Full Torque at Zero RPM - electric vehicle

[Rjay]

Gentlemen,

Newbie here...but an EE too so I can speak bit, byte and flux. I have been researching various manufacturers of AC drives. I am looking for a drive that will yield full torque at zero RPM and so far, AB has a paper outlining their Force technology for this. They walk you through V/Hz. sensorless, sensor, flux control and then, finally, field oreiented flux control. The claim is that this field oriented control with encoder feedback will do the trick.

Do other drives do this? AB commands a premium for their stuff so I am interested in alternatives. Will a "boost" work as a substitute (I see this on some of the drives). It seems that I must run vector control with feedback to be able to handle fast load transients but I can't tell which drives will provide full torque at zero RPM.

Any thoughts from your experience/knowledge?

Thanks,

Rjay

 

[504bloke]

SSD Drives 690+ and 890 on closed loop with encoder will both give full torque at 0 rpm.

 

[Rjay]

Hi Dave,

Thank you for the response. I am aware of the SSD drives but I can't find this spec. They mention Flux Vector Control with a feedback encoder but AB shows a curve (5th up from the bottom) outlining the difference between Flux Vector and their patented Field Oriented control.

http://literature.rockwellautomation.com/idc/groups/literature/documents/wp/drives-wp002_-en-p.pdf

They show that Flux Vector Control has a sharp drop in Torque in the first few RPM. Their Force technology using Field Oriented Control shows full torque at zero RPM.

Do you know how to find this spec on any given drive?

Thanks,

-Rjay

 

[BTalbot]

Check out Emerson/Control Techniques UnidriveSP:
http://www.emersonct.com/products/ac_drives/unidrive_sp_-_high_performance/overview.aspx

"Closed Loop Vector – Ultimate precision control of induction motors offering full motor torque at zero speed."

 

[LadderLogic]

Does this have to be an induction motor? Any servomotor with a servo drive would produce full torque at zero RPM...

What is the approximate horsepower requirement for this app?

 

[Rjay]

The application..

Guys,

It is a 40 HP continuous, 90 HP peak, 3 phase AC motor. It is a Siemens 1PV5133 and should run about 240 RMS on the AC side. I want to put it in a car. AC drivers are not available for this motor. Actually, there are not too many AC drive solutions for this market. Whatever is out there can cost from $5K and up. A guy named Eric Tischer just did this and it was featured in Wired magazine. He bought an SSD drive and replace the IGBTs on the output side. On the input side, he bypassed the rectifiers and went straight DC into the drive (240 RMS*1.41 - 340VDC from batteries in series). Anyway, I like the concept and want to do the same thing. On a general note, anything you guys can throw out as potential obstacles would be helpful.

I am trying to pick which drive to purchase. AB drives are expensive but well documented and have this "Force" technology that is supposed to have zero torque at low RPM. I just can't see that in the specs for the other drives. I see that folks on this site pretty much give a thumbs up to the SSD/Eurotherm/Parker drives too. I am not opposed to them, I just want to make sure that I can get full torque at zero RPM. Sorry for not mentioning this all sooner. I know that the more information provided, helps you guys. I just didn't want to put anyone to sleep!

-Rjay

 

[Rjay]

Btalbot,

Thanks...I'll read up on Emerson stuff this afternoon.

-Rjay

 

[leitmotif]

Guys, It is a 40 HP continuous, 90 HP peak, 3 phase AC motor. It is a Siemens 1PV5133 and should run about 240 RMS on the AC side. I want to put it in a car.
REPLY I intend to convert 56 Chev pickup with 50 HP Leeson 3550 RPM.

You may want to search the forum for the posts i have put here over last several years on this subject. Got very good responses and encouragement and education.

AC drivers are not available for this motor. Actually, there are not too many AC drive solutions for this market.
REPLY The AC drive providers in the EV business say you can only use their drive with their motor. this may be true if you go with their plug n play package where they have the entire setup all prewired. Azure does this and to some extent Siemens. Generically any VFD will drive any squirrel cage 3 phase motor. See below if using off shelf industrial VFD.

Whatever is out there can cost from $5K and up.
REPLY Oh yes


A guy named Eric Tischer just did this and it was featured in Wired magazine. He bought an SSD drive and replace the IGBTs on the output side. On the input side, he bypassed the rectifiers and went straight DC into the drive (240 RMS*1.41 - 340VDC from batteries in series). Anyway, I like the concept and want to do the same thing.
REPLY I looked over his website quite closely. This guy is one sharp cookie have no doubt. He is in the automation business so he should know something about drives. In essance what he did is take the output module ie IGBTs sized for his need and connect them to the brains out of a smaller drive. He reported several bugs still left. I suppose I could do the same thing and would have no problem electrically BUT he had the software and the programming expertise to change the software - that I do NOT have.

On a general note, anything you guys can throw out as potential obstacles would be helpful.

I am trying to pick which drive to purchase. AB drives are expensive but well documented and have this "Force" technology that is supposed to have zero torque at low RPM.
REPLY Please send me the link - would like to look at it.

I just can't see that in the specs for the other drives. I see that folks on this site pretty much give a thumbs up to the SSD/Eurotherm/Parker drives too. I am not opposed to them, I just want to make sure that I can get full torque at zero RPM. Sorry for not mentioning this all sooner. I know that the more information provided, helps you guys. I just didn't want to put anyone to sleep!-Rjay

I am intending to use off shelf industrial rated drive for several reasons
1. Mass production should lower cost
2. In field use for at least two years
3. More applications should ensure that someone else has worked out the bugs.
4. Local support - parts, technica l, warranty etc.
5. Industrial is what I have worked with for many years.


Off shelf industrial VFD -- The trick is to
1. get the 325 VDC to the DC bus. I intend to use 2/0 copper for cabling but VFDs seem to rate their terminals for 1/0
2. You need to keep the capacitors to allow VFd to stay energized to let EEPROM do what ever before deenergize
3. connecting battery direct to DC bus bypasses any precharge circuitry. This is needed to prevent capacitors from blowing as result of uncontrolled inrush current I am told.

In theory you could use a planar "volt/freq" drive. They will give full torque at low RPM. The problem is they do not correct for slip. So assume 3600 RPM motor with 50 RPM slip
and 3600 motor RPM = 60 mile per hour = 60 Hz.
At low speed ie 1 mph with planar the command is 1 Hz and motor should do 60 rpm but subtract the 50 rpm slip now you have only 10 rpm maybe. Vector takes care of this so at 1 mph or 1 Hz command you will get 60 rpm out of motor.


I am deliberating between Hitachi SJ 300 or AB vector drives
AB designed for purpose NO ADAPTATION ie connect direct to DC 325 bus price tag $6,000
Hitachi not designed to connect direct WITH ADAPTATION to DC bus - will have to build precharge circuit and bypass relay and some other tinkering around. Terminals are rated for 1/0 - I want to use 2/0. Price tag 4400.

I think the VFD and the motor part is the easiest to implement on vehicles. The hard part is getting correct components ie relays disconnect switches fuses rated above 325 VDC and am finding will end up with stuff in 5 to 600 VDC range. This is all industrial stuff so they are big and packing them into a car is a trick.

You may PM me if you want more info.
I will not hog the forum further with more of my findings
UNLESS there are five votes for me to keep blabbing onward.

Dan Bentler

 

[LadderLogic]

Whatever tricks may be used in order to simulate 'full torque at zero RPM' in a squirrel-cage motor, nothing would beat a motor with strong permanent magnets inside.

I just checked the line-up of Yaskawa servo drives and found that Sigma-II series has motors rated up to 50 kW (i.e. about 67 hp) with a 55 kW drive designed to work on 3 X 400 VAC.

Now, it may be expensive, heavy and big solution. But it would work. Servomotors produce constant torque in almost the entire range from zero to the rated RPM.

 

[Stationmaster]

Dan,

"BABBLE" onward all you want.... the readers can choose what to read and what to skip.

Stationmaster

 

[DickDV]

Pretty much any drive with flux vector control can acheive full rated torque at zero speed if an encoder is used for motor shaft speed feedback.

Pretty much no drive with sensorless vector control can acheive full rated torque at zero speed except ABB's Direct Torque Control drives. DTC is patented technology and I have used it numerous times to get full torque at zero speed without any speed feedback device.

There is one limitation however. If the shaft must hold perfectly still while developing zero speed, then an encoder is still required. But if a little "jitter" is acceptable, DTC is the technology of choice in my experience.

 

[Rjay]

I hear what you guys are saying but here is what AB says in their paper (the link is in my second post).

PLOT 1 (Below, credit to AB).. Is Flux Vector Control with or without the sensor. Note the low current at zero RPM.

PLOT 2 (Below, same paper).. Is AB's patented Force technology using Flux Oriented Control. They say that it doesn't really matter whether you have a sensor or not. Note the full torque at zero RPM.

So here is my question. Do all the drives using FVC do what is outlined in plot one? How can I tell? This doesn't seem to be a manufacturers specification.

To be sure, adding a feedback encoder would definately improve the response time to transient loads i.e. a good idea. The motor I have, has an encoder.

To the fella with the Chevy....my approach is not purchase a large drive. Manufacturers make their money i.e. the top line retail price as a function of the parts X 3 to 6 or more depending on volume. Choosing 5, just to illustrate the point...if a drive costs, say $3000, the are building it for $600 in parts. My guess is that well over 2/3 the cost in those IGBTs...say $400. So, if you buy the "smarts" (microprocessor...algorithms, etc.), replace the output IGBTs with your own, you can probably build a 40 HP drive for under $1000. Even though the IGBTs will cost you more at retail, it will not be as large a multiplier.

-Rjay

 

[Rjay]

Btalbot,

Thank you!

[FONT=Arial,Bold][FONT=Arial,Bold]

2.3.3 Closed loop vector mode​

[/FONT][/FONT]

For use with induction motors with a feedback device installed.
The drive directly controls the speed of the motor using the feedback
device to ensure the rotor speed is exactly as demanded. Motor flux is
accurately controlled at all times to provide full torque all the way down​

to zero speed.

Right out of their manual!

-Rjay

 

[504bloke]

Thats an old article and i'm sure the new 890's have better technology than that, although i cant find info on it!

Ill get some info from SSD next week when i'm at their factory (UK).

I do know that they are actually using 4 drives in a battery car here in the uk 890's i think on common bus dc and im sure the 4 motors are special permanent magnet motors.

Again i'll find out some more info.

 

[504bloke]

I just spoke with an ssd guy and the 690+ and 890 both use field orientation as the vector mode so do produce full torque at 0 rpm when used with an encoder.

You can get both of these drives as common bus supplies i.e. you feed them DC onto the bus rather than having 3 phases.

VW Passat here using brains from an ssd 690+ invertor, running in sv mode with feedback from a hall effect sensor modified to give a 5v quadrature encoder signal out.