AC motors, multipoles, torque

AC, DC, amps, volts and kilowatt. It's all discussed in here
Post Reply
User avatar
Richo
Senior Member
Posts: 3685
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

AC motors, multipoles, torque

Post by Richo » Tue, 06 Jan 2009, 05:18

woody wrote:How far can we push this ? Performance is not about efficiency!


The e-motor already has 30% more torque than the petrol orignal.
The eV BMW should leave the original petrol burner with the dinosaurs Image

I'm sure that the more you put in the less extra you'll get out. Image
Hard to test real values without a dyno+load.
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:

AC motors, multipoles, torque

Post by Johny » Wed, 07 Jan 2009, 17:38

acmotor wrote:BGA, I am with you on the slip rate thinking.
It's interesting that slip rate was what we were talking about months ago with that little MC3PHAC speed controller. You (acmotor) had suggested comparing real motor speed with it's internal model of motor speed and maintaining a given slip for a given RPM to achieve torque-like control.
acmotor wrote:I have thought of a micro between the pedal and controller that controlled the (speed) demand to the controller and adjusted the slip (torque) based on motor rpm, dc bus etc etc.
For the VFDs we already have.
If speed was controlled by the throttle and torque was linked to throttle with an offset so that we still had braking torque at zero throttle THEN maximum torque was based on motor RPM - wouldn't that be enough?

Do you think that the DC Bus really needs to be taken into consideration?



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

AC motors, multipoles, torque

Post by bga » Wed, 07 Jan 2009, 18:20

Hi Johnny,

I have been thinking about how throttle pedal interacts with the torque demand vs speed. I would want a pedal relationship that is similar to an ICE, which was are well trained to operate this and is seems to be effective for speed regulation vs acceleration.

Slip rate is the most important factor to control and maintain within the motor's operating limits. There may be an optimisation of field strength vs slip rate at various speeds/torques that will determine the most efficient target slip rate at various operating conditions.
(wanted: A motor that's optimised for slip rate/efficiency and ignores high slip DOL starting performance)

I agree that using the throttle to permit regen braking makes sense. ICEs do this if the revs are above Idling, although most cars are geared and weigh so that the engine braking is not very effective, not compared to acmotor's wheeel locking regen capability.

I want to do an experiment where I use a pot to select a continuous range of graded regen options:
-no regen
-mild regen on brake only
-mild regen on brake and throttle below coast.
-strong regen on throttle below coast.
I think that this is a control that may be useful to operate according to the traffic conditions.

I would like to use a pot on the brake pedal so that the slack before the hydraulic brakes operate can be used for proportional regen control. Also at hard hydraulic braking, regen has to be disabled to not upset the brake operation.

A brake lamp override on regen operation would be good to reduce rear-enders.

Speed limit obervation may be easier of the throttle system cogs slightly at the speed limits 40,50,60,70,80,90,100, causing the vehicle to prefer these speeds to the intermediates. ?good/bad?

Cheers

BGA

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Wed, 07 Jan 2009, 18:21

Available (max) torque is a function of present RPM and available voltage. It is very dependent on over sync revs and controller is not naturally thinking we want to push the emotor to near pullout.

Voltage can vary from 470V (low voltage cutout point) to 750V (fully charged after regen). This means that the motor voltage can be from 336VAC to 536VAC and this has a considerable effect on max torque.
If we weren't squeezing everything we can out of the emotor by taking it above Tn and sync revs then it would be less of an issue.

You have me thinking re the use of accelerator to feed both a speed and torque signal to the controller (if that was what you meant).

iMiEV MY12     105,131km in pure Electric and loving it !

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

AC motors, multipoles, torque

Post by Johny » Wed, 07 Jan 2009, 18:27

I understand that DC Bus voltage will effect max. torque but I was wondering if this had to be taken into account to produce the signal for maximum torque for any given RPM. I wouldn't think so.
It seems to me that we have to create a curve for the maximum torque figure for given RPM for each motor.

Yes, use accelerator for both speed and torque but torque has preset minimum value (even when accelerator is zero) and maximum value based on RPM. This latter one (max. torue value) has to be determined perhaps empirically perhaps by using a separate torque control and noting value for optimum torque at various RPM.

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

AC motors, multipoles, torque

Post by Johny » Wed, 07 Jan 2009, 18:39

bga. Yes, my ideal (but not carried away) control system would be moderate regen when accelerator is backed off and a hand control that could increase it. I think having it on the brake pedal may be dangerous. Picture hammering into a wet corner and just touching the brakes as a slight sped adjustment - not good if only the rear wheels do the early braking.
That's why a user controlled more-regen would be more intuitive for ex-ICE drivers.
I discuss the controlling asspects with my wife often and she does not like the Honda Jazz CVT AT ALL due to a feature thay call "Grade Logic" which slides the engine revs up on downhills when you have your foot off the accelerator to create more engine braking.
She wants the default condition to be VERY ICE like.

I originally didn't think an extra-regen hand control was easily possible with Danfoss/Lenze but now I can see a way - by increasing the minimum torque offset in the scheme in my previous posts. A purpose built EV VFD having regen control input could be used either way though.

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

AC motors, multipoles, torque

Post by bga » Wed, 07 Jan 2009, 19:12

Agreed,
Max torque will fry the motor in short order. It wants to be operated in the linear low slip region (25-50RPM slip, prpobably) and within its safe operating area. The torque curves indicate that there is a slip rate for maximum torque onthe motor. I had thought this to be one of the limits, with normal operation being in the mfr's reccommended range.

Speed and torque was what I was thinking, but I haven't developed the idea sufficiently to say what the relationship should be. It's likely to be system dependent. A good experiment subject. How to calibrate this is another question.

The battery voltage variation is an interesting problem. This points to a higher voltage battery or lower voltage motor being desirable to prevent the low end limit from being an problem issue.
Voltage and current issues as usual. Rewinds are not very attractive because of the $$.
I wonder how a 500V battery and a universal (US) 230V (thinks international chinese product) motor would go for regen on the high bus voltage?
It'd need a higher Amp controller because of the lower motor volts.

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Wed, 07 Jan 2009, 19:41

Speed cogging would be a topic on its own.
First impression is that it would make for uncomfortable driving in traffic that was not at a cog speed. A bit like using cruise control in traffic. But like cruise, an enable/disable would solve that and the cog may then be handy.

Regen seems to have no issue with a lower voltage motor as the regen voltage is easily several times the motor nominal voltage. I hope to experiment with this on the Gemini.
iMiEV MY12     105,131km in pure Electric and loving it !

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

AC motors, multipoles, torque

Post by bga » Wed, 07 Jan 2009, 20:44

We can forget about speed cogging.

Johnny, I agree that regen braking could cause surprises. I think that it has to be sufficiently subtle that it doesn't cause control problems.

It may be more effective on a front wheel drive where the regen axle is also the primary braking axle. I think that the algorithm should accomodate the properties of RWD vs FWD (logic programming?) and dry vs wet (adjustable parameter - the regen demand control).
With a bit of clever thought, the settings on the analog demand control could mimic an ICE. Perhaps a '4-cyl' and 'V8' mark on the dial?

I suspected that Regen would be happy with higher bus volts. This gives some hope to different motor voltages, except that this isn't the USA where lower voltage motors are common.

fuzzy-hair-man
Groupie
Posts: 118
Joined: Wed, 12 Nov 2008, 16:40
Location: Canberra

AC motors, multipoles, torque

Post by fuzzy-hair-man » Wed, 07 Jan 2009, 20:54

Most of this topic (slip etc) seems over my head but I figure a manual ICE like handling of regen would be good even to the extent of having indexed slots on your regen pot box to simulate engine braking in various gears in the ICE. Then I thought direct drive would pose a safety challenge for this, suppose I last used regen at a slow speed (eg 2nd gear in ICE) I have since accelerated to 100km in an ICE my gear changes would have brought the gears up as well so any regen/engine braking engaged would be like 4th or 5th gear, if in a direct drive BEV it might still be stuck in 2nd and as soon as I back off I'm spinning donuts down the road Image

So perhaps the regen would need to be initially set according to RPM (lower revs = higher regen torque) and therefore car speed and then you get the option to shift up or down from that point to further control your speed? Like shifting down gears into a corner in a ICE. But this doesn't let the driver hold on to a gear so they have extra engine braking available...

What if the user selects a position relative to the default regen level (which is decided based on RPM) so if I'm cruising slowly or coasting I'll select gentle regen relative to the default, if I'm fanging it or in town (where in an ICE I might hang onto a gear longer so increased engine braking is available) I might select harsher regen relative to the default level.

Not sure what electrickery would be needed to achieve this though, but it would avoid someone being overly/dangerously surprised but the regen level they had selected. Image Perhaps it would be possible to combine the 2? ie you can select the initial regen level and then allow changing up or down gears from that position? but it's getting rather complicated by this point... Image it should be reasonably intuaitive to use though, you set the initial regen level according to the kind of driving your doing and any 'changing down into corners' is done by the driver and resets after the accelerator is pressed?

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Wed, 07 Jan 2009, 23:35

It is good to think about the finer points of the application of regen.
Don't be concered though. There is plenty of control over what is done.

Inherent ABS, traction control and stability control (and the general redundancy of mechanical brakes) are all the future upsides of EVs.

Having pushed a (dirty hybrid) prius with regen, TCS and ABS hard on tight hilly roads and been delighted with the result, I can assure you the future looks bright.
If anything, I thought the regen was conservative as I guess toyota need to be conservative with the 'ICE' presentation.

The prius had two components - 'engine braking' simulated regen and top end of the brake pedal regen. (no real certainty of what was going on under heavy braking)
Note. The prius has a 'B' mode for extra engine braking where the CVT runs the (little 1.5L) ICE at higher revs. (gear 'paddle' has F N R and B)
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
woody
Senior Member
Posts: 1713
Joined: Sat, 21 Jun 2008, 02:03
Real Name: Anthony Wood
Location: Mt Colah

AC motors, multipoles, torque

Post by woody » Wed, 07 Jan 2009, 23:44

thoughts were along the same lines as fuzzy -- use the gear lever or at least the same numbering / lettering system to select the amount of regen.
I think Johny's wife is probably typical - people don't like change - accellerator, brake, clutch, gearstick + steering wheel have been pretty standard since 1930s I think - we should use them in the same way as much as possible.

Another possible input is the rear suspension height AKA the brake proportioning valve - this should put a realtime ceiling on the regen - perhaps in co-ordination with a wet/dry switch...
Planned EV: '63 Cortina using AC and LiFePO4 Battery Pack

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

AC motors, multipoles, torque

Post by Johny » Thu, 08 Jan 2009, 00:35

I don't have a gear-stick - it is (WAS) an auto (sob!). The column mechanism is so slack that I have been looking forward to ditching it. I almost can't get reverse anymore and I have already replaced the brass bushes and pins. Truth be told, that's the only reason I'm converting it to an EV - to get rid of the slack in the auto-gear selector. Image

Anyways - I'll opt for a purpose designed spring loaded thingy on the left of the steering column - or if that gets too hard to do, a push button for "more".

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Thu, 08 Jan 2009, 00:46

A NOS button in an EV ! Image

Does that mean you will require a 'less' button to cancel the 'more' ?

Image   
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

AC motors, multipoles, torque

Post by Electrocycle » Thu, 08 Jan 2009, 00:59

nah you put it on a timer, say 10 seconds of boost - with some sound effects and an LCD screen flashing "DANGER TO MANIFOLD!"


actually, my bike's controller has a setting to allow automatic boost for up to 10 seconds at a time when you accelerate hard.
It doesn't work now since I set the main current limit to maximum so there's no boost left :P
Last edited by Electrocycle on Wed, 07 Jan 2009, 14:01, edited 1 time in total.
The Engine Whisperer - fixer of things

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Thu, 08 Jan 2009, 01:22

It is like the 'turbo' button on old PCs.
Why would you bother turning the turbo off ?

I'm glad to see there are others who's policy on controller settings is 'if I can have it at all, then I want it all the time' !

Image
iMiEV MY12     105,131km in pure Electric and loving it !

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

AC motors, multipoles, torque

Post by Johny » Thu, 08 Jan 2009, 04:08

More REGEN you nutbags - and it's s momentary so you have to hold it in.
Geez.

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Thu, 08 Jan 2009, 04:43

Ok, now I get it. You have to hold on.

... sorry, just re-read the post, it was 'hold it in'.
At least it is only momentary ! Image
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
Mesuge
Groupie
Posts: 168
Joined: Wed, 05 Mar 2008, 07:37
Contact:

AC motors, multipoles, torque

Post by Mesuge » Wed, 14 Jan 2009, 12:37

acmotor wrote: Voltage can vary from 470V (low voltage cutout point) to 750V (fully charged after regen). This means that the motor voltage can be from 336VAC to 536VAC and this has a considerable effect on max torque.


Sorry if explained in other thread (couldn't find it), but how do you go about 704VDC (3.2V x 220) nominal in your future Rodeo/TS when 220batts. x 4.2V = 924VDC, I thought that Danfoss and 415V emotor combo tops at 750VDC bus input?
ImageCzech EV Club | blog, gallery, video | http://old.elektromobily.org

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Wed, 14 Jan 2009, 18:50

Yet to be finalised numbers, however this I have noted...

Danfoss applies power to external brake resistor (shunting DC bus) at 750V.
Danfoss allows DC bus voltage to rise to 850V when resistor braking selected (even if resistor is not there). At this point it lets go the regen on the emotor.

It is almost impossible to rapid charge TS in series to 4.2Vpc and keep the balance with 1C or more charge, so working out 220 x 4.2 =924V is academic only. Anyway, pack charging from regen will never get there. (Danfoss will limit this to 850VDC)
TS are some 98%? charged at 3.8V so no need to keep going, at least on a regular basis. I am trying to confirm this number and TS essential charging req.

A fully charged TS will only present 3.8Vpc after a short rest, 220 x 3.8V = 836V, and 3.3V after a longer rest, 220 x 3.3V = 726V. Both within Danfoss working range.

The target is that with voltage droop under 3C load the pack is still above 220 x (3.2 - .16 x 3) = 598VDC = 427VAC (more like 415V in practice).

With regen on red suzi I monitor DC bus and don't go above about 700VDC in the first km of driving, after that regen is not likely to get the SLAs back to full charge (unfortunately !).

Does this clarify or mudify ?
Last edited by acmotor on Wed, 14 Jan 2009, 09:01, edited 1 time in total.
iMiEV MY12     105,131km in pure Electric and loving it !

User avatar
Mesuge
Groupie
Posts: 168
Joined: Wed, 05 Mar 2008, 07:37
Contact:

AC motors, multipoles, torque

Post by Mesuge » Thu, 15 Jan 2009, 01:14

Clarify & Mudify, lol, thanks. Yes, I guesstimated you plan to charge upto 3.8V only. I'll try to consult the DC bus question with the other vfd brand specialist as to whether or not this is yet another DAN feature only..

So, to summarize this thread, in order to have very light & aero passanger car going ~120-130km/h in reasonable time and lowest budget, one has to have 18-22kW 415V (umodded) emotor and >60kW nominal (*capable ~120kW = 200% overtorqued) VFD? Another question mark, in fact is any major vfd brand able to supply its available max current to the emotor, in this example short impuls *415VAC & 290A giving ~450% over the motor nominal rating or this is again DAN only?

And with ~3:1 gearing/diff. the car will reach maximum power/top speed near 3500/4000rpm, torque falling from ~1500rpm? In case this is a realistic approach why not everybody is using it massively by now??

Also given the fact that listprice of VFDs is usually being quoted by dozens of % higher than the final checkout pricetag, I've seen as high as 40% discounts lately on them (ex works), so 30-40% off the price realistically. The weight and dimensions are usually giant for this desired output, so not suitable for every conversion..
Last edited by Mesuge on Wed, 14 Jan 2009, 21:00, edited 1 time in total.
ImageCzech EV Club | blog, gallery, video | http://old.elektromobily.org

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

AC motors, multipoles, torque

Post by acmotor » Thu, 15 Jan 2009, 01:50

I don't think it is Danfoss only, they are just the upmarket units most of the time with useful features.
Many VFDs are just HVAC models designed to run fans and pumps not process or torque control, however there are probably dozens of manufactureres of EV useable VFDs.

The industrial AC direction is still just the poor man's version of the high price AC drives e.g. ACpropulsion AC150, Siemens drives, etc.
I think the Chinese will fill the void soon with low cost AC drives.

The above sync area is still an issue for the industrial AC direction.
iMiEV MY12     105,131km in pure Electric and loving it !

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

AC motors, multipoles, torque

Post by Johny » Thu, 15 Jan 2009, 04:13

The Lenze is pretty much the same. Low voltage boost, 150% overload for 60 seconds etc. In considering the AC motor performance, provided the drive isn't an HVAC and runs totally from the DC Bus, then the VFD really doesn't appear to come into it much (Danfoss devotees may disagree).

The main reason I think that industrial components haven't been used much is that there are a lot of loose ends that you have to tie up. it's not the kit approach that commercial EV motor/controller combos provide.

acmotor made a point too, the sync speed and/or V/F ratio is the glitch. Life would be so much easier if we could easily get 240V 3 phase motors (or better still 180V). Then again, the rewind of $600-800 (even on a new motor) is not expensive when you consider the commercial EV AC motor prices.

User avatar
Mesuge
Groupie
Posts: 168
Joined: Wed, 05 Mar 2008, 07:37
Contact:

AC motors, multipoles, torque

Post by Mesuge » Thu, 15 Jan 2009, 10:42

I just like to add that from power/torque curves (could be dated) for Azure/Solectria AC55 system, they apparently peaked in power at 4000rpm and 1500-2000rpm in torque, so that sounds to be in line with the "hacked" industrial low speed VFDs approach..

On the other hand, the top tier guys like Siemens, ACP/Tesla, and others are quite a different league at 2-3x the max rpm or even more..
Last edited by Mesuge on Wed, 14 Jan 2009, 23:44, edited 1 time in total.
ImageCzech EV Club | blog, gallery, video | http://old.elektromobily.org

User avatar
Mesuge
Groupie
Posts: 168
Joined: Wed, 05 Mar 2008, 07:37
Contact:

AC motors, multipoles, torque

Post by Mesuge » Thu, 15 Jan 2009, 15:55

woody wrote:
Richo wrote: Here is a comparison between an ACIM and the std BMW 318i E30 motor.
Image

The ACIM would be a 11kW 4-pole 132-frame Tn=72Nm Tm=2.9xTn=209Nm.
Except it is rewound for around 200V.
So it can be used up to 2 x sync freq (3000RPM=22kW)
It would weigh about 60kg.

From the above graph you can see the ACIM torque is better than the original ICE upto about 4200RPM.
Which is typical driving rev range.
Also the Tm of 2.9xTn is quite conservative ie no requirement for high breakout torques such as 3.6 which is hard to find.

With the 22kW rating it should be good for 120kph depending on car.
I think it makes a good compromise for vehicles that are keeping the gearbox.
I still reckon the "field weakening zone" above sync speed means your torque drops off proportionally to the square of the inverse of the frequency, but even so you'll have more torque until about 3800 RPM. This just means that revving the crap out of your ACIM is counter-productive :-)

Also at peak Power (3000 rpm) you'll be pushing 8 times nominal factory current through the windings which the motor should handle for a minute - you shouldn't need that long unless you go lead acid...


Did anybody (apart from Kermit's experiments) perform this hack on any (smaller output) VFD and motor combo? Will the efficiency map for this emotor plotted against RPM remain essentially the same or there might be 1-2x efficiency gaps? Improved forced air cooling, water cooling, heatpipes, peltier, ... Image
Last edited by Mesuge on Thu, 15 Jan 2009, 05:21, edited 1 time in total.
ImageCzech EV Club | blog, gallery, video | http://old.elektromobily.org

Post Reply