GS450H Discussion

Topics concerning the Toyota and Lexus inverter drop in boards
Post Reply
xp677
Posts: 435
Joined: Sat Jul 27, 2019 10:53 am
Location: UK
Has thanked: 1 time
Been thanked: 13 times

Re: GS450H Discussion

Post by xp677 »

I'm also using an RX8 pedal. There is an offset on the RX8 between the two values, it's 115 ticks.

My code added error checking, limp mode, etc for if one of the pedal sensors is faulty. That stuff would work on any pedal. Here's my code, you'll have to port it yourself to work with the VCU, it outputs to the variable "aps", you can probably just map it to the max torque rather than 4095 and return it if you like. I mix my pedal reading with the cruise control functions to allow acceleration during cruise, etc, so it's handled differently. Should still be usable for porting to the VCU.

The output from the pedal is in the 5V range but it must not go above 3.3V as I just used a 1K resistor in series and it never maxes out the ADC.

Code: Select all

//accelerator pedal limits
#define APSMIN 160 //160
#define APSMAX 780 //780
#define APSLOW 230 //230
#define APSHIGH 660 //715
#define APSDELTA 20 //20

int16_t aps1, aps2, apscheck, aps = 0;
uint8_t apsgood = 0;
bool limp_mode = 0;

//aps1 203-665
//aps2 318-780

void get_aps()
{
  aps1 = analogRead(pin_aps1);
  aps2 = (analogRead(pin_aps2) - 115);
  apscheck = aps1 - aps2;

  if ((apscheck >= -APSDELTA) && (apscheck <= APSDELTA) && (aps1 >= APSMIN) && (aps1 <= APSMAX) && (aps2 >= APSMIN) && (aps2 <= APSMAX)) //normal
  {
    apsgood = 3;
    aps = map(aps1, APSLOW, APSHIGH, 0, 4095);
  }
  else if ((aps1 >= APSMIN) && (aps1 <= APSMAX)) //app1 still valid, limp mode
  {
    apsgood = 1;
    limp_mode = 1;
    aps = map(aps1, APSLOW, APSHIGH, 0, 4095);
  }
  else if ((aps2 >= APSMIN) && (aps2 <= APSMAX)) //app2 still valid, limp mode
  {
    apsgood = 2;
    limp_mode = 1;
    aps = map(aps2, APSLOW, APSHIGH, 0, 4095);
  }
  else //cant use the pedal
  {
    apsgood = 0;
    aps = 0;
  }

  if (limp_mode) aps /= 4;

  if (aps < 0)aps = 0;
  if (aps > 4095)aps = 4095;

}
User avatar
Gigas
Posts: 86
Joined: Mon Nov 09, 2020 6:30 pm

Re: GS450H Discussion

Post by Gigas »

Thanks man - But I'd rather just use a pedal with no coding needed as it's a little out of my league. I'm all thumbs and welding lol

xp677 wrote: Mon Nov 30, 2020 9:54 pm I'm also using an RX8 pedal. There is an offset on the RX8 between the two values, it's 115 ticks.

My code added error checking, limp mode, etc for if one of the pedal sensors is faulty. That stuff would work on any pedal. Here's my code, you'll have to port it yourself to work with the VCU, it outputs to the variable "aps", you can probably just map it to the max torque rather than 4095 and return it if you like. I mix my pedal reading with the cruise control functions to allow acceleration during cruise, etc, so it's handled differently. Should still be usable for porting to the VCU.

The output from the pedal is in the 5V range but it must not go above 3.3V as I just used a 1K resistor in series and it never maxes out the ADC.

Code: Select all

//accelerator pedal limits
#define APSMIN 160 //160
#define APSMAX 780 //780
#define APSLOW 230 //230
#define APSHIGH 660 //715
#define APSDELTA 20 //20

int16_t aps1, aps2, apscheck, aps = 0;
uint8_t apsgood = 0;
bool limp_mode = 0;

//aps1 203-665
//aps2 318-780

void get_aps()
{
  aps1 = analogRead(pin_aps1);
  aps2 = (analogRead(pin_aps2) - 115);
  apscheck = aps1 - aps2;

  if ((apscheck >= -APSDELTA) && (apscheck <= APSDELTA) && (aps1 >= APSMIN) && (aps1 <= APSMAX) && (aps2 >= APSMIN) && (aps2 <= APSMAX)) //normal
  {
    apsgood = 3;
    aps = map(aps1, APSLOW, APSHIGH, 0, 4095);
  }
  else if ((aps1 >= APSMIN) && (aps1 <= APSMAX)) //app1 still valid, limp mode
  {
    apsgood = 1;
    limp_mode = 1;
    aps = map(aps1, APSLOW, APSHIGH, 0, 4095);
  }
  else if ((aps2 >= APSMIN) && (aps2 <= APSMAX)) //app2 still valid, limp mode
  {
    apsgood = 2;
    limp_mode = 1;
    aps = map(aps2, APSLOW, APSHIGH, 0, 4095);
  }
  else //cant use the pedal
  {
    apsgood = 0;
    aps = 0;
  }

  if (limp_mode) aps /= 4;

  if (aps < 0)aps = 0;
  if (aps > 4095)aps = 4095;

}
What was I doing?
User avatar
ggeter
Posts: 133
Joined: Wed Nov 18, 2020 1:56 pm
Location: Houston, TX
Has thanked: 4 times
Been thanked: 15 times

Re: GS450H Discussion

Post by ggeter »

Shopping for inverters to drive the 450h trans, and hard to find at decent price in the States. Let's say I'm dropping the trans in an MG Midget... Will the Camry inverter from the same generation (selling for often <$200 and plentiful) work well enough? Given the price difference, are there mods that would be worth the $400-$600 difference in price to get it to work?

-g.
xp677 wrote: Tue Sep 08, 2020 5:58 pm
TFirenza wrote: Tue Sep 08, 2020 9:00 am does anyone have experience with G9201-30010 its a 2008 RHD gs450h , its from the same period so I wouldn't expect it to be an issue but it also isn't listed on the hardware wiki.
As far as we know, all the "square" inverters from the Gen1 GS450h/Camry are the same when it comes to control.

The Camry inverters have different connectors, the LHD inverters have the ports on the other side (mirrored casing), but when it comes to using them, they are all the same.

I've ran both a 2006 and a 2010 RHD GS450h inverter, no issues.
Houston, Texas, USA
EV Newbie
1979 MG Midget + GS450h = "Mexus?"
IT Consultant
User avatar
Gigas
Posts: 86
Joined: Mon Nov 09, 2020 6:30 pm

Re: GS450H Discussion

Post by Gigas »

ggeter wrote: Thu Dec 03, 2020 12:43 am Shopping for inverters to drive the 450h trans, and hard to find at decent price in the States. Let's say I'm dropping the trans in an MG Midget... Will the Camry inverter from the same generation (selling for often <$200 and plentiful) work well enough? Given the price difference, are there mods that would be worth the $400-$600 difference in price to get it to work?

-g.
xp677 wrote: Tue Sep 08, 2020 5:58 pm
TFirenza wrote: Tue Sep 08, 2020 9:00 am does anyone have experience with G9201-30010 its a 2008 RHD gs450h , its from the same period so I wouldn't expect it to be an issue but it also isn't listed on the hardware wiki.
As far as we know, all the "square" inverters from the Gen1 GS450h/Camry are the same when it comes to control.

The Camry inverters have different connectors, the LHD inverters have the ports on the other side (mirrored casing), but when it comes to using them, they are all the same.

I've ran both a 2006 and a 2010 RHD GS450h inverter, no issues.
Check all your local junk/parts yards - I managed to find one that had hybrids and had a 450h and I got the inverter for 150$ took a few weeks but worth it
What was I doing?
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: GS450H Discussion

Post by nkiernan »

ggeter wrote: Thu Dec 03, 2020 12:43 am Shopping for inverters to drive the 450h trans, and hard to find at decent price in the States. Let's say I'm dropping the trans in an MG Midget... Will the Camry inverter from the same generation (selling for often <$200 and plentiful) work well enough? Given the price difference, are there mods that would be worth the $400-$600 difference in price to get it to work?

-g.
xp677 wrote: Tue Sep 08, 2020 5:58 pm
TFirenza wrote: Tue Sep 08, 2020 9:00 am does anyone have experience with G9201-30010 its a 2008 RHD gs450h , its from the same period so I wouldn't expect it to be an issue but it also isn't listed on the hardware wiki.
As far as we know, all the "square" inverters from the Gen1 GS450h/Camry are the same when it comes to control.

The Camry inverters have different connectors, the LHD inverters have the ports on the other side (mirrored casing), but when it comes to using them, they are all the same.

I've ran both a 2006 and a 2010 RHD GS450h inverter, no issues.
I just ran two GS450H gearboxes side by side at the weekend, one with a Camry inverter and one with a GS450H inverter. Got the same results as Arturk on the forum previously got. The GS450H inverter runs the gearbox with Damian's VCU smoothly and in both directions. However, the Camry inverter with the same setup only runs the gearbox in one direction, but in the opposite direction it stutters back and forth. Also, for the same code and input signals, the forward direction from both inverters drives the gearboxes in different directions. Interestingly, both inverters are not identical in height, the Camry is smaller in height!

So the Camry must need some modifications to the VCU code and is not directly replaceable for the GS450H inverter. It would be great to get this figured out. Wonder is there anyone on the forum that could log data from a Camry driving!???
User avatar
mdrobnak
Posts: 692
Joined: Thu Mar 05, 2020 5:08 pm
Location: Colorado, United States
Has thanked: 1 time
Been thanked: 5 times

Re: GS450H Discussion

Post by mdrobnak »

One thing to note, which I find interesting - Camry is a front-wheel-drive vehicle, with a very different drivetrain. So I'm kind of amazed they work at all - UNLESS they are all configured to have the same resolver offset.

-Matt
User avatar
ggeter
Posts: 133
Joined: Wed Nov 18, 2020 1:56 pm
Location: Houston, TX
Has thanked: 4 times
Been thanked: 15 times

Re: GS450H Discussion

Post by ggeter »

Thanks for the replies.

One thing I really need in the planning of my conversion is dimensions of the transmission and inverter. I know Damien has rough dimensions on the wiki, but any chance someone can post some pics with a tape measure next to them? For example, what is the length and width of the pan? It seems awfully large.
Houston, Texas, USA
EV Newbie
1979 MG Midget + GS450h = "Mexus?"
IT Consultant
User avatar
Gigas
Posts: 86
Joined: Mon Nov 09, 2020 6:30 pm

Re: GS450H Discussion

Post by Gigas »

ggeter wrote: Thu Dec 03, 2020 10:49 pm Thanks for the replies.

One thing I really need in the planning of my conversion is dimensions of the transmission and inverter. I know Damien has rough dimensions on the wiki, but any chance someone can post some pics with a tape measure next to them? For example, what is the length and width of the pan? It seems awfully large.
I plan to go over mine in a video on my youtube channel over the weekend , but I can do this - the inverter is tiny but the transmission motor is HUGE
What was I doing?
AdvancedEnergySys
Posts: 26
Joined: Sun Aug 30, 2020 7:16 pm
Location: United States

Re: GS450H Discussion

Post by AdvancedEnergySys »

I've just recorded a VERY rough measurement template, but I'll get more dimensions of the inverter and gs450h trans. I've also got a gen 1 and 2 camry inverter along with a RHD LS600HL inverter to compare against the 450h inverter as well.
If you can see the rough dimensions written down, take note to add an extra zero to the end as I recorded in cm and wrote down mm :?
Image
User avatar
ggeter
Posts: 133
Joined: Wed Nov 18, 2020 1:56 pm
Location: Houston, TX
Has thanked: 4 times
Been thanked: 15 times

Re: GS450H Discussion

Post by ggeter »

Awesome. Thanks!
Houston, Texas, USA
EV Newbie
1979 MG Midget + GS450h = "Mexus?"
IT Consultant
User avatar
SciroccoEV
Posts: 369
Joined: Thu Oct 10, 2019 1:50 pm
Location: Luton UK
Been thanked: 15 times

Re: GS450H Discussion

Post by SciroccoEV »

Whilst I applaud the use of millimetres as a proper engineering unit, rather than centimetres, you've still fallen into the off by an order of magnitude trap.
User avatar
ggeter
Posts: 133
Joined: Wed Nov 18, 2020 1:56 pm
Location: Houston, TX
Has thanked: 4 times
Been thanked: 15 times

Re: GS450H Discussion

Post by ggeter »

Will definitely fit in the Midget!
Houston, Texas, USA
EV Newbie
1979 MG Midget + GS450h = "Mexus?"
IT Consultant
RE3Rotor
Posts: 74
Joined: Thu Feb 06, 2020 11:37 pm
Location: Vancouver, Canada

Re: GS450H Discussion

Post by RE3Rotor »

Been MIA for a while. I just picked up a clutch disc and confirmed it fits the input shaft on the GS450h gearbox.

Toyota part no. 31250-35452
User avatar
sfk
Posts: 289
Joined: Mon Jan 14, 2019 8:29 pm
Location: Wellington, NZ
Has thanked: 2 times

Re: GS450H Discussion

Post by sfk »

Toyota part no. 31250-35452
2015 Toyota Tacoma Base, Pre Runner, TRD Pro 4.0L V6 - Gas
2014 Toyota FJ Cruiser Base 4.0L V6 - Gas
2014 Toyota Tacoma Base, Pre Runner 4.0L V6 - Gas
2013 Toyota FJ Cruiser Base 4.0L V6 - Gas
2013 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2012 Toyota FJ Cruiser Base 4.0L V6 - Gas
2012 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2011 Toyota FJ Cruiser Base 4.0L V6 - Gas
2011 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2010 Toyota FJ Cruiser Base 4.0L V6 - Gas
2010 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2009 Toyota FJ Cruiser Base 4.0L V6 - Gas
2009 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2008 Toyota FJ Cruiser Base, Trail Teams Special Edition 4.0L V6 - Gas
2008 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2007 Toyota FJ Cruiser Base, TRD Special Edition 4.0L V6 - Gas
2007 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2006 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2006 Toyota Tundra Base, Limited, SR5 4.0L V6 - Gas, 4.7L V8 - Gas
2005 Toyota Tacoma Base, Pre Runner, X-Runner 4.0L V6 - Gas
2005 Toyota Tundra Base, Limited, SR5 4.0L V6 - Gas, 4.7L V8 - Gas
-< Mazda Eunos JC Cosmo rotary -> EV conversion w/ Lexus GS450H gear >-
User avatar
sfk
Posts: 289
Joined: Mon Jan 14, 2019 8:29 pm
Location: Wellington, NZ
Has thanked: 2 times

Re: GS450H Discussion

Post by sfk »

Can someone define how torque can be transmitted directly from the ICE to the rear differential without assistance from the MG1? As far as I can see, because of the 1st planetary gearset, if you try to simply use the ICE to drive away from a standstill, the MG1 will just start spinning because this is the path of least resistance.

Is it the case that MG1 must be held stationary (effectively locking it up) by applying power to produce a torque value higher than the torque required to move the vehicle? Thus the path of least resistance would then be through the rest of the transmission and diff to the rear wheels.

The only alternative I can see is that there is never any direct connection between the ICE and the output of the transmission. The only way power is transmitted is by the the ICE spinning the MG1, MG1 generating power which is sent to MG2 which does all of the propulsion work. This contradicts the power flow display inside the car, but I can't really trust that since it only depicts 1 electric motor and shows pseudo activity that clearly isn't happening at some times.
-< Mazda Eunos JC Cosmo rotary -> EV conversion w/ Lexus GS450H gear >-
User avatar
mdrobnak
Posts: 692
Joined: Thu Mar 05, 2020 5:08 pm
Location: Colorado, United States
Has thanked: 1 time
Been thanked: 5 times

Re: GS450H Discussion

Post by mdrobnak »

I understood it as when MG1 RPM = Engine RPM -> Torque to output. By varying MG1s RPM you could change how much power is going to the axle vs being used to charge the batteries.

But I may be wrong.
User avatar
SciroccoEV
Posts: 369
Joined: Thu Oct 10, 2019 1:50 pm
Location: Luton UK
Been thanked: 15 times

Re: GS450H Discussion

Post by SciroccoEV »

sfk wrote: Thu Dec 10, 2020 9:20 pm The only alternative I can see is that there is never any direct connection between the ICE and the output of the transmission.
That's right only MG2 has a direct torque path to the driven wheels, the ICE and MG1 are just inputs to the power split device, same as the Prius.

WeberAuto on YouTube have some very good videos on the operation of the Prius transaxles.
User avatar
celeron55
Posts: 774
Joined: Thu Jul 04, 2019 3:04 pm
Location: Finland
Has thanked: 27 times
Been thanked: 110 times
Contact:

Re: GS450H Discussion

Post by celeron55 »

You can't use the ICE to directly drive away from standstill because it cannot produce torque from 0 rpm. But by gradually braking down MG1 to 0 rpm while the ICE is spinning you can start moving just like with a clutch. The power generated by MG1 when braking is directed to MG2 as extra propulsion torque and to battery for storage. If MG1 reaches zero rpm the only power going to it is inverter and winding losses, as zero rpm makes zero net power regardless of torque. It doesn't have to stop completely though as any power lost to it is regained at MG2 or the battery. In that way you can create a virtual CVT ratio.
User avatar
ggeter
Posts: 133
Joined: Wed Nov 18, 2020 1:56 pm
Location: Houston, TX
Has thanked: 4 times
Been thanked: 15 times

Re: GS450H Discussion

Post by ggeter »

Newbie regen question - I'm building a 307v pack and of course will be wiring straight into the bus per Damien's instructions.

How is regen managed? How to adjust, if possible, the voltage going back to the pack? What about amps? My pack is designed to handle over 500 amps peak.

I'm having nightmares that as long as the motor is spinning, it's making power, and since the converter is out of the picture, am I shunting all that back to the pack?

Thanks for the kind responses in advance...
Houston, Texas, USA
EV Newbie
1979 MG Midget + GS450h = "Mexus?"
IT Consultant
Isaac96
Posts: 656
Joined: Sat Oct 05, 2019 6:50 pm
Location: Northern California, USA
Been thanked: 1 time
Contact:

Re: GS450H Discussion

Post by Isaac96 »

If it can handle discharge it can handle regen. You'll be fine, Lexus engineers have done the hard work already :)
Now, voltage can theoretically go too high -- if you have a full charge and start your day with a long downhill drive. Johannes' code handles that by turning regen off, not sure what the almighty Damien VCU does. But that's almost never a concern.
User avatar
PatrcioEV-ATX
Posts: 181
Joined: Sun Dec 13, 2020 5:25 pm
Location: Austin, TX, USA
Has thanked: 40 times
Been thanked: 53 times

Re: GS450H Discussion

Post by PatrcioEV-ATX »

I’ve read through the threads but I still have a (maybe) dumb question. In pre-CAN cars, how do you tell the VCU to change between drive and reverse? I think one could use a simple actuator on the transmission shift linkage to engage/disengage park. From what I’ve read that would disable motor movement.
1998 Ford ZX2 - DC EV conversion(sold) http://evalbum.com/2093
2012 Nissan Leaf (sold)
2016 Mercedes B250e (sold)
2023 Volvo C40

Current: 1964 Rambler Classic 660 w/ GS450h set up. 36kwh Tesla batteries from B250e.
https://www.instagram.com/rambler_660e/
User avatar
celeron55
Posts: 774
Joined: Thu Jul 04, 2019 3:04 pm
Location: Finland
Has thanked: 27 times
Been thanked: 110 times
Contact:

Re: GS450H Discussion

Post by celeron55 »

The GS450H transmission has a shift actuator that mechanically engages park, and electronically outputs information about other gear positions. You probably want to grab an automatic shifter to connect it to using a rod or a mechanical cable.
User avatar
ggeter
Posts: 133
Joined: Wed Nov 18, 2020 1:56 pm
Location: Houston, TX
Has thanked: 4 times
Been thanked: 15 times

Re: GS450H Discussion

Post by ggeter »

PatrcioEV-ATX wrote: Thu Dec 31, 2020 1:38 pm I’ve read through the threads but I still have a (maybe) dumb question. In pre-CAN cars, how do you tell the VCU to change between drive and reverse? I think one could use a simple actuator on the transmission shift linkage to engage/disengage park. From what I’ve read that would disable motor movement.
I have a pre-CAN 1979 Midget...

My plan is to use the mechanical linkage to engage Park only and use a switch of some kind for Reverse and Drive. Damien's VCU takes a simple signal on one pin for Drive/Reverse. If I can figure out how to read the positions on the original selector, I can use that to then command the transmission to switch to Drive or Reverse. There is also a position past Drive for the M mode. But, my time is probably better spent rigging a solenoid to get in and out of Park, then just switches for R and D.

Bottom line, you need a switch or button to signal the Reverse pin. The "Park" position on the transmission does indeed lock the transmission. You can bypass using the mechanical linkage altogether by putting the selector in any position other than Park, leave it there, and control Drive and Reverse with buttons and software only. Of course, you'll need to use the parking brake for "Park" then.
Houston, Texas, USA
EV Newbie
1979 MG Midget + GS450h = "Mexus?"
IT Consultant
Domt177
Posts: 219
Joined: Sun Aug 30, 2020 6:05 pm
Location: West Midlands, UK
Been thanked: 11 times

Re: GS450H Discussion

Post by Domt177 »

So if I plan to use this

https://zero-ev.co.uk/product/gear-sele ... nversions/

Then I’ll need an actuator to engage park, unless I keep the gearbox in neutral and use physical park brake
User avatar
PatrcioEV-ATX
Posts: 181
Joined: Sun Dec 13, 2020 5:25 pm
Location: Austin, TX, USA
Has thanked: 40 times
Been thanked: 53 times

Re: GS450H Discussion

Post by PatrcioEV-ATX »

ggeter wrote: Thu Dec 31, 2020 4:14 pm
PatrcioEV-ATX wrote: Thu Dec 31, 2020 1:38 pm I’ve read through the threads but I still have a (maybe) dumb question. In pre-CAN cars, how do you tell the VCU to change between drive and reverse? I think one could use a simple actuator on the transmission shift linkage to engage/disengage park. From what I’ve read that would disable motor movement.
I have a pre-CAN 1979 Midget...

My plan is to use the mechanical linkage to engage Park only and use a switch of some kind for Reverse and Drive. Damien's VCU takes a simple signal on one pin for Drive/Reverse. If I can figure out how to read the positions on the original selector, I can use that to then command the transmission to switch to Drive or Reverse. There is also a position past Drive for the M mode. But, my time is probably better spent rigging a solenoid to get in and out of Park, then just switches for R and D.

Bottom line, you need a switch or button to signal the Reverse pin. The "Park" position on the transmission does indeed lock the transmission. You can bypass using the mechanical linkage altogether by putting the selector in any position other than Park, leave it there, and control Drive and Reverse with buttons and software only. Of course, you'll need to use the parking brake for "Park" then.
I'm still in the planning stage obviously and am looking at several different motor/inverter options. Looking at the Thunderstruck controller for the LEAF, they have a one pin input for NDR as well. 0-1.5V for neutral, 1.6-4.0V for forward, and 4.1-5V for reverse (If I remember them correctly). If that's the same case with Dami's VCU, then we're good.

this looks slick and would fit in nicely with the chrome accents on, oh say, a 1963 Rambler Ambassador:

https://www.ebay.com/itm/2017-2020-FORD ... Swkm1fdiZb
1998 Ford ZX2 - DC EV conversion(sold) http://evalbum.com/2093
2012 Nissan Leaf (sold)
2016 Mercedes B250e (sold)
2023 Volvo C40

Current: 1964 Rambler Classic 660 w/ GS450h set up. 36kwh Tesla batteries from B250e.
https://www.instagram.com/rambler_660e/
Post Reply