Some technical NX info I have learned.
#32
I think I finally thought of a way to 'easily' explain the eCVT function... using a visual aid that can be used to show every eCVT condition with actual rpm and direction of engine, MG1, and vehicle speed/MG2. (The concept I was thinking of before was mostly analogous; but, not always.) The WeberAuto video is great if you can wrap your head around all the relationships; but, I don't think I fully understood until I thought through this visualization.
Can anyone tell me how I could (easily) make a nomogram?
It will just be three parallel scales... no curves. At this point, I might be able to make something good enough in the form of an Excel chart with primary, secondary, and tertiary Y-axes... it might not be very pretty. It might come out better if I just drew it in Photoshop or something... unless someone knows an easier way.
Sad story: I thought today would be the day I would collect, hopefully all, the data I need; but, the Mrs decided to go somewhere after work and I had to leave for my thing before she got back.
Can anyone tell me how I could (easily) make a nomogram?
It will just be three parallel scales... no curves. At this point, I might be able to make something good enough in the form of an Excel chart with primary, secondary, and tertiary Y-axes... it might not be very pretty. It might come out better if I just drew it in Photoshop or something... unless someone knows an easier way.
Sad story: I thought today would be the day I would collect, hopefully all, the data I need; but, the Mrs decided to go somewhere after work and I had to leave for my thing before she got back.
Last edited by CdO; 05-05-23 at 12:39 AM.
#33
Can anyone tell me how I could (easily) make a nomogram?
I'm pretty sure it will just be three parallel scales... no curves. At this point, I might be able to make something good enough in the form of an Excel chart with primary, secondary, and tertiary Y-axes... it might not be very pretty. It might come out better if I just drew it in Photoshop or something... unless someone knows an easier way.
I'm pretty sure it will just be three parallel scales... no curves. At this point, I might be able to make something good enough in the form of an Excel chart with primary, secondary, and tertiary Y-axes... it might not be very pretty. It might come out better if I just drew it in Photoshop or something... unless someone knows an easier way.
By playing with the ICE and MG2 rpms, you can see the effect on MG1 and also understand why (on non plug ins) hybrids can only go up to ~ 42mph with the ICE at 0rpm.
https://archive.org/details/Toyota_P...r_Split_Device
Last edited by Droid13; 05-04-23 at 05:47 AM. Reason: More detail
#34
Nice. Thanks, I hadn't come across anything like this before... which I thought was weird...
Me: "why hasn't anyone done this before?" Well, good to know my thinking was correct... they totally stole my idea (15 years before I thought of it). Guess I need to work on my googling skills.
Yeah, that one is old school Prius, I guess. The gear ratios are off and the rpm are quite limited; so, while the principles are the same, the numbers are not applicable to the NX (e.g. the NX motors can go up to 17,000 rpm... at 124 mph... MG2 specifically; MG1 spins about 20% slower in EV, MGR ~4% <MG2... always). So far, I don't see a reason for the 84 mph 450h+ EV limit. AFAIK, the powertrains of NX HEVs and PHEVs (and other Toyota?) are effectively identical... except for the size of their batteries (i.e. power output). The 350h will shut off the engine going down a hill / decelerating at highway speeds, no?
Turns out Excel doesn't really do tertiary y-axes... you can sort of kluge it; but, absent an actual nomogram making application, I think I'll just make a template image and modify it for various conditions to discuss. No fancy animations... yet... probably never.
Last edited by CdO; 05-05-23 at 01:50 AM.
The following users liked this post:
clmk (05-12-23)
#35
Yeah, that one is old school Prius, I guess. The gear ratios are off and the rpm are quite limited; so, while the principles are the same, the numbers are not applicable to the NX (e.g. the NX motors can go up to 17,000 rpm... at 124 mph... MG2 specifically; MG1 spins about 20% slower in EV, MGR ~4% <MG2... always). So far, I don't see a reason for the 84 mph 450h+ EV limit. AFAIK, the powertrains of NX HEVs and PHEVs (and other Toyota?) are effectively identical... except for the size of their batteries (i.e. power output). The 350h will shut off the engine going down a hill / decelerating at highway speeds, no?
.
#36
Can anyone find (or knows) all the gear teeth counts in the NX P810 (eCVT)?
I was preparing some calculations and quickly realized that there are differences to a few gears (4?) compared to the P710 in the Weber video... he made it sound like it just had a bigger motor and maybe just a tweaked differential. I haven't been able to find any good info.
I was able to find the numbers for the rear.. the Q610. It is listed at a 10.781 ratio. That might seem odd to you. It's really a 3.944 (71/18) differential with an additional reduction of 2.733 (41/15).
I found it strange that they list the rear like that, but list the front as 3.412.I guess that has to make it 58/17 (instead of 78/20). I was able to find some numbers that seem to work with that (approximately match my limited/low speed driving data); but, I'd like to know for sure... I don't really need to know.
I was all set up to get some driving data last week; but, when I went to top off my battery I found that they changed the charger hours. I didn't want to proceed with limited battery power... it's always something.
I was able to find the numbers for the rear.. the Q610. It is listed at a 10.781 ratio. That might seem odd to you. It's really a 3.944 (71/18) differential with an additional reduction of 2.733 (41/15).
I found it strange that they list the rear like that, but list the front as 3.412.
I was all set up to get some driving data last week; but, when I went to top off my battery I found that they changed the charger hours. I didn't want to proceed with limited battery power... it's always something.
Last edited by CdO; 06-05-23 at 12:11 AM.
#37
Nice. Thanks, I hadn't come across anything like this before... which I thought was weird...
Me: "why hasn't anyone done this before?" Well, good to know my thinking was correct... they totally stole my idea (15 years before I thought of it). Guess I need to work on my googling skills.
Yeah, that one is old school Prius, I guess. The gear ratios are off and the rpm are quite limited; so, while the principles are the same, the numbers are not applicable to the NX (e.g. the NX motors can go up to 17,000 rpm... at 124 mph... MG2 specifically; MG1 spins about 20% slower in EV, MGR ~4% <MG2... always). So far, I don't see a reason for the 84 mph 450h+ EV limit. AFAIK, the powertrains of NX HEVs and PHEVs (and other Toyota?) are effectively identical... except for the size of their batteries (i.e. power output). The 350h will shut off the engine going down a hill / decelerating at highway speeds, no?
Turns out Excel doesn't really do tertiary y-axes... you can sort of kluge it; but, absent an actual nomogram making application, I think I'll just make a template image and modify it for various conditions to discuss. No fancy animations... yet... probably never.
Yea, I'm not really sure this latest gen handles ratios in the eCVT for say the regular NX hybrid. My 2007 Camry Hybrid could only do 42mph in EV, and even though in my RX hybrid the max MG1 rpms were supposed to be upwards of 10,000rpm as opposed to the older Prius depicted on this web site, it also still was limited to 42mph without the engine spinning.
#38
Can anyone find (or knows) all the gear teeth counts in the NX P810 (eCVT)?
I was preparing some calculations and quickly realized that there are differences to a few gears (4?) compared to the P710 in the Weber video... he made it sound like it just had a bigger motor and maybe just a tweaked differential. I haven't been able to find any good info.
I was able to find the numbers for the rear.. the Q610. It is listed at a 10.781 ratio. That might seem odd to you. It's really a 3.944 (71/18) differential with an additional reduction of 2.733 (41/15).
I found it strange that they list the rear like that, but list the front as 3.412. I guess that has to make it 58/17 (instead of 78/20). I was able to find some numbers that seem to work with that (approximately match my limited/low speed driving data); but, I'd like to know for sure... I don't really need to know.
I was all set up to get some driving data last week; but, when I went to top off my battery I found that they changed the charger hours. I didn't want to proceed with limited battery power... it's always something.
I was preparing some calculations and quickly realized that there are differences to a few gears (4?) compared to the P710 in the Weber video... he made it sound like it just had a bigger motor and maybe just a tweaked differential. I haven't been able to find any good info.
I was able to find the numbers for the rear.. the Q610. It is listed at a 10.781 ratio. That might seem odd to you. It's really a 3.944 (71/18) differential with an additional reduction of 2.733 (41/15).
I found it strange that they list the rear like that, but list the front as 3.412. I guess that has to make it 58/17 (instead of 78/20). I was able to find some numbers that seem to work with that (approximately match my limited/low speed driving data); but, I'd like to know for sure... I don't really need to know.
I was all set up to get some driving data last week; but, when I went to top off my battery I found that they changed the charger hours. I didn't want to proceed with limited battery power... it's always something.
Wouldn't the ratio change if the sun gear spins in the the direction of axle vs. reverse when the Electric motor forces it the other way?
#39
Right now, I'm trying to make sure I have all the other gears sorted out. Here's how they go in the P710:
MG2------Planetary (30/23/78)
16----------53
------49
------20
------78
They list the RAV4 P710 as 3.605 final drive ratio (front) which is 49/53*78/20.
NX hybrids are listed as 3.412 (from Lexus Japan; USA doesn't say).
Last edited by CdO; 05-25-23 at 03:14 AM.
#40
I wasn't fully behind the 58/17 final drive ratio. It seems for some reason, Toyota (and others?) always round gear ratios down and you have to round 58/17 up to get 3.412 (3.4117...). Since there are no two gears with a reasonable number of teeth that round down to that ratio, I figured it had to be the ratio of four gears. With some better data and some more trial and error, I am pretty confident I have found the correct answer.
NX P810
MG2------Planetary (30/23/78)
17----------56
------49
------20 (19 RX P810)
------78 (79 RX P810)
So, 56/49*20/78 = 3.4125 (exactly, and rounds down to the listed 3.412).
For the RX = 56/49*19/79 = 3.6381579... (listed at 3.638).
Bonus: I found a Chinese article that supposedly has a cut away of the P810 and it looks like the counter drive gear does indeed have 56 teeth... maybe the only gear you could accurately count.
Now that I have that all sorted out, I think the best measure of speed is MG2 rpm / X. Techstream has six different vehicle speed measurements. One is "Vehicle Speed"... it's pretty useless... if you accelerate quickly it shows something like 0, 6, 11, 17 etc. There is one called SP1 Speed (?)... it's pretty good; but, is a bit lower (?) than the others which are FL, FR, RL, RR Wheel Speed. Those are all pretty good; but, the fronts and rears are a little different and they seem to have a slight a delay (and won't be the same if you are turning). As these cars are available with different wheels/tires and there may be some weird speedo laws, I think using MG2 rpm will be most accurate. (IIRC, Tesla had some speedo related recall. Musk complained that they ran afoul of one stupid law by trying to comply with a different stupid law.)
NX P810
MG2------Planetary (30/23/78)
17----------56
------49
------20 (19 RX P810)
------78 (79 RX P810)
So, 56/49*20/78 = 3.4125 (exactly, and rounds down to the listed 3.412).
For the RX = 56/49*19/79 = 3.6381579... (listed at 3.638).
Bonus: I found a Chinese article that supposedly has a cut away of the P810 and it looks like the counter drive gear does indeed have 56 teeth... maybe the only gear you could accurately count.
Now that I have that all sorted out, I think the best measure of speed is MG2 rpm / X. Techstream has six different vehicle speed measurements. One is "Vehicle Speed"... it's pretty useless... if you accelerate quickly it shows something like 0, 6, 11, 17 etc. There is one called SP1 Speed (?)... it's pretty good; but, is a bit lower (?) than the others which are FL, FR, RL, RR Wheel Speed. Those are all pretty good; but, the fronts and rears are a little different and they seem to have a slight a delay (and won't be the same if you are turning). As these cars are available with different wheels/tires and there may be some weird speedo laws, I think using MG2 rpm will be most accurate. (IIRC, Tesla had some speedo related recall. Musk complained that they ran afoul of one stupid law by trying to comply with a different stupid law.)
Last edited by CdO; 06-05-23 at 12:14 AM. Reason: MG2, not MG1.
#41
EV Acceleration
Finally, sorting through some data. The EV data looks fine; but, the Hybrid (total system power) data is suspect. I did two accelerations for both modes. Both hybrid accelerations have a similar drop in engine power/torque over a certain rpm range. On one acceleration, I thought I felt a little wheel slip... no traction light, and no indication in the motor rpm or wheel speed data... so, I don't know yet if it will always do that or if I need to find a stickier surface.
Anyway, here's a graph of EV acceleration (in mph). It reached 60 at about 9.8 seconds. Car mags seem to subtract the time it takes to go one foot... how long does that take? IDK. Before Car and Driver got their VBox, they used to approximate this by starting at 3 mph. That would knock off about 0.5 seconds!
Notice, the highest acceleration occurs from ~5-15 mph. This is where the motors (MG2 and MGR) have the most torque (and the battery reaches its output limit at ~17 mph). The data rate is 21 Hz. The battery started at 81.96% and ended at 78.03% (actual SoC... ~94 to 91 App%). The second run was just a bit slower... I forgot to turn off the A/C.
I used MG2 rpm to calculate the speed using the exact gear ratios above and a tire circumference equal to 709 revolutions per mile. Tire manufacturers sometimes give the revs/mile... they don't for my tires; but, they do for a similar Bridgestone Alenza 235/50R20. The Pirellis I saw were also 709. Michelins were 710 (wee bit smaller). There's a 690 Goodyear (almost 3% bigger for the same 235/50R20)! 699 Khumo.
Anyway, here's a graph of EV acceleration (in mph). It reached 60 at about 9.8 seconds. Car mags seem to subtract the time it takes to go one foot... how long does that take? IDK. Before Car and Driver got their VBox, they used to approximate this by starting at 3 mph. That would knock off about 0.5 seconds!
Notice, the highest acceleration occurs from ~5-15 mph. This is where the motors (MG2 and MGR) have the most torque (and the battery reaches its output limit at ~17 mph). The data rate is 21 Hz. The battery started at 81.96% and ended at 78.03% (actual SoC... ~94 to 91 App%). The second run was just a bit slower... I forgot to turn off the A/C.
I used MG2 rpm to calculate the speed using the exact gear ratios above and a tire circumference equal to 709 revolutions per mile. Tire manufacturers sometimes give the revs/mile... they don't for my tires; but, they do for a similar Bridgestone Alenza 235/50R20. The Pirellis I saw were also 709. Michelins were 710 (wee bit smaller). There's a 690 Goodyear (almost 3% bigger for the same 235/50R20)! 699 Khumo.
Last edited by CdO; 06-05-23 at 12:38 AM. Reason: MG2/Tires
The following 4 users liked this post by CdO:
#44
General update:
As I mentioned above, I need to gather more data to sort out a couple of things. I understand the mechanical limitations of the system and the battery power limitations. I thought that would be the end of it; but, I seem to have discovered another limitation (possibly two). It may just be a software limit (so it's not faster than their sportier cars), a hardware limit, or a combination.
Getting a good complete data set is tricky. You have to know what you are looking for and which ECU can give it to you. If you log everything in the 'Hybrid Control' ECU (415 PIDs) it only updates once every ~6 seconds. You can log two ECUs at the same time; but, the data rate becomes even slower. I seem to get a max rate at about <20 PIDs... logging multiple PIDs for each of the Engine, MG1, MG2, MGR, Battery, etc. add up pretty quickly.
Also, I may have to draw this out a couple more months as we have a vacation coming up and I have made exactly zero plans. Got my new passport, though!
The following users liked this post:
lsft (06-07-23)
#45
Related to the MG1 and the simulation above...
This weekend I was doing some long distance driving. No EV range left and driving in HV at speeds ~120km/hour.
At those speeds, I would expect MG1 to be spinning backward to be providing overdrive gearing. But, where does that energy come from? I actually saw my battery go down to 1 bar which I have never seen before... So is it pulling energy from the spinning MG2 or MGR to feed power to MG1?
This weekend I was doing some long distance driving. No EV range left and driving in HV at speeds ~120km/hour.
At those speeds, I would expect MG1 to be spinning backward to be providing overdrive gearing. But, where does that energy come from? I actually saw my battery go down to 1 bar which I have never seen before... So is it pulling energy from the spinning MG2 or MGR to feed power to MG1?