all sorts of restrictions and prohibitions on the transport of Li-Ion batteries much bigger than cell phone batteries for this reason. PLEASE BE CAREFUL!

* The other knock on the use of Li-Ion batteries for vehicle applications is that their current output capability is very small compared to NiMH and Lead-Acid batteries of the same capacity. Not a problem for consumer electronics, but it definitely could limit your acceleration capabilities if you're not careful. (But maybe pre-charging the ultra-caps could help?)

 

[itsmoked]

cswilson; Are you sure about your last point? I see discharge rates of 10C and 12C used on Li-po batteries all the time. R/C helicopters drawing 40A from the small battery packs they loft.

 

[abcd3286]

RE
Are you sure about your last point? I see discharge rates of 10C and 12C used on Li-po batteries all the time. R/C helicopters drawing 40A from the small battery packs they loft.

I'm just a former submarine electrician. The 10C and 12C values correspond to what I learned as the 10 and 12 hour rates. Our battery was 5,000 AH 250 VDC. So
1 Hr was 5,000
2 hr 2,500
4 hour 1,250 (this was one we most commonly referred to)
10 hour was 500.
The 40 amp draw you quote is somewhat meaningless by itself what was battery rating?

Electric vehicles can (at start) draw at the 14 min or maybeless rate (2 to 500 amp to get rolling). Steep hills can easily hit the one hour rate.

I am doubtful about Li batteries because
1. YOu cannot get replacements for the Ford pickup EXCEPT thru a "sanctified" Ford shop.
2. They have not been in commercial use long enough and completely de bugged for my taste.

I am going to go with the lead acid on my vehicle and accept the lower power to weight or volume penalty
but get
well proven technology
ability to buy anywhere
long term availability
fairly low cost
I understand them

Even though I am used to 250 VDC
WITH the associated 300 volt charge voltage
Remember TVG = 2.62 times cells minus .003(T - 80)
(am journey electrician) I still would like to keep voltage below 200 if I can to stay with equipment that hs dielectric rating of 250 (or is it 300?) Any thing above that gets more expensive.

Dan Bentler

 

[bill318]

itsmoked I live in Central Oregon and as a matter of fact, I fly an electric RC helicopter. I use a 9.6V/3.7Ah NiMH battery pack. I have average currents a shade over 20A into a 3-phase AC motor. And you are right, 40A is not unheard of in these craft.

Lionel I was wondering about the low voltage cut-out. The LiIon batteries do have a wide voltage range throughout their discharge cycle. I'll do some more looking around for the DC bus versions.

cswilson & Dan B. You are right, LiIon power density is a bit lower, but the energy density of LiIon is much greater than the other battery chemistries. While acceleration may be limited, the range is enhanced greatly. As for availability, Thunder Sky

http://www.thunder-sky.com/en/battery.htm

is in full production, although I would agree that the quality is not the greatest. SAFT on the other hand should have a higher quality battery but I'm sure you would have to pay for it.

http://www.saftbatteries.com/120-Te...chno=Lithium+systems&Intitule_Produit=HEliion

I know Nevada ships some advanced compounds used in Lithium battery construction overseas, but I have yet to find a US manufacturer of large Lithium batteries available to the public.

As for the dangers, anything with high power/energy density deserves respect; both petrol fuels and batteries. Yes LiIon batteries can catch fire and it is usually from being overcharged. A battery management system is a must for sure.

I guess the one advantage I see of LiIon over NiMH is that the cell voltage is 4.25V for LiIon and 1.2V for NiMH. Which may make the management system a bit easier to deal with. Not that I'm against NiMH, because I think they are a great battery.

There I go, I should be concentrating on drives and motors in this thread... Oh well, I guess it's all related!

-Bill-