Transaxle Modifications

[Jack Bauer]

Please understand I'm not trying to give anyone a hard time. The toyota stuff is very promising. Examples being the gs450h and prius gen 3 inverter both of which I am comfortable selling control boards for as I have proven them first hand in my own cars. I just wish someone, anyone , would forget about perfection and just throw something together to prove the viability before we get tied up with bench tuning motors etc. Again, please don't missunderstand I'm just trying to give of my experience both past and present of things that "everyone" said wouldnt work that subsequently did and vice versa.

[RetroZero]

Yup, my case is a little more complicated than the rest of you. Very tough rules, so throwing something together is just what authorities are waiting to see and make a lesson of it.. Aside the gear lock up, it's keep it simple and get something rolling.

[Kevin Sharpe]

Yup, my case is a little more complicated than the rest of you. Very tough rules, so throwing something together is just what authorities are waiting to see and make a lesson of it..

You misunderstand... we are suggesting a test car that you drive around a field or track not on the road... AFAIK the modified transaxle is unproven and the concern is that people are spending a lot of time and money without any idea whether something basic has been overlooked.

[RetroZero]

Gotcha. Better find me 4 wheels and a steering wheel then.

[GaryClarke]

I know you've put a lot of work into the fix for the gen 2 tranaxle so that it will pass the strict rules in France for modifying OEM components. but can I ask why you've not considered the Gen 3?

Unless I am missing something important, the Gen 3 only needs an external modification to mobilise the ICE input shaft. It is designed and therefore already approved to run in EV mode up to 61mph using MG2 alone. There is potential to boost further with MG1. T

Elsewhere on the site, there's a lot of work being put into the MGR unit but on paper with internal cooling the Gen 3 unit seems to need the least hacking for use as EV propulsion unit.

What is it that I'm missing do you think?

[RetroZero]

Hi Gary, at this stage it's the cost of Gen 2 as apposed to Gen 3, along with availability.
I would like to start with "most readilly available" as I plan on converting 20 vehicles with exact same "kit".
Then I can start to upgrade and look at other options...again, with 20 kits put together (to absorb homologation costs for each model)
Already found 3 Gen 2 Prius locally and untouched, with no one to take care of them. I believe they are waiting for buyers from certain East European Countries to come and make them an offer, if not, they just sit there......Then we find them on internet for stupid prices and pay to get them transported back here.....

[Kevin Sharpe]

Unless I am missing something important, the Gen 3 only needs an external modification to mobilise the ICE input shaft. It is designed and therefore already approved to run in EV mode up to 61mph using MG2 alone. There is potential to boost further with MG1. T

In this video Professor Kelly suggests that the Gen 3 transaxle will run much faster than 61mph (he speculates the speed restriction is due to battery limitations not MG1/MG2). If I understand correctly you have the option of an external lock on the ICE input shaft or even a third electric motor

[Jack Bauer]

The gen 3 transaxle requires welding of the power split device as like the gen 2 the oil pump is mechanical and driven by the ice input shaft.

[Kevin Sharpe]

The gen 3 transaxle requires welding of the power split device as like the gen 2 the oil pump is mechanical and driven by the ice input shaft.

Any reason we couldn't run a small motor on the ICE input shaft to keep the oil pump happy? Keeping the transaxle sealed has a lot of advantages especially if we're converting Gen 3 Prius to EV

[ZooKeeper]

What has me confused is that unless I missed something, no one would ever know the power-split was welded, except whomever did it, so why and how would that matter? It is the most cost-effective method to achieve the goal.

[Kevin Sharpe]

The gen 3 transaxle requires welding of the power split device as like the gen 2 the oil pump is mechanical and driven by the ice input shaft.

Any reason we couldn't run a small motor on the ICE input shaft to keep the oil pump happy? Keeping the transaxle sealed has a lot of advantages especially if we're converting Gen 3 Prius to EV

A related question... the Gen 4 (P610) transaxle has a similar mechanical oil pump that is driven by the ICE but the car also has extended EV driving modes... does the transaxle rely on an external oil pump when in EV mode and if so, why do we need the mechanical pump?

[Jack Bauer]

Only the Prius plug in has the external oil pump on the p610 as it has the ability to reach motorway speed in ev mode.

[Matthew100]

The gen 3 transaxle requires welding of the power split device as like the gen 2 the oil pump is mechanical and driven by the ice input shaft.

So IF the gen2 ICE was restricted from starting and driven around in just EV mode for extended periods of time harm could occur to the system?

I was looking at this one example where modifying CAN bus messages prevented the ICE from warming up, assisting or contributing any power to the system. Driven in EV mode up to 70mph /112kmph.

"You may be wondering how did I get this EV mode with no speed limit? I had to totally disconnect the engine ECM from the CAN bus and run it on it's own second CAN bus from my BMS board. The Hybrid (HV) ECU normally asks the Engine ECM for ICE power over the CAN bus, but basically if you prevent these CAN frames from going in, and instead send different frames with a power request of zero, then the Engine will not use fuel. So in hybrid mode I just send the frames as-is, but in EV mode I manipulate the frames to request no ICE power and keep the fuel cut. The HV ECU can then still spin the ICE as needed to lube and protect the transaxle and MG's, but use no fuel, this gives you the best PHEV use scenario, and you are still protecting your valuable hardware under the hood!"

He unfortunately didn't figure out how to get the inverter to output more than the 20kw, so it was very slow at the top end. IF only there was a large group of inverter enthusiasts around right ?

Would a modification like this need an electric oil pump? If the ICE doesn't warm up and doesn't spin until it is needed to prevent over speed of MG1.

Here is a link outlining all the details of how it was done https://priuschat.com/threads/update-an ... ect.86836/

[sfk]

He unfortunately didn't figure out how to get the inverter to output more than the 20kw, so it was very slow at the top end. IF only there was a large group of inverter enthusiasts around right ?

I thought 20kw was a limit of the buck/boost hardware itself. "Mainlining" the battery pack directly into the busbars of the inverter is the way to avoid this, but not in software alone.

[Isaac96]

He unfortunately didn't figure out how to get the inverter to output more than the 20kw, so it was very slow at the top end. IF only there was a large group of inverter enthusiasts around right ?

I thought 20kw was a limit of the buck/boost hardware itself. "Mainlining" the battery pack directly into the busbars of the inverter is the way to avoid this, but not in software alone.

Right, but seems like the person who did that was using a mostly stock vehicle, minus a couple of CAN hacks. So the boost converter would need to stay.

[RetroZero]

After watching Damiens video on set-up with Gen3 transaxle. There was the point that when the 2 MG's were connected, MG1 would fight rotation (permanent magnets) if not excited at the same time.....which I think is where the whole ICE comes in.
If buck/boost is limited to 20 kW (100A/200V), the rest of the amps "arrive" via that generated by MG1.(kind of like running MG1 in regen mode, but the ICE helping to rotate it). Like a huge alternator.
The inverter then redirects the generated kW to MG2 (I assume this is where the 400V generated by MG1 would come in handy)
That's my 2 cents of how it works (I am by no means an expert)
IF this is the case, increasing Battery voltage will enable more KW. BUT, it needs to arrive after the buck/boost converter (limited to 100A/200V)
Now I ask myself, what would happen with 300V/150A battery pack set-up? (If my assumptions are correct) Thats 50Kw in my calculations.

[Matthew100]

I went out to the prius plugged in techstream to get all the live data flowing. Confirmed what I believed to be the case that the engine doesn't rotate when in ev mode. Of course it wouldn't otherwise mg1 would never be able overspeed and we wouldn't have those ev mode speed limits imposed on the system. Double checked, and can move forward.
I haven't taken the transaxle apart, I haven't seen inside and so I don't understand how oil is pumped when the ICE is off. Ordinarly on start up the ICE will go into a warm up period, once it is up to temperature it shuts down. With the OEM HV battery you go a km or 2 in ev mode and the ICE will turn back on. Charge batteries and go again. With the larger battery packs you could go significantly longer without needing to recharge and fire up the ice. If the oil wasn't getting pumped in EV mode damage or excessive wear and tear would occur, but I haven't heard of that happening and so I am lead to believe it must be getting pumped somehow.

In regards to the inverter power output. I see the HV battery BMS communicates limits on the charge and discharge rate. Depending on temperature of the battery and the SOC. Those limits are usually 25kw charge / 21kw discharge when it is happiest. With a new programable BMS and a suitable battery, lifting the charge and discharge limit to example 50kw each. As well as bypassing the buck/ boost converter in the inverter. What else would be required? Does the inverter need to be told to use more power from the hybrid main brain? or is it self limiting it self?

[RetroZero]

Good to know about 25kw charge/21kw discharge values.

The oil pump in transaxle rotates when the Input shaft from ICE rotates. Check the notch at the end of the shaft in previous photo I posted, it drives the pump.
If you lock up the power split device, the pump rotates when there is wheel rotation. I see 4 options:
1 - lock up PSD (MG1 turns with MG2 when in EV drive mode), but can't have any ICE present.
2 - Enable MG1 to rotate permanently with MG2 ( Dual motor mode), but that means more kw needed to power 2 motors.
3 - Cut off end of Input shaft that connects to oil pump, and make up adapter to have it permanently connected to MG2. The most complicated....and expensive.
4 - add external oil pump.
From what I learned from Damien, keep it simple. I have a fancy lock up device, all assembled and running on the bench, but not yet tested in real life mode.
If I tell you how many months it took to get that done here where I live, along with the price tag (hint - could have bought another Inverter & transaxle) and have no concrete proof, it's for nothing. So I am going to try to get this up and running without more mods besides 2nd adapter board for MG1, and then work out how to get more than 20 kw. BUT, my set up is full electric, so option 1 is simplest and cheapest.

[vin]

Please understand I'm not trying to give anyone a hard time. The toyota stuff is very promising. Examples being the gs450h and prius gen 3 inverter both of which I am comfortable selling control boards for as I have proven them first hand in my own cars. I just wish someone, anyone , would forget about perfection and just throw something together to prove the viability before we get tied up with bench tuning motors etc. Again, please don't missunderstand I'm just trying to give of my experience both past and present of things that "everyone" said wouldnt work that subsequently did and vice versa.

Hi all, when setting the parameters on the MG2 of the Prius Gen 3 Transaxle, I've managed to figure out the resolver exciter pins, however I seem to be getting the same plot results from all 4 of the other pins on the resolver. The results are very similar and all appear to be giving decent reading according to Damien's FOC tuning video. This is a photo of the pinout, pairs, and resistance values. Has anyone got this working successfully or have the pinout? Also are there some parameters that I can compare mine to please? I'm using a Prius Gen2 inverter and a Prius Gen 3 transaxle.

Thanks in advance

[vin]

Please understand I'm not trying to give anyone a hard time. The toyota stuff is very promising. Examples being the gs450h and prius gen 3 inverter both of which I am comfortable selling control boards for as I have proven them first hand in my own cars. I just wish someone, anyone , would forget about perfection and just throw something together to prove the viability before we get tied up with bench tuning motors etc. Again, please don't missunderstand I'm just trying to give of my experience both past and present of things that "everyone" said wouldnt work that subsequently did and vice versa.

Hi all, when setting the parameters on the MG2 of the Prius Gen 3 Transaxle, I've managed to figure out the resolver exciter pins, however I seem to be getting the same plot results from all 4 of the other pins on the resolver. The results are very similar and all appear to be giving decent reading according to Damien's FOC tuning video. This is a photo of the pinout, pairs, and resistance values. Has anyone got this working successfully or have the pinout? Also are there some parameters that I can compare mine to please? I'm using a Prius Gen2 inverter and a Prius Gen 3 transaxle.

Thanks in advance IMG_6981.jpg

It turns out that on the picture I've uploaded

Pin 1 = Sign
Pin 2 = Co Sign
Pin 3 and 4 = Exciter
Pin 5 and 6 = Ground

Thanks

[GaryClarke]

The latest teardown video from Weber University.

It seems that MG1 is designed to over speed to provide extra overdrive. MG1 takes power from the HV battery to spin faster than the ICE to add mechanical drive.

For EV conversion purposes, I'm taking this to mean that there is no need to weld the PSD. At most the ICE driveshaft may need clamping. The added benefit is that the oil pump will operate.

[lsh3rd]

With the welded modification made so MG1 and MG2 turn in unison, what are the thoughts on coupling the protruding MG1 splined shaft with the transmission for a RWD vehicle?

If it's feasible then could remove the running gears to reduce weight. Forgive me if this was the plan all along and I'm just stating the obvious!

Replying to a very old comment after watching some of Weber State's videos. This is a very interesting idea since you could potentially very easily make a coupler using the Prius's stock clutch disc and whatever clutch disk you had for the RWD vehicle. The quality of the welding on the sun gears to carrier would need to be very good since all of MG2's torque would be routed back through that gearset.

However, I don't believe it is feasible to remove the running gears for one simple reason. In addition to the very small trochoid pump, these gears sling transmission fluid throughout the case and provide cooling. I think the ring gear above all seems to be responsible for this. You probably should remove the spider gears and S spring, however!