GS450H gearbox with Camry inverter G9200-33171 [SOLVED]

[PetersonOctavius]

Thanks johu!!
I ll google to see if i can find a diagram.

How do i get my bms to balance?

[Bigpie]

Zombieverter also has 2 digipots. I use them for my fuel gauge in the BMW. There's 2 params, DigiPot1Steps and DigiPot2Steps that set the 'steps' of each of the 10k pots, I can get about 5% resolution using them.

[PetersonOctavius]

thanks Jamie... i will try to use them

on the webinterface i can only set a fixed value i see.
how do i make it change based on SOC?

[PetersonOctavius]

i managed to get total of roughly 118km or 74miles on one charge.
max speeds were around 80-90kmh

with regen enabled on the vcu, the car becomes extremely jerky, sometimes even ignores brake input. so i left all regen settings at 0.
my gear shifting method is push buttons, so i have to buttons, press one car goes forward, press the other one car goes in reverse, press both car goes neutral.
so what i normally do to get some "regen"....is while coating, is press the reverse button, then when i apply throttle it regens based on the amount of throttle i press.
so far im getting used to it.

i think i can round my range down to 100km per charge as when i add a/c to the car there will be more load on my pack.
i will charge and run 1 or 2 more range tests to confirm the range.

on another note, the vcu is pretty serious on the low voltage safety, i had my lower limit at 320v, around 327v-325v it felt like what moder ev's call limp mode. at 320v torque request was 0. amazing work developers !!

[PetersonOctavius]

the went through inspection fine, soo finally road legal.
on mothersday i took my wife and mom and the baby on a date, after some 5minutes driving, i gs450h lost torque, MG2 was spinning freely.
oil pump was not pumping. right at an intersection. upon restart, the car got going for a few meters and then the same thing.
i managed to drive back home, but i dont know what the problem is. later the same day, i rechecked all my connection, all were fine, no loose contact (as far as i could see and feel). went for another drive with the wife....after 10kms...same thing...drove back home, checked the oil controller temp, warm, i could hold my hand on it for as long as i wanted, so it wasnt burning hot. the oil pump on the other hand was burning hot.

has this happened to any gs450h users before? " oil pump stops pumping, restart the car, oil pump does its job but i dont know for how long its going to keep pumping"

i have a tesla front drive unit, im not sure if it will fit in the rear subframe, that would be awesome, no oil pump noise, no expensive oil pump unexpected breakdown.
i think for now im going to explore the tesla sdu option.

[PetersonOctavius]

i've seen some conversation about this hydraulic pump, but no actual tests in real life.
found one for cheap one ebay, i bit the bullet.

wish me luck.
i still need to get an adapter, there is one on ebay but it doesnt fit in the budget yet.

[robertwa]

Couple of theories:
1) Pump bearings are shot. After driving some distance, the bearings heat and seize. You won't notice it until you stop driving since the internal oil pump keeps everything at required pressure. Once you stop you cannot restart again until the pump cools enough to free the bearing again
2) Inadequate oil fill causing pump cavitation. Pump runs too fast cavitating and heats up. Again, this only becomes apparent after a stop as the internal pump keeps things going. When you wait a while, enough fluid drains back to the pump to allow it to start pumping again, enough to allow the internal pump to take over.

No real world experience with either of these, just something to think of before trying new hardware.

[Gregski]

sorry to hear about your pompulator wowes, this is why I picked up a cheap GS450h and plan to tear into it and see what can be done to lock the input shaft /power split internally causing it to activate the internal mechanical oil pump, fingers crossed it can be done

[PetersonOctavius]

Hi Robert, thanks for the tips.

i did change the bearings around 1 month ago, maybe i should have put some grease around the new bearings? i still have a set of new bearings left.

what is inadequate oil fill? not enough? i put in fresh fully synthetic cvt oil when i changed the bearings, not the toyota one, i filled it till the fill plug starting pouring out.
you also mentioned the internal pump, im only using mg2, my input shaft is not locked.

[robertwa]

Are you running an external oil cooler? If so, you may have to hang it lower than the transmission during the fill process. Also, I believe the fill process calls for checking the fluid level when the transmission is up to working temperature. No idea how critical this is but something to try.

If you can replicate the situation, can you check if the pump motor is actually spinning?

[PetersonOctavius]

im not using an oil cooler,
replicate the situation where the pump stops pumping? im afraid i cant cause its unpredictable.
it happened yesterday at the KFC drive thru, car was not moving, i could hear the oil pump raspy sound went less loud, then it stopped making noise all together. i gave it some throttle, mg2 was spinning freely, no car motion.

[PetersonOctavius]

sorry to hear about your pompulator wowes, this is why I picked up a cheap GS450h and plan to tear into it and see what can be done to lock the input shaft /power split internally causing it to activate the internal mechanical oil pump, fingers crossed it can be done

another user was trying to do the same, konstantin8818
he welded the power split part, and he made a solid clutch plate thing to eliminate the the need of the external pump all together.
but no update.

i hope you can find something with your gs450h. i wanna open mine so bad right now

[Gregski]

another user was trying to do the same, konstantin8818
he welded the power split part, and he made a solid clutch plate thing to eliminate the the need of the external pump all together.
but no update.

i hope you can find something with your gs450h. i wanna open mine so bad right now

i see, well Artur has a good question he is wondering if it will work, in other words will we be able to get moving from a dead stop if there is no electrical pump circulating the transmission fluid when we start to move, since the internal oil pump won't pump without an engine if we are not moving, or will there be a short delay till the pressure builds up, hope you follow what I am sayin'

[PetersonOctavius]

i "THINK" most gs450h conversions only use the high gear of the Ravigneaux gear system in the rear end of the gearbox, the oil pressure is essential for that part (clutch plates, etc)
if we could eliminate that part, so instead of having the 2 stages, (High and Low), mechanically lock it into the high gear output, so no oil pressure would be needed, the oil would only be used for lubricating and heat transfer.

i dont if im thinking about it too simple. i m no mechanic by any means.
i downloaded the template of the high gear clutch pack that konstantin8818 uploaded in another thread. i really wanna let a shop make it and give it a try! but i only have one tranmission

[Gregski]

i "THINK" most gs450h conversions only use the high gear of the Ravigneaux gear system in the rear end of the gearbox, the oil pressure is essential for that part (clutch plates, etc)
if we could eliminate that part, so instead of having the 2 stages, (High and Low), mechanically lock it into the high gear output, so no oil pressure would be needed, the oil would only be used for lubricating and heat transfer.

i dont if im thinking about it too simple. i m no mechanic by any means.
i downloaded the template of the high gear clutch pack that konstantin8818 uploaded in another thread. i really wanna let a shop make it and give it a try! but i only have one tranmission

if that is the case than we may not even need to run the internal mechanical pump at all, and just rely on the gears splashing the trans fluid all over inside the box as if it was a manual gear box

can you share that template, and is that something Send Cut Send could make for us?

[PetersonOctavius]

i think Send cut send could do it.
find the .dxf below.
it should be either 30mm or 32mm in thickness

viewtopic.php?t=644&hilit=gs450h+discus ... rt=250#top

[tom91]

that is the case than we may not even need to run the internal mechanical pump at all, and just rely on the gears splashing the trans fluid all over inside the box as if it was a manual gear box

Uhm no! I would review the way the oil is routed and where it sprays out.

Note that there is also a lot of stacking of shafts with oil based bearings. https://toyota-club.net/files/faq/21-12 ... n.htm#L110

[PetersonOctavius]

2) Inadequate oil fill causing pump cavitation. Pump runs too fast cavitating and heats up. Again, this only becomes apparent after a stop as the internal pump keeps things going. When you wait a while, enough fluid drains back to the pump to allow it to start pumping again, enough to allow the internal pump to take over.

No real world experience with either of these, just something to think of before trying new hardware.

i was indeed running the pump at full speed i think.
on the webinterface i did put in 30% but since im using GS300H option as inverter, the vcu disregards the oil pump speed input.
i put a separate pwm generator at 1kHz and 20% duty cycle, pump is running quiet now, ive driven 20km+ now no issue

[PetersonOctavius]

my batteries can output max 200A, i can never reach more than 129A. even on full throttle
im limited to 70-80kmh, downhill i might reach 90-100kmh.
i read Zechy's thread, BMW Z3, but couldnt find the fix. Jamie sent him a bin and his 110A-120A max issue was fixed.

im only using mg2, input shaft is not locked.

[tom91]

im only using mg2, input shaft is not locked.

This is why.

[PetersonOctavius]

thanks tom,
i'll see if i can find a matching spline to lock the input shaft, as i can't find the original clutch plate

[robertwa]

There might be more to it than just using MG2 alone. Others have got up to ~300amps and higher speeds with MG2 only. Here's one:

[PetersonOctavius]

I thought so too, is it some limitation in the gs300h code?
I havent got the time to do some testing
I m thinking of plotting mg2 speed, torque and idc.

[Gregski]

I think battery pack density has something to do with it as well, and word on the street is that hybrid modules push more amps than some 100% electrics, think small packaging that requires lots of juice to propel the car

jehu just happens to comment on that here (and I also heard AEM engineers talking about that as well in one of their dyno test videos):

[PetersonOctavius]

My pack is made of samsung sdi cells, 67Ah, 3C peak
96 cells
That would be 67 x 3 = 201A
Fully charged, isa shunt max Amp report is around 120-129A, throttle all the way down

It feels more like a torque request limit