ALL-Electric Prius

All kidding aside with that one, Are you intending on replacing the whole transaxle? If not, You'd have to do extensive mechanical rework to the transaxle and then you'd still be left with a pretty lethargic EV. I'm going the PHEV route, so then I have the best of both worlds, EV in town for short trips, but I can go for a road trip anytime burnin' Dinosaurs.

 

Personally, I think that's a bad way to go. You'd be much better off replacing the complex Prius transaxle with a single electric motor. From the ORNL 2004 teardown:

With 3 electric motors, and the PSD, your system efficiency will be pretty low, not to mention a control nightmare, as you will have to hack the CAN bus to the Engine ECU to generate a torque signal for your new motor.

Just take a transaxle from a Yaris and bolt a single motor to it. Substantially less loss, and then the control system is super-simple.

 

I'm not sure of that! Two things to consider here. First of all, as we've discussed in other forums, the battery can't supply all of the power to the motor (MG2) that it is capable of using. Typically the rest comes from MG-1 when the engine is running. I believe this is a bottleneck in the inverter itself.

The being the case, using a custom 3-phase AC motor-controller, it could be possible to drive MG2 at is full potential.

Second thing - I saw a thread a few years back on some other website where a guy mentioned that he removed the engine and he welded something together inside the PSD that basically allowed him to use MG1 and MG2 together for driving the car. I don't remember which part, but I want to say the planet-carrier. Anyway, the point is that with MG1 and MG2 together, the car had plenty of power. Again, being controlled with a custom motor controller, battery pack, etc.

 

Ok, if you build your own inverters (you'd need 2), then maybe you have something. The existing drive in EV is limited mostly by what the boost converter is capable of. According to ORNL, this is ~20kw, whereas in PSD mode, shifting power from MG1 to MG2 is handled all on the boosted bus so it's limited only by the IGBTs in the inverter.

The other option would be to keep the inverters, but install a 500v pack and connect it directly to the IGBT bus. Then you'd only be limited to building your own HV ECU replacement. (which is still a non-trivial amount of work!)

I'd love to experiment with this myself. If I ever came across a cheap used/salvage Prius, maybe with a blown engine or MG1, It would be interesting to remove everything except MG2 and see what kind of peak power it's capable of. If you could keep it cool, and not depolarize your neodymium, you might have something cool.

But adding another electric motor is not very sensible in my eyes.

Also: I can probably see how much parasitic loss the engine adds while spinning in valves-off mode. I suspect it's probably not as bad as we think. There is no pumping loss, which is the biggie, so it's down to the oil/water pump and internal friction. Given that a normal small 4 cylinder (with pumping loss) takes less than 2kw to crank, I suspect it's going to be less than the total losses in the transaxle. (some if which are almost impossible to avoid, such is in the diff)

Hypoid gears are cut to be quiet, not efficient. You want efficient, you are likely going to have a noisy powertrain!

A cool solution would be 2 wheel motors in the back, no drive mechanics up front, only batteries. Too bad wheel motors aren't "there" yet, and then you also have the pesky unsprung mass problem.

 

I'm replacing the battery ECU with one of my own design. It's the last device on the CAN bus, so it can talk to any of the other CAN nodes in addition to doing it's normal job, so I plan to add lots of different functionality.

One of the goals is EV mode up to 70mph. (yes, the ICE still must spin with valves disabled over 52mph) I've built my own charger and have assembled a 7.2kwh A123-based Lithium pack.

 

It probably hasn't been done much because it isn't a great idea. I think before the engine is removed just run the car out of gas and see what the situation is.

Because really the limiting factor as far as I know isn't the Prius electric motors, it is the batteries. So unless you're putting more batteries in, there isn't a benefit, but if you do put more batteries in you might as well keep the engine and either do the "52 mph hack" or else do as was done in the video and remove the engine components and leave a bare block, which to me the only benefit there is a bit of weight reduction. The PICC conversion has already accomplished 70 mph in EV only mode, but unfortunately they won't market just that part of their kit, so hopefully pEEf or somebody else will come along and figure it out too. Obviously it can be done, which is the first hurdle to any project.

It will always come back to the range of the batteries, not the speed of the car. And unfortunately, as the speed of the car increases, the range of the battery drops quickly.

 

The biggest limit to EV power is actually the boost converter. It takes the battery voltage (201.6v nom) and boosts it up to whatever is needed at the time. This is needed because the Prius uses permanent-magnet based brushless DC motors (DC is a misnomer, as it's actually 3-phase AC). As these motors spin faster, the develop more back EMF (the voltage raises). To keep the motor delivering high power, you must have a high voltage, as back EMF is a function of RPM. Here's ORNL's table:



MG2 is rated at 50kw according to ORNL, but the boost converter is only rated for a little more than 20kw. This means in battery mode you can't really get more than that total unless you do some serious hacking. Normally the ICE is spinning MG1 which is producing power that's put back into the invert high-voltage bus, so that's how the Prius can come up with 50kw for MG2. (~30kw from MG1, ~20kw from the battery via the boost converter)

So really, the Prius just isn't made to be an EV, you are much better off just putting a new motor with lower voltage windings, coupled to a simple transmission. Yeah, you can hack all kinds of things, but by the time you do all that you are going to have WAY too many hours into it.

Now as long as you have the ICE for power when you need it, the ~20kw or so that EV mode can provide is just about enough to sustain the car at speed on a flat surface. So if you have an upgraded battery, can can make good use of it and just keep the ICE for rapid acceleration and long trips. This way you only need 1 car for all your trips.

I currently own 2 EVs in addition to my Prius, and they are fine for city driving. Until we have charging stations available on our highways and packs that can do less than a 10 minute fast-charge, we are stuck using the ICE I'm afraid.

I'm doing what I can within my means to reduce the petroleum I use and trying to limit my emissions. My goal is to get the Prius so it can serve both purposes and give one the EVs to someone who can benefit more from it. (it's for sale BTW)

 

It is simply a matter of physics, not beliefs. Now, it is true what pEEf says about power, but that is comparing apples to oranges, because (almost) everything is about range in my opinion. Yes, you can't take 30 minutes to get it up to 30mph, but it doesn't need to go 80 mph either (not that pEEf said that).

As far as I know, the PICC uses strictly battery power, with absolutely no input from the engine, and can (no matter how long it takes or how inefficient it supposedly is) get the Prius up to 70 mph. But what is the range at that speed? It obviously is much lower than doing 30 mph.

Now, I actually don't want my car generating CO2 any more than absolutely necessary, but there is a limit to what is practical to do with the Prius as it was manufactured. And, although I'm not to the level of pEEf or mrbigh or others, I have done what is practical for myself at the present time and put in the Hymotion kit. The last tank went over 1100 miles, at about 125 mpg. I think by following a few tips from The Force, I could easily double the mpg. But that is for my situation and the relatively short commutes I'm making in the car. For a longer trips, no matter what, it would need the gasoline engine. There is no battery pack that would give 200 or 300 miles of range, and can be recharged in less than an hour while you stop for lunch. And by removing the gasoline engine, you cut that aspect of the Prius out of the equation and thus require another car. This is essentially what pEEf said too.

But actually, I did speak a bit too soon, and there is one possibility when such an idea would be possible. If there was an otherwise pristine Prius with a blown engine, then such a project might make sense. But still, I don't think you'll be saving 300 lbs. because you've got to leave the engine block and the crankshaft. Just running the tank dry saves about 88 lbs, so you're more than 1/4th the way there just doing that and nothing else, which is MUCH easier. And, at some point, you can return to as manufactured condition just by putting gasoline in the tank and clearing the codes. And no matter what, with such a project you will never have a "normal working" Prius. The video showed that during the first minute or so.

 

Yes, as the Ewerts have shown, the car will do 70 on a flat surface even with the transaxle and ICE spin losses. That's pretty much the upper limit of the boost converter, as it's going to be pulling about 100 amps to give you the ~20kw. Now if your ICE is warmed up, there's little wrong with using some of it's power to get you to that speed. That way you get up to speed in a reasonable amount of time without pissing everyone on the road off and/or creating a safety hazard.

You can calculate it. It's pulling around 20kw from the batteries. If you have a 20kwh usable capacity pack, you can't expect more than an hour or 60ish miles. 5kwh? About 15 minutes/15 miles. I'm putting in a 7kwh (usable) pack which I believe is a good compromise. It weighs 170 pounds with all it's goo installed, but I'll be removing the stock pack, so that's a net gain of around 100lbs.

Thanks for the Compliments!

The PHEV modifications for me will cover most of my in-town needs. I probably won't have to visit the pumps even as often as once a month! But I like to take trips, and then I don't need a separate car.

Kind of OT, But I've made a piece of plywood that goes in the back to span the gap when you fold down the seats. Slide the front seats all the way forward and drop the plywood in. Then unroll a foam mattress and I've got some of that mylar-coated bubble wrap cut to fit all the windows. Presto! Mini camp-anywhere Prius for 2! If it's hot outside I can even leave the car in ready mode and run the A/C. The PHEV pack can run the A/C for about 10 hours if it's fully charged. This is amazingly comfortable and works well. I don't need to bother with Hotels. If I can find a place to plug in I can even run my A/C indefinitely from a 120v outlet!

I need to make a hack which will let me run the electric heating strips without causing the Prius to start the ICE, then I can have a little heat as well! I suppose I could also just order a small EU-spec space heater (240v) and modify it to run on DC pack voltage.

You don't need the crank. Steve Woodruff's "conversion" (the one in the video) left the empty aluminum engine block in black mainly as a mounting support. It looks like it only weighs about 34lbs:



I bet you could either fabricate a new mount or at least cut away most of the metal in the block if you wanted to go that route.

Then go into the transaxle and remove MG1, the PSD, etc. Leave only MG2 as it's really all you can use anyway. That would give you a 20kw EV.

To summarize: There is no amount of mechanical fiddling with the PSD, MG2/MG1, gear ratios, etc. that will give you more than 20kw power unless you do a lot of electrical work to somehow deliver more than 40A @ 500V to the transaxle!

If you were to hack the inverter and improve the boost converter, that will give you 50kw. (you could install a 500v pack instead and bypass it) 50kw would be a tolerable EV, and it would get you to highway speeds. That's probably the best you could hope for keeping the Prius transaxle.

If someone wants to send me a wrecked Prius, I'll be happy to figure out all the technical details and (probably need to) come up with a replacement for the HV ECU to make it work well. If you beefed up the boost converter or added a 500v pack, you'd have no choice.

 

Hi All,

I did not take the time to read through all this. But, it seems, the way to do this is to get three standard prius packs, put them in series, and feed that to a third-party three phase VFD (Variable Frequency Drive)....hmmm three stock Gen 1 inverters in series too?

 

Several problems, one is 3 Prius packs will give you about 3kwh of usable energy. That's enough to only go about 3 miles on the highway!

Secondly, you are going to have considerable trouble locating an affordable 50kw (67hp) VFD that can handle a BLDC-type motor. Pretty much all industrial VFD's are built to run AC induction motors. While the drive algorithms are similar, they will need to be different to run the Prius BLDC motor properly. They also don't build big VFD's like that to be either small or light-weight.

You'd be better off building a 500v Thundersky Lithium pack. If you use the 40ah blocks, that will get you about 20kwh which is more reasonable for a small EV. Then, just run this into the boost bus inside the stock inverter, and buy a motor control development kit from any of the big-name microcontroller manufacturers and connect it to the Prius inverter. You'll have to also interface the encoder, Accel pedal, and do some amount of hacking on the brake system, along with some decent amount of software work. But then you should end up with a decent small EV!

 

Bottom line is, unless you happen to have old Prii laying around like Steve Woodruff, you'd be better off selling your Prius and buying a Nissan Leaf.

You'll be much more happy! (and probably end up spending less!)

There are just too many little details and stumbling blocks here. Has anyone asked Steve to see what became of his Prius EV project?!

 

From what I recall, It never came to be

There were too many problems with the BMS that they were using(a prototype made by Plug-In Supply and the Ewerts)

The car was eventually sold I think...

 

I think you'd be far better off converting a Geo Metro into an EV, the Prius is so complicated to begin with....