Running HV EV Motors at 48V?

[cloudy]

For IVA approval of one-off scratch built cars, there is an electrical approval (v. expensive) exemption only when using up to 48V for traction. With tesla inverters able to deliver ~1000amps you should in theory be able to deliver about 48kW/64hp with a bunch of paralleled 10S packs. I guess there are all sorts of efficiency issues with this approach, but it does open up cheap donor parts. Will these motors generally be able to draw these currents on such a low voltage?
A small Tesla motor happily spins even at 40V, but no idea what sort of power it's able to generate...

[Kevin Sharpe]

seen this?

https://openinverter.org/wiki/Legalities/United_Kingdom

[cloudy]

I have, but that only applies to conversions rather than built from scratch (which require an IVA) - and for IVA with over 48V - ECE R100 test certification is required beforehand...

[celeron55]

It comes down to the maximum torque of the motor overall and the rpm of the motor at 48V.

Let's take the Leaf motor, which with the factory inverter using 350V input voltage outputs about 250Nm (500A) up to 3000rpm and from there upwards to 10krpm about 80kW. At 48V it's still able to output 250Nm, but speed will be reduced directly in relation to voltage to 1400rpm. That's slighly less than half the rpm that it starts outputting the constant 80kW which means the output power will be slightly less than 40kW.

Inverter, motor and wiring efficiency and lifetime will suffer due to high currents.

You will need to have a manual transmission and clutch, because the motor never gets to its constant power band and you will always want to keep it at the maximum possible rpm for maximum power.

So, yes, for a small car you could do this, if you really have to. It won't be fun to drive, but you can get it to highway speed if it has tall enough gearing to do that at the low motor rpm you will be limited to.

No small car has tall enough gearing to do that without modifications. Maybe a rear wheel drive vehicle where you can lower the final ratio to 2:1 or so?

It's a ridiculous idea though. Go for at least 100V. That way you don't have to modify the gear ratios and can enjoy some of the constant power band goodness.

EDIT: That having been said, the leaf motor, when opened, suspiciously looks as if you could rewire it in some way, maybe to half the voltage and double the current? You could do some research there.

[doobedoobedo]

Have you considered sticking an ICE in for the IVA then converting it? Or a 48V forklift motor and upgrading it after it's passed to save re-classification?

[slow67]

EDIT: That having been said, the leaf motor, when opened, suspiciously looks as if you could rewire it in some way, maybe to half the voltage and double the current? You could do some research there.

They Leaf motor is Wye wound. You could undo the Wye winding and wire it delta. The motor would spin ~1.73 times faster on the same voltage. Keep in mind it will take 1.73x more amps for the same torque.

[arber333]

They Leaf motor is Wye wound. You could undo the Wye winding and wire it delta. The motor would spin ~1.73 times faster on the same voltage. Keep in mind it will take 1.73x more amps for the same torque.

Well I would think leafmotor wiring is wound with constant 500A in mind. Big 50mm2 cables remember...
But I also think motor could be rewound to some lower voltage. You could also rewind it for lower pole count. Maybe 4poles instead of 8. Then you would get less torque but higher speed at the same v/Hz applied.
What would be the benifit there though...
Just to satisfy the under 60V rule?
Why not use forklift DC motor then? They are made for that application.

[slow67]

They Leaf motor is Wye wound. You could undo the Wye winding and wire it delta. The motor would spin ~1.73 times faster on the same voltage. Keep in mind it will take 1.73x more amps for the same torque.

Well I would think leafmotor wiring is wound with constant 500A in mind. Big 50mm2 cables remember...
But I also think motor could be rewound to some lower voltage. You could also rewind it for lower pole count. Maybe 4poles instead of 8. Then you would get less torque but higher speed at the same v/Hz applied.
What would be the benifit there though...
Just to satisfy the under 60V rule?
Why not use forklift DC motor then? They are made for that application.

You can’t rewind it for lower poles without making a custom rotor.

Changing from Wye to Delta is like rewinding, but without having to have the entire motor rewound. if you can crimp/solder you can do this. FYI the Leaf motor is already a pretty high rpm motor compared to industrial AC motors as it stands. Also a controller to do thousands of amps and low volts probably isn’t on the shelf at this point.

[arber333]

You can’t rewind it for lower poles without making a custom rotor.

True, my experience with changing motor poles is from ACIM motor world.

Then i will try to get my dual Ampera inverter to make 1000A into low voltage battery. Maybe stacking multiple Volt battery 48V sections in parallel could do it? 48V x 1000A is 48kW. Could make 2000rpm and a lot of field weakening from then on to get at least to 4000rpm....

[slow67]

You can’t rewind it for lower poles without making a custom rotor.

True, my experience with changing motor poles is from ACIM motor world.

Then i will try to get my dual Ampera inverter to make 1000A into low voltage battery. Maybe stacking multiple Volt battery 48V sections in parallel could do it? 48V x 1000A is 48kW. Could make 2000rpm and a lot of field weakening from then on to get at least to 4000rpm....

I would still reconfigure from Delta to Wye. Then he could get a little over 2400 rpm before field weakening, possibly 4,000 rpm with fw. Also the phase cables would be double the size at the motor.

You could even go a little crazier, do 1 whole volt inverter per phase. At 3k amps per phase at 48V makes 144kw (obviously only for short bursts). You should be able to break up a Volt battery pack into 8 sections of 48V and wire them in parallel also. Under a 3,000 amp load, the batteries would only see 8C, not too much of a load. Make sure your cable sizes and contactors are up to snuff!

[celeron55]

Oh man, you'll be spending thousands for contactors and cables at that point assuming you'll find ones. And they'll weigh hundreds of pounds!

3000 amps? No, just no.

[cloudy]

Great discussion! This really is just an thought exploration of what realistically could be squeezed from under 48V using donor parts. It is entirely possible (and dare I say, likely) the UK also moves this legislation onto conversions too in time. Due to the certification cost - we'd all be in the same boat if we wished to build/convert our own. My use is for an extremely lightweight (sub 400kg) build, so I think it is feasible to have "car like" performance at 48V, but not sure whether it can be done using the common conversion parts ie: Teslas/Leafs etc or whether you need to get exotic...

[arber333]

Oh man, you'll be spending thousands for contactors and cables at that point assuming you'll find ones. And they'll weigh hundreds of pounds!

3000 amps? No, just no.

Well then I would first traverse from Igbt control to mosfet.
Once I had a mosfet inverter ZAPI ACe5 which could develop 120kW power from 120V battery. Inside it used 70 mosfet pairs per phase to develop 1000A. I imagine best way would be to make a new power section. But that is a lot of work.

Cheap mosfets i would use are up to 80Vdc. That is plenty for such use.
At such a low voltage we could use one Prius Gen2 capacitor for each phase. It would be a bit larger inverter, but we could make it in triangular shape like tesla makes their inverters. Now we only need to design cooling system for each phase branch. If we use such low voltage we could use copper pipes for current conduits. That would cool them down and allow transfer of more than 1000A per phase.
I think Johannes gen2 inverter brain lends itself to any voltage. Current sensors in a chip such as Melexis could be used at a propper distance.
Since we would use LV and there really is no possibility of driving 3phase motor with DC short circuited i imagine we could get away with mechanical contactor.
A spring and scissor mechanism that would be operated by a cable from the cabin. It would require regular maintenance (brushing, cleaning) and a set of 500A fuses to brake the current in case of real SC.

Anyone care to begin?

[arber333]

On other thought... what about 3x DC motor controlers commanded by Johannes brain?

It really is more academic, since 50mm2 cable can reliably carry about 500A for extended time and about 1000A for a burst. I imagine Leaf motor would have difficulty running at that low voltage for any extended time.