RX400H transaxle

AC, DC, amps, volts and kilowatt. It's all discussed in here
Ilya
Noobie
Posts: 4
Joined: Fri, 10 Jun 2011, 10:13
Real Name: Ilya
Location: Nizhny Novgorod

RX400H transaxle

Post by Ilya »

I measured the temperature sensor from MG2 Toyota Prius by Calibrator, and I'm sure that RX400H transaxle is the same) This NTC thermistor is 3900-3950K 47kohm!Image
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

Thanks for the graph squire. I like the compactness of this electric transaxle too & the relative ease of mounting but the main problem with this IPM motor is finding a controller to run it. The Tritium Wavesculptor can drive it but is currently limited to 450v whereas the motor needs 650v to develop it's rated output.
I'm sort of in Limbo at the moment. Ryan (in the USA) is going ahead with his project but I worry that a 450v build won't develop enough horse power to be worthwhile.
bga
Senior Member
Posts: 492
Joined: Mon, 01 Sep 2008, 19:27
Real Name: Bruce Armstrong
Location: Perth WA

RX400H transaxle

Post by bga »

Is it possible to get a motor rewinder to attack it and change the wire gauge so that the operating voltage comes down to something better for the batteries and controller?

The same rules as any 3phase motor should apply.

It's not the end of the world, but I can see it from here.
Ilya
Noobie
Posts: 4
Joined: Fri, 10 Jun 2011, 10:13
Real Name: Ilya
Location: Nizhny Novgorod

RX400H transaxle

Post by Ilya »

Hello! Have a look at the schedule, whether you need to use it at 100% speed?

The inverter will do himself. As an option to use an inverter from the Toyota Prius ZHV30 or modernized industrial frequency converter.
At the expense of TRITIUM - The WaveSculptor uses 600V IGBTs as the power switching elements.
It is evident he is very well done, once such a small margin for voltage at the IGBT, but I would not operate it on 450V DC.
All that I know of converters operate with a twofold margin of voltage transistor! In order to use the rear motor from RX400H - Use modules to 1200V!
If you have the desire and experience - something you can rewind any motor.
The simplest thing to lower the voltage - connected delta winding.

I not make out the engine, he assembled precision and filled with oil. If the photo from outside are needed - i can do;)

Ilya. Image
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

Ignorance is my big stumbling block Bruce. If I take it to a rewinder I've got no idea what gauge wire to ask for.

Hey nice graph Ilya, what other gems you got tucked away over there in Nizhny Novgorod? You're right about the Wavesculptor I think, load voltage is more like 360v (450v was the max pack voltage after a charge).

TJ tested my IPM @ 360v & reckons it would be marginal in anything but an ultralight car. I could drop it into something like a Suzuki cappuccino I guess but I fret about driving a vehicle that small on todays roads with 2 tonne SUVs on the loose. I was hoping for something more like a Celica donor.

Do you have any info on the Lexus controller that used to run this motor in the Rx400h Ilya?
Tritium_James
Senior Member
Posts: 683
Joined: Wed, 04 Mar 2009, 17:15
Real Name: James Kennedy
Contact:

RX400H transaxle

Post by Tritium_James »

Our controller can run at full load at 450V, no problems. We have better power stage layout than most industrial drives, and lower inductance bus caps = less overshoot on switching edges.

The main reason we specify 450V max instead of even higher is because we need enough time to react and shut things down if the batteries are disconnected when doing full regen current, before the voltage can get to 600V and damage the power silicon.

Ilya, unfortunately the bit you have drawn in on red on your graph is not correct, regarding the reduction in speed vs voltage. This is an interior permanent magnet motor, not the usual permanent magnet type where speed scales linearly with voltage. For an IPM, only about 1/3 to 1/2 of the torque (and speed) comes from the permanent magnets, depending on the ratio the motor designer has chosen. The other 1/2 to 2/3 is from the switched reluctance part of the design (saliency on iron, etc)

Running from a reduced voltage will also mean that the point where torque starts dropping off ('base speed' = about 4500rpm in your graph) will also be lower. Therefore power output is lower too. If you have a nominal 360V pack (eg using our controller, 450V top of charge, partially discharged cells, under load) that base speed will only be at 650/360*4500 = 2500rpm. Power will also be less by around the same ratio, ie 28kW. This is going to be pretty wimpy in anything except a lightweight vehicle.
Ilya
Noobie
Posts: 4
Joined: Fri, 10 Jun 2011, 10:13
Real Name: Ilya
Location: Nizhny Novgorod

RX400H transaxle

Post by Ilya »

I read your forum, a lot of interesting and useful for yourself found it!

James, your converter according to the description of - High End Inverter!
You are absolutely correct on schedule and by the power! IPM motor has a soft mechanical characteristics (torque / speed). By reducing the voltage curves of torque and power will shift to the left and down the chart.

Pictures about sensor-resolver. Toyota uses a sensor, digital converter company Tamagawa.
Image
Image
Image
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

It would seem from Ryan's experience in the USA with his Saturn Sky that 900Nm (130Nm motor x ~7:1 gearbox)just isn't enough starting torque for a 1000kg car. It sort of bugs me that I've purchased a pup because the numbers looked OK on paper, the old AC24LS motor/AT1200 gearbox combo for example was 92Nm x 10:1 = 920Nm & AzureDynmaics claimed it was good for vehicles up to 1600kg.

Anyway numbers from the Leaf et al seem to be point to about double that amount of torque for 'sparkling' acceleration with 1000kg. I feel like someone has moved the goal posts these past few years.

So I find myself back at square one but I still really like the idea of a drop in electric transaxle. The obvious option is an off-the-shelf motor mated to something like the 31-03 but getting one is not looking easy.

Does anyone know of any available alternatives differentials/transfer cases with a ~8:1 reduction?
Last edited by PlanB on Sun, 27 May 2012, 08:06, edited 1 time in total.
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

I pulled the end plate of the electric transaxle today, naively hoping to find n wires (where n is a single digit number). I guess I was hoping it might be set up as a 650v wye that I could reconfig as a delta to run from a 375v battery pack.
What I found was 12 wires to each of the 3 phase terminals. Now I know how Pandora must have felt! If anybody has got any clue on why an 8 pole motor has 36 coil ends & why are they are terminated in 3 bundles of 12 & what that means exactly in terms of running it from a lower voltage I'd be very grateful.
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

RX400H transaxle

Post by Johny »

My take is that you have LOTS of options.
How do you know that it's an 8 pole motor?
I read on an engineering forum recently where someone had a trivial problem (relative to yours) of 6 wires and was lead through how to determine start-end of each winding by injecting a DC pulse and noting the direction of induced voltage on the other windings.
I guess the first step is to figure where these windings would theoretically be in an 8 pole motor.
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

That's my problem John, too many options. An N phse, M pole motor should have NM/2 coils. So the pug, with 3 phases & 8 poles should have 12 coils. But, with 36 tails, I figure I've got 18?

Does anyone know how slots figure in all of this? This thing as 48 slots, if I unrolled the stator it would look like this......
LLllllLLllllLLllllLLllllLLllllLLllllLLllllLLllll
where L= lots of wires & l=less wires
Last edited by PlanB on Wed, 30 May 2012, 12:24, edited 1 time in total.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

RX400H transaxle

Post by weber »

PlanB wrote:Does anyone know how slots figure in all of this? This thing as 48 slots, if I unrolled the stator it would look like this......
LLllllLLllllLLllllLLllllLLllllLLllllLLllllLLllll
where L= lots of wires & l=less wires

Until you posted that unrolled stator diagram I was going to point out that reluctance motors often have a different number of stator poles to rotor poles. I was going to suggest that what you had was an "8 pole 12 slot" motor, where the "slots" referred to here are really the gaps between the stator poles (and so there are the same number as there are stator poles), as opposed to the physical slots in the stator iron. These are often the same thing, but not in your case.

See http://www.icee-con.org/papers/2008/pdf/P-071.pdf

Actually, your stator diagram can't be right. If this is a 3 phase motor, the number of stator poles must be a multiple of 3. You've shown 8 stator poles.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

Definitely 48 holes with wires in them in 8 groups of 6. Thanks for the link Dave, nothing like a little light reading before dinner!
Image
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

RX400H transaxle

Post by coulomb »

All those pieces of coloured spaghetti may be where they've wired coils in series.

Though it may be where they have put coils in parallel, but if it's wound for 650 V, it seems unlikely that many coils are in parallel.

Then yet again, it's hard to imagine why 12 wires come to one terminal, unless a lot of them are in parallel.

Very strange.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

RX400H transaxle

Post by weber »

Ah! Now that I see the photo ...

I think you will find the same amount of wire in every one of those 48 slots. I'm pretty sure there are two more hidden sets of 8 (pole-windings), offset by +- 2 slots from the set we can see in the photo. The innermost set we can see, is just for one phase.

I believe the explanation for why there are 12 wires at each terminal is that the windings are made with either 3, 6 or 12 wires in parallel. I'd bet on 6 wires in parallel, all 8 pole-windings in series within a phase, and a delta configuration.

[Edit: Added hyphen to "pole-windings"]
Last edited by weber on Sun, 08 Jul 2012, 17:04, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

RX400H transaxle

Post by coulomb »

weber wrote: I'd bet on 6 wires in parallel, all 8 pole-windings in series within a phase, and a delta configuration.

Ah, a sort of hex-filar winding (cf. bifilar, but with 6 wires instead of 2)? Basically to keep the current carrying capability high and the resistance low, but using thinner wire because it is easier to bend and to wind compared with wire that is 6 times the cross-sectional area?
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

RX400H transaxle

Post by weber »

coulomb wrote:Ah, a sort of hex-filar winding (cf. bifilar, but with 6 wires instead of 2)? Basically to keep the current carrying capability high and the resistance low, but using thinner wire because it is easier to bend and to wind compared with wire that is 6 times the cross-sectional area?

Right. But I thought I'd better find out the proper terminology, as used by motor winders. It's called "6 in hand" winding.

And what I called a "pole-winding" above is called a "pole phase group" or "PPG", sometimes shortened to just "group" when context makes it clear. This is particularly confusing terminology in the case of the motor under discussion as each "group" consists of a single "coil". Typically each pole phase group consists of 2 or 3 concentric coils to approximate a sinusoidal field distribution.

So the above motor appears to have 8 poles x 3 phases = 24 pole phase groups (PPGs) and 1 coil per PPG. And is laid out as a basket winding.

I learned the terminology from this wonderful resource:
http://www.uiitraining.com/b51a/100/15104training.htm
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Richo
Senior Member
Posts: 3737
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

RX400H transaxle

Post by Richo »

PlanB wrote: It sort of bugs me that I've purchased a pup because the numbers looked OK on paper, the old AC24LS motor/AT1200 gearbox combo for example was 92Nm x 10:1 = 920Nm & AzureDynmaics claimed it was good for vehicles up to 1600kg.


I have repeatedly emphasised that the AC24LS/AT1200 was no good for 1600kg cars.

Very bold claims.

Our average 1600kg car in Australia (Commodore) has 300Nm then through a gearbox then the diff ie ~10:1 = 3000Nm
Then you have all the other 5 gears for acceleration.
Being stuck in first with 1/3 of the Torque would be terrible.
So the short answer is NO but the long answer is YES.
Help prevent road rage - get outta my way!
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

Thanks for the collective wisdom lads. It's an intriguing beast, why they went for a 650v winding on a 288v battery pack I'll never know, maybe it's corporate mish mashing? Like the MD notar helicopter. The original (tail rotor) chopper had a step down gearbox for the tail rotor drive shaft. To get the fan speed for the notar MD then added a step up gear box after the step down, very elegant.
So what next, anybody know any brave motor rewinders in Sydney?
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

RX400H transaxle

Post by weber »

PlanB wrote: Thanks for the collective wisdom lads. It's an intriguing beast, why they went for a 650v winding on a 288v battery pack I'll never know, ... So what next, anybody know any brave motor rewinders in Sydney?

Did I miss something? Why would you pay to get it rewound when you don't know if you can get a controller to work properly with its IPM-ness?

And if you're prepared to get it rewound, you've got nothing to lose by having a bit of a poke around the windings to see if you can find the midpoint jumper of each phase, which you could cut in order to wire the two halves of each phase in parallel to convert it to half voltage.

You'd start by cutting the string and teasing the sleeved wires away from the coils so you can see more. Be careful not to nick any of the enamelled wire with any hard or sharp tools.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

TJ has run it previously on a wavesculptor Dave, it's internal PMness worked just lack lustrely. Do we know for sure there is a midpoint jumper in there? I know I've probably got nothing to loose but poking around in there like a witch doctor at an extispicy spooks me a bit.
Let me see if I've got this? You reckon the per phase innards might look like this..
1B-c-j-c-1A            where c = coil
2B-c-j-c-2A            and j = join
3B-c-j-c-3A            with A & B being 2 of the 3 motor terminals
4B-c-j-c-4A            
5B-c-j-c-5A
6B-c-j-c-6A
7B-c-j-c-7A
8B-c-j-c-8A
9B-c-j-c-9A
10B-c-j-c-10A
11B-c-j-c-11A
12B-c-j-c-12A
Last edited by PlanB on Sun, 03 Jun 2012, 06:41, edited 1 time in total.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

RX400H transaxle

Post by weber »

No Kris. I recommend these two tutorials, if you haven't already read them, http://www.uiitraining.com/b51a/100/15104training.htm, so we are talking the same language.

From the limited information I can glean from that single photo, my guess is that it is connected in delta and so the 12 wires you see at a terminal are 6-in-hand for one phase and 6-in-hand for another phase. "6-in-hand" means that a bundle of six wires is treated as if it was one wire. And I guess that, for each phase, that bundle of 6 wires goes continuously through all 8 coils before returning to another terminal.

A "jumper" does not involve any joins. It is merely the name given to a part of the winding where the bundle of wires hops from one coil to the next. So if all 8 coils are in series, as I'm guessing, then there is, by definition, a middle jumper. Whether you can get to all 3 of them is another question. Being a basket winding, there is an inner phase, an outer phase and a middle phase. The middle jumper of the middle phase may be difficult to get to.

To find the middle jumper of a phase you should follow the ends of the phases from the terminals to see what slots they enter. The middle jumper of a phase will jump between the two slots that are 180 degrees from the start and finish slots.

You can determine which are two ends of the same phase by which slot numbers they enter. I've shown slot numbers for each phase on a different line below. A hyphen represents a coil.

1-6 7-12 13-18 19-24 25-30 31-36 37-42 43-48
3-8 9-14 15-20 21-26 27-32 33-39 40-44 45-2
5-10 11-16 17-22 23-28 29-34 35-40 41-46 47-4

For example, if two ends (from terminals) go into slots 6 and 43, then the middle jumper of that phase will go between 19 and 30.

But that photo is really very little to go on. Maybe there is a star-point buried there somewhere. Maybe each phase is already done in two halves which are paralleled.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
T2
Groupie
Posts: 121
Joined: Sat, 08 May 2010, 04:51
Location: ON,CANADA

RX400H transaxle

Post by T2 »

AC24LS/AT1200 was no good for 1600kg cars
Azure Dynamics now in receivership

they went for a 650v winding on a 288v battery pack
Are we talking Lexus?

Until the arrival of the 2004 Prius the inverter bus was tied to the 273Vdc battery which explains the anaemic performance of previous Prii.

With the addition of the voltage Upconverter - which surprised some in the industry - not only was the power band extended over a larger speed range but there were probably cost savings in requiring fewer battery modules to build the 201Vdc battery. This battery has now become a standard item.

In some models Normal mode remains at 500Vdc but with an available Sport mode sitting at 650Vdc. MG1 and MG2 both turn much faster with the higher voltage while inverter current remains the same.

This philosophy is good for constructors of pure electrics also but it should be borne in mind that only 21Kw is being boosted (100A) here.

OTOH I was thinking the principle could be used to maintain V/Hz while an ACIM accelerates beyond twice its base speed.
T2
T2
Groupie
Posts: 121
Joined: Sat, 08 May 2010, 04:51
Location: ON,CANADA

RX400H transaxle

Post by T2 »

I've since read the front end of this thread more thoroughly and need to redo my previous post.

"AC24LS/AT1200 was no good for 1600kg cars"

Complaints about 100N-m into 10 : 1 ?

AS a matter of interest back in the day, each of the Impact's (prototype vehicle for the EV1) front wheel motors generated 47lbs-ft. About 125N-m total, giving an 8sec to 60mph performance from 320V of Pb-acid. The Nissan Leaf is still a couple of seconds slower than this I believe.

In this light the AT1200 gearbox spec requiring just 100N-m is not far off the mark at all. The achilles heel of AZURE was their motor/controller kit. Their ACIM was junk by all accounts and on introspection of the torque curves it's clear the motor winding has an excessive V/Hz, besides which their controller was dated and meagre in output.

they went for a 650v winding on a 288v battery pack
I see your problem. You don't have the controller with the upconverter - you have just the rear transaxle ?

You may know that the RX400h, unlike Prius, has two traction motors MG2 & MG3. With the combined torque from two driven axles it is permissible to have MG3 avoid the more extreme ratio of 10 : 1. Plus this vehicle is spec'd to go beyond the Prius' 100mph limit. The 7 :1 ratio in effect throws away potentially 30% of the motor torque that could reach the road for the dubious pleasure of 120mph operation.

From what I have learned over the years I would recommend as best practice for an electric transmission to be a 10krpm motor working into a 10:1 gearbox to provide 100Km/hr. I would add the proviso that the motor V/Hz must not be compromised below around 6600 rpm. And the icing on the cake would be the ability to operate at up to 300% nominal current below this particular rpm without exceeding the battery 4C rate since what is termed a "good launch" is key to establishing useful acceleration benchmarks.

Although I have rejected the idea of a Prius electric-only conversion in the past it was mainly because the electronic interface is not too constructor friendly, voltagewise that is. Then recently, I realized that a donor Prius may in fact bring something to the table after all if the constructor is prepared to discard the two motors, MG1 and MG2, together with their electronics package. A technology you cannot master is a liability not an advantage.

In essence the engine and then MG1 need to be removed first up.
Next, the planet carrier/gear cage has to be mechanically restrained somehow so that it is prevented from rotating relative to the chassis.
An induction motor as specified should then be mated into the sun gear of the HSD planetary in place of MG1.

Doing this enables the HSD planetary to deliver a reduction ratio between the 30t sun gear and 78t ring gear of 2.6 : 1.
The ring gear of the HSD couples to the front wheel axle through a conventional geartrain and differential with a cumulative ratio of 4.11 (PRIUS MY2004-9) Combined with the HSD planetary ratio this gives an overall ratio of 10.7 : 1 which is in the ballpark for a non compromised design.

At 100km/hr the motor will be rotating at around 10Krpm causing the four 24t planet gears in the stationary carrier to rotate around 12.5Krpm. Faster speeds may be possible depending on the durability of the planet gears.

I believe with T-J's wavesculpture drive there is available a range of high speed induction motors with copper rather than aluminum rotors. The use of these specific motors will take the guesswork out of the setup and at the same time push electrical effcy into the nineties.

Regarding lubrication, a suggestion was made years ago on the Yahoo Group, Prius_Technical_Stuff to reduce gearing losses   by replacing the OEM transmission oil with Mobil ONE in the transaxle. Possibly this low viscosity oil will reach the HSD as well or some provision might be made.

- PlanB You were asking for suggestions so feel free to tear this to bits.
T2
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

RX400H transaxle

Post by PlanB »

An interesting read T2. I went for the Lexus all electric rear transaxle because I liked the idea of a drop in 40kg 50kw module in a 700kg glider, but it would seem to be a flawed device. Ryan tells me that it can't be over torqued with extra current for fear of depolarising the magnets & TJ won't sell me the HV version of his wavesculptor to run it in the sweetspot anyway.
I don't have the where with all to rejig Prius transmissions, still fancy the general idea of a drop in & a Tritium motor with a single speed box like the BorgWarner 31-03 would be a nice combo except that 31-03s are a bit thin on the ground.
Meanwhile my little EV stash keeps earning interest & I guess someday a manufacturer will bring out something less ugly than a Leaf & I'll finally have something to plug in.
How are things with that French outfit in Canada? Blade was using there controller/motor combo for a bit. Any sign of an off-the-shelf gearbox in use with any of their kit over there?
Post Reply