Building an IS350 6MT
Thread Starter
Joined: Apr 2009
Posts: 5,211
Likes: 746
From: California
FWIW I did a project like this for my street bike. In short Yamaha limited ignition advance in 1st, 2nd, and to a much lesser extent 3rd gear by comparing the CPS signal to the final drive to tell the ECU what gear the bike is in. I made a circuit that made it think it was in 3rd gear and up. However with all the data on a single data line, it also hacks with the reported speed and distance on the multi-function meter / speedometer so i removed it. The hardest part was trying to mimic the output from a hall like sensor where the voltage goes up exponentially with shaft speed as it induced errors the ECU didn't care for. The ECU can be flashed but they want $600 USD to throw a couple of ignition maps at it so I declined.
I know this project would be huge but it opens up the possibilities to do some fun swaps if done right.
I know this project would be huge but it opens up the possibilities to do some fun swaps if done right.
As for housing bushings, yes, I know the 250 housing is different but went looking for TRD parts for the later housing which Lexus seems to put in just about everything 2013 and up. Likely just go with poly as it's not likely to need maintenance like a suspension bushing wanting lube every 9 months.
Thanks for you're input.
Thanks for you're input.
Jeff
JM2C for anyone who wants a 3.5L 6MT: we have snow on the ground, the car's more or less mine to take apart and play with, and I've already put myself into a CEL fever dream with all my second-guessing of solder joints and signal noise...so it seems the fastest/easiest way for me to make the thing faster before next year is to add displacement. We all know there's no replacement for it. Also gives me a much broader range of variables to keep tabs on, so I have a better idea of what does and doesn't like being played with. I'm very much at the BSEL mod point of OCing, here.
My idea is to just skip over bolting this ported/polished pair of heads up to our currently running mostly-factory 4GR, get a JDM 2GR, and swap the 6-injector 4GR topend onto the 2GR's bottom end, keeping the 4GR 6MT electronics. I'd be willing to bet the 6-injector topend will out-produce the 12-injector one in torque (esp w/o the slushbox). While doing the swap, I'd be sure to keep the 2GR's camshaft cradles with the 6-injector topend. Swapping 4GR cam cradles for 2GR ones seems a worthwhile thing to do, even on factory 2.5L 4GRs.
Anyway, I figure that'd be a slightly more elegant solution, since, IIRC, Jeff's engine doesn't run the direct injectors, just the ports. Unless RR hooked you up somewhere I didn't catch...
My idea is to just skip over bolting this ported/polished pair of heads up to our currently running mostly-factory 4GR, get a JDM 2GR, and swap the 6-injector 4GR topend onto the 2GR's bottom end, keeping the 4GR 6MT electronics. I'd be willing to bet the 6-injector topend will out-produce the 12-injector one in torque (esp w/o the slushbox). While doing the swap, I'd be sure to keep the 2GR's camshaft cradles with the 6-injector topend. Swapping 4GR cam cradles for 2GR ones seems a worthwhile thing to do, even on factory 2.5L 4GRs.
Anyway, I figure that'd be a slightly more elegant solution, since, IIRC, Jeff's engine doesn't run the direct injectors, just the ports. Unless RR hooked you up somewhere I didn't catch...
Spoiler
Thread Starter
Joined: Apr 2009
Posts: 5,211
Likes: 746
From: California
My idea is to just skip over bolting this ported/polished pair of heads up to our currently running mostly-factory 4GR, get a JDM 2GR, and swap the 6-injector 4GR topend onto the 2GR's bottom end, keeping the 4GR 6MT electronics. I'd be willing to bet the 6-injector topend will out-produce the 12-injector one in torque (esp w/o the slushbox). While doing the swap, I'd be sure to keep the 2GR's camshaft cradles with the 6-injector topend. Swapping 4GR cam cradles for 2GR ones seems a worthwhile thing to do, even on factory 2.5L 4GRs.
Anyway, I figure that'd be a slightly more elegant solution, since, IIRC, Jeff's engine doesn't run the direct injectors, just the ports. Unless RR hooked you up somewhere I didn't catch...
Anyway, I figure that'd be a slightly more elegant solution, since, IIRC, Jeff's engine doesn't run the direct injectors, just the ports. Unless RR hooked you up somewhere I didn't catch...
The closest to your idea I've seen people do is running the 3GR-FSE with the 4GR ECU and then either retuned it (for forced induction, etc.) or let the ECU compensate for the additional displacement. I can appreciate that you've already spent the time to port and polish the 4GR heads, but it seems like it would also be a bit of a waste not to utilize the larger bore diameter for the larger 2GR valves. Additionally, if what you're looking to do is exclusively run direct injection, it seems almost trivial to delete the port injectors from the 2GR intake.
I'll be interested to see what you come up with either way, sounds like it should be fun.
Jeff
Well, the idea is to run a 3.5l engine in front of the RA62. Chances are significantly greater than not that anyone attempting this won't have access to much, if any, in the way of ECU programming ability, without going through what will effectively be a unique means of sourcing such, thus excluding everyone else (I predict a post-coffee edit here, that was a mess)
If, given just a Toyota/Lexus junkyard and hand tools, nobody can really recreate your car, Jeff. Your route is essentially unique, like Thunder28's approach to LS-swap. I'm looking for a route that one person with basic mechanical competence can follow without having to ask someone else for help. Hence, swap the bottom end and see what the ECU does to fueling when the MAF sees more than the ~100gm/s it's used to. I'm betting the bump in CR between 3GR and 4GR is to keep air demands and low-end torque similar between the two engines so that a new set of airing/fueling tables didn't need to be made. I'm also betting that the small-valve, high-restriction 6-injector heads with the SCV and ACIS doodads in the way will probably be a good 25hp or more down p-t-p on the 12-injector setup a 2GR comes with, but--much like putting a mild set of 283 heads&intake on a warm-cam 383--torque will be significantly higher than either produce on their own due to the in-chamber turbulence caused by the higher port velocity from the higher air demands at the valve. I bet they had that much torque to the wheels in mind when they killed us with the 3.73 ratio. When driving a manual transmission, having a fat torque curve is remarkably more noticeable than on a slushbox just by the way each work...manuals have a physical connection from crank to differential, autos will always have a fluid gap. So I'm not too worried about peak HP so much as always having more than one "right gear" for any given speed under 70.
And yes, I understand the 283/383 is typically carbureted...as we know from the SAE papers redspencer posted, the 3GR topend was designed for high turbulence, already. The 2GR's V-slit injector is to help fuel the "top" of the cylinder while the heavy fuel-charged port-injected air is barging it's way past the valvestem/head to the "bottom" of the chamber at higher velocities. The 3/4GR's flat spray is intended to stratify the charge at the expense of higher flow, which produces a higher chamber pressure at peak torque, but starts "losing control" of the charge at higher RPMs. I may need to run a 3GR's SCV and intake setup to give the 3.5l bottomend enough plenum to suck on, but I really doubt I'll be doing anything the factory 6MT ECU can't adapt to. If it chokes for air in the last thousand RPM, that's what nitrous is for. The SCV plate seems to be made with n2o in mind; you're always just an S bit and 1/8npt tap from being faster...or a pedestrian.
Certainly a better idea than the current one I'm running with...
If, given just a Toyota/Lexus junkyard and hand tools, nobody can really recreate your car, Jeff. Your route is essentially unique, like Thunder28's approach to LS-swap. I'm looking for a route that one person with basic mechanical competence can follow without having to ask someone else for help. Hence, swap the bottom end and see what the ECU does to fueling when the MAF sees more than the ~100gm/s it's used to. I'm betting the bump in CR between 3GR and 4GR is to keep air demands and low-end torque similar between the two engines so that a new set of airing/fueling tables didn't need to be made. I'm also betting that the small-valve, high-restriction 6-injector heads with the SCV and ACIS doodads in the way will probably be a good 25hp or more down p-t-p on the 12-injector setup a 2GR comes with, but--much like putting a mild set of 283 heads&intake on a warm-cam 383--torque will be significantly higher than either produce on their own due to the in-chamber turbulence caused by the higher port velocity from the higher air demands at the valve. I bet they had that much torque to the wheels in mind when they killed us with the 3.73 ratio. When driving a manual transmission, having a fat torque curve is remarkably more noticeable than on a slushbox just by the way each work...manuals have a physical connection from crank to differential, autos will always have a fluid gap. So I'm not too worried about peak HP so much as always having more than one "right gear" for any given speed under 70.
And yes, I understand the 283/383 is typically carbureted...as we know from the SAE papers redspencer posted, the 3GR topend was designed for high turbulence, already. The 2GR's V-slit injector is to help fuel the "top" of the cylinder while the heavy fuel-charged port-injected air is barging it's way past the valvestem/head to the "bottom" of the chamber at higher velocities. The 3/4GR's flat spray is intended to stratify the charge at the expense of higher flow, which produces a higher chamber pressure at peak torque, but starts "losing control" of the charge at higher RPMs. I may need to run a 3GR's SCV and intake setup to give the 3.5l bottomend enough plenum to suck on, but I really doubt I'll be doing anything the factory 6MT ECU can't adapt to. If it chokes for air in the last thousand RPM, that's what nitrous is for. The SCV plate seems to be made with n2o in mind; you're always just an S bit and 1/8npt tap from being faster...or a pedestrian.
Certainly a better idea than the current one I'm running with...
Hi, i have a 2006 is250 i am pretty much interested in swapping is350 setup in the is250 i would like to ask is it possible that if i have a wrecked is350 i can swap its engine/tranny/driveshaft/differential and ecu. Will i need to do something regarding vdim or since i am using everything from the is350 it will be a simple swap?someone told me you will need standalone since the is350 ecu will throw check engine lights on the is250 bec there are some options that are not available in the is250.
Hey Jeff, I was trying to PM you but guess I don’t have enough permission yet... I was hoping you could steer me in the right direction, I’m swapping my IS250 auto to manual but I’m running an LS swap, hoping you could help me out with the Auto2Manual wiring? How’d you wire in the clutch switch? Did you have any issues with push to start?
im using 2 ecu’s for now until I figure out how/if I can use 1 only btw...
im using 2 ecu’s for now until I figure out how/if I can use 1 only btw...
Thread Starter
Joined: Apr 2009
Posts: 5,211
Likes: 746
From: California
Well as my car was already a manual the clutch switch was already there. That said, I was having issues with the ECU thinking that every time I clutched in I was in neutral and so for a short period of time I actually switched over to an automatic power source control ECU so that I could start the car with the brake pedal instead. Eventually I figured out a way to separate the clutch pedal starting functionality from the ECU's detection and got it working how it should be.
So what exactly are you looking to achieve? You want to start the car with the clutch pedal instead of with the brake pedal, or are you just trying to find a way to get the stock ECU out of there?
Is the car already running and driving at this point?
Jeff
So what exactly are you looking to achieve? You want to start the car with the clutch pedal instead of with the brake pedal, or are you just trying to find a way to get the stock ECU out of there?
Is the car already running and driving at this point?
Jeff
Well, the idea is to run a 3.5l engine in front of the RA62. Chances are significantly greater than not that anyone attempting this won't have access to much, if any, in the way of ECU programming ability, without going through what will effectively be a unique means of sourcing such, thus excluding everyone else (I predict a post-coffee edit here, that was a mess)
If, given just a Toyota/Lexus junkyard and hand tools, nobody can really recreate your car, Jeff. Your route is essentially unique, like Thunder28's approach to LS-swap. I'm looking for a route that one person with basic mechanical competence can follow without having to ask someone else for help. Hence, swap the bottom end and see what the ECU does to fueling when the MAF sees more than the ~100gm/s it's used to. I'm betting the bump in CR between 3GR and 4GR is to keep air demands and low-end torque similar between the two engines so that a new set of airing/fueling tables didn't need to be made. I'm also betting that the small-valve, high-restriction 6-injector heads with the SCV and ACIS doodads in the way will probably be a good 25hp or more down p-t-p on the 12-injector setup a 2GR comes with, but--much like putting a mild set of 283 heads&intake on a warm-cam 383--torque will be significantly higher than either produce on their own due to the in-chamber turbulence caused by the higher port velocity from the higher air demands at the valve. I bet they had that much torque to the wheels in mind when they killed us with the 3.73 ratio. When driving a manual transmission, having a fat torque curve is remarkably more noticeable than on a slushbox just by the way each work...manuals have a physical connection from crank to differential, autos will always have a fluid gap. So I'm not too worried about peak HP so much as always having more than one "right gear" for any given speed under 70.
And yes, I understand the 283/383 is typically carbureted...as we know from the SAE papers redspencer posted, the 3GR topend was designed for high turbulence, already. The 2GR's V-slit injector is to help fuel the "top" of the cylinder while the heavy fuel-charged port-injected air is barging it's way past the valvestem/head to the "bottom" of the chamber at higher velocities. The 3/4GR's flat spray is intended to stratify the charge at the expense of higher flow, which produces a higher chamber pressure at peak torque, but starts "losing control" of the charge at higher RPMs. I may need to run a 3GR's SCV and intake setup to give the 3.5l bottomend enough plenum to suck on, but I really doubt I'll be doing anything the factory 6MT ECU can't adapt to. If it chokes for air in the last thousand RPM, that's what nitrous is for. The SCV plate seems to be made with n2o in mind; you're always just an S bit and 1/8npt tap from being faster...or a pedestrian.
Certainly a better idea than the current one I'm running with...
If, given just a Toyota/Lexus junkyard and hand tools, nobody can really recreate your car, Jeff. Your route is essentially unique, like Thunder28's approach to LS-swap. I'm looking for a route that one person with basic mechanical competence can follow without having to ask someone else for help. Hence, swap the bottom end and see what the ECU does to fueling when the MAF sees more than the ~100gm/s it's used to. I'm betting the bump in CR between 3GR and 4GR is to keep air demands and low-end torque similar between the two engines so that a new set of airing/fueling tables didn't need to be made. I'm also betting that the small-valve, high-restriction 6-injector heads with the SCV and ACIS doodads in the way will probably be a good 25hp or more down p-t-p on the 12-injector setup a 2GR comes with, but--much like putting a mild set of 283 heads&intake on a warm-cam 383--torque will be significantly higher than either produce on their own due to the in-chamber turbulence caused by the higher port velocity from the higher air demands at the valve. I bet they had that much torque to the wheels in mind when they killed us with the 3.73 ratio. When driving a manual transmission, having a fat torque curve is remarkably more noticeable than on a slushbox just by the way each work...manuals have a physical connection from crank to differential, autos will always have a fluid gap. So I'm not too worried about peak HP so much as always having more than one "right gear" for any given speed under 70.
And yes, I understand the 283/383 is typically carbureted...as we know from the SAE papers redspencer posted, the 3GR topend was designed for high turbulence, already. The 2GR's V-slit injector is to help fuel the "top" of the cylinder while the heavy fuel-charged port-injected air is barging it's way past the valvestem/head to the "bottom" of the chamber at higher velocities. The 3/4GR's flat spray is intended to stratify the charge at the expense of higher flow, which produces a higher chamber pressure at peak torque, but starts "losing control" of the charge at higher RPMs. I may need to run a 3GR's SCV and intake setup to give the 3.5l bottomend enough plenum to suck on, but I really doubt I'll be doing anything the factory 6MT ECU can't adapt to. If it chokes for air in the last thousand RPM, that's what nitrous is for. The SCV plate seems to be made with n2o in mind; you're always just an S bit and 1/8npt tap from being faster...or a pedestrian.
Certainly a better idea than the current one I'm running with...
I'm thinking about doing this swap and the ECU/the electronics are the only thing keeping me from it.
And since my car is a 2007 model so it doesn't have this VDIM thing often seen on this thread I would need to have a different aproach to the ECU problem. I have also had this idea that you could use the 4GR 6MT ECU with the 2GR running only 6 injectors. Is it a must to swap the 4GR topend to the 2GR bottom if you want to use the 250 wiring or is it possible to use that wiring with the 2GR topend?
The 12-injector ECU and the 6-injector ECU aren't swappable without reprogramming, as far as I know. The 6-shooter has ACIS and SCV to control, while the 2GR-FSE makes up for that with a combination of V-slit injector nozzles and port injectors that fill in the blind spots from the V.
Now, the 6-shooter ECU from a 4GR can run a 4GR-FSE topend (heads, intake, wiring) on a 3.5l bottom end, and my prediction is that you'll max the 4GR flat-spray injectors out at right about 300WHP. Without supplemental fueling, I believe the V-spray injectors from the 2GR-FSE would result in hot spots inside the chamber that'd lead to the ECU pulling timing back from optimum. Leaving a 2GR-FSE sealed up and trying to run a 4GR ECU and injectors in a 3.5l engine would likely result in a sluggish engine due to the lack of ACIS/SCV or supplemental fueling. The 3GR and 4GR require the intake air charge to be very high energy, while the 2GR-FSE is essentially running a nitrous shot without the nitrous.
Now, the 6-shooter ECU from a 4GR can run a 4GR-FSE topend (heads, intake, wiring) on a 3.5l bottom end, and my prediction is that you'll max the 4GR flat-spray injectors out at right about 300WHP. Without supplemental fueling, I believe the V-spray injectors from the 2GR-FSE would result in hot spots inside the chamber that'd lead to the ECU pulling timing back from optimum. Leaving a 2GR-FSE sealed up and trying to run a 4GR ECU and injectors in a 3.5l engine would likely result in a sluggish engine due to the lack of ACIS/SCV or supplemental fueling. The 3GR and 4GR require the intake air charge to be very high energy, while the 2GR-FSE is essentially running a nitrous shot without the nitrous.
The 12-injector ECU and the 6-injector ECU aren't swappable without reprogramming, as far as I know. The 6-shooter has ACIS and SCV to control, while the 2GR-FSE makes up for that with a combination of V-slit injector nozzles and port injectors that fill in the blind spots from the V.
Now, the 6-shooter ECU from a 4GR can run a 4GR-FSE topend (heads, intake, wiring) on a 3.5l bottom end, and my prediction is that you'll max the 4GR flat-spray injectors out at right about 300WHP. Without supplemental fueling, I believe the V-spray injectors from the 2GR-FSE would result in hot spots inside the chamber that'd lead to the ECU pulling timing back from optimum. Leaving a 2GR-FSE sealed up and trying to run a 4GR ECU and injectors in a 3.5l engine would likely result in a sluggish engine due to the lack of ACIS/SCV or supplemental fueling. The 3GR and 4GR require the intake air charge to be very high energy, while the 2GR-FSE is essentially running a nitrous shot without the nitrous.
Now, the 6-shooter ECU from a 4GR can run a 4GR-FSE topend (heads, intake, wiring) on a 3.5l bottom end, and my prediction is that you'll max the 4GR flat-spray injectors out at right about 300WHP. Without supplemental fueling, I believe the V-spray injectors from the 2GR-FSE would result in hot spots inside the chamber that'd lead to the ECU pulling timing back from optimum. Leaving a 2GR-FSE sealed up and trying to run a 4GR ECU and injectors in a 3.5l engine would likely result in a sluggish engine due to the lack of ACIS/SCV or supplemental fueling. The 3GR and 4GR require the intake air charge to be very high energy, while the 2GR-FSE is essentially running a nitrous shot without the nitrous.
It is doable to make such a device, and with some math one could get the volume calculations close enough to start tuning it. I'd guess it would need some form of gain control as well as duty cycle/duration.
To your point, then you wonder about the MAF seeing huge volumes and the VVTi which we know so little about.
The 12-injector ECU and the 6-injector ECU aren't swappable without reprogramming, as far as I know. The 6-shooter has ACIS and SCV to control, while the 2GR-FSE makes up for that with a combination of V-slit injector nozzles and port injectors that fill in the blind spots from the V.
Now, the 6-shooter ECU from a 4GR can run a 4GR-FSE topend (heads, intake, wiring) on a 3.5l bottom end, and my prediction is that you'll max the 4GR flat-spray injectors out at right about 300WHP. Without supplemental fueling, I believe the V-spray injectors from the 2GR-FSE would result in hot spots inside the chamber that'd lead to the ECU pulling timing back from optimum. Leaving a 2GR-FSE sealed up and trying to run a 4GR ECU and injectors in a 3.5l engine would likely result in a sluggish engine due to the lack of ACIS/SCV or supplemental fueling. The 3GR and 4GR require the intake air charge to be very high energy, while the 2GR-FSE is essentially running a nitrous shot without the nitrous.
Now, the 6-shooter ECU from a 4GR can run a 4GR-FSE topend (heads, intake, wiring) on a 3.5l bottom end, and my prediction is that you'll max the 4GR flat-spray injectors out at right about 300WHP. Without supplemental fueling, I believe the V-spray injectors from the 2GR-FSE would result in hot spots inside the chamber that'd lead to the ECU pulling timing back from optimum. Leaving a 2GR-FSE sealed up and trying to run a 4GR ECU and injectors in a 3.5l engine would likely result in a sluggish engine due to the lack of ACIS/SCV or supplemental fueling. The 3GR and 4GR require the intake air charge to be very high energy, while the 2GR-FSE is essentially running a nitrous shot without the nitrous.
I thought it was too good to be true, too, which is why I bought a 3GR the same week a member on here posted their success with our supposedly too-good-to-be-true idea. Lemme dig up who that was...
Edit:
Edit:
Lexus tech here and just finally got the time with the new baby to swap up to the 3.5l and glad I didn't just to a complete long block rebuild on the original and spent the money for the long block rebuild on the big brother.
Upgrade upgrade upgrade is my vote as well. All in all around 1700 bucks in parts and total time to pull the old 2.5 and rebuild and install the much better bigger brother was 13 hours. Glad I took my Sunday to do so. At the dealership either rebuild and long block rebuild I get to do takes me 10 hours.
Upgrade upgrade upgrade is my vote as well. All in all around 1700 bucks in parts and total time to pull the old 2.5 and rebuild and install the much better bigger brother was 13 hours. Glad I took my Sunday to do so. At the dealership either rebuild and long block rebuild I get to do takes me 10 hours.
Last edited by Ultra4; Sep 22, 2021 at 04:40 PM.
What is the difference on the 3GR and 4GR heads? Is it just "minor" things like the intake and exhaust valves being bigger on the 3GR? Which I would think is the case. I was thinking that the 3GR topend on top of the 2GR bottom would make slightly more power than the 4GR topend. But I'll propably go with the 4GR topend since I already have it. My project will start next spring to hopefully be ready at the beginning of summer because its not worth it (nor do I have the time or money) to start it now since winter is coming and its gonna be snowing before the car would be ready to hit the roads again. So I wouldn't be able to drive it properly in the snow and make the adjustments that most likely would need to be done. And I would love to video it and put it on youtube cause I think there are some people who would like to see that. And thank you for clearifying that it is in fact relatively easy to do this swap.
Edit:
You'll need some serious Wuhan-windmills to spit on that M3.😂 Good luck with that hope you do smoke him.
Edit:
You'll need some serious Wuhan-windmills to spit on that M3.😂 Good luck with that hope you do smoke him.
Last edited by Slais250; Sep 22, 2021 at 05:30 PM.
As far as I've been able to tell without taking the heads off my 3GR, the shape and size of the air paths are the only difference, and I suspect only on the exhaust side. 3GR exhaust port looks like a 2GR, while 4 GR is oval. Camshafts are cast with different identifiers, too, so that might also be a difference, though I've seen two places claim 3 and 4 GR share cam geometry.
Edit: he sold the M3 to get a plane, but thankfully F80s are popular up here, so my benchmark still remains.
Edit: he sold the M3 to get a plane, but thankfully F80s are popular up here, so my benchmark still remains.
I was reading some earlier posts on this thread and started to wonder if it is possible to use the 4GR intake with the 2GR heads? And if so that would mean that the port injectors would be taken out of the game and then maybe one could use some bigger flat injectors instead of the V- injectors. And this way one could utilize the bigger bore diameter of the 2GR valves. And the ECU would be fine since by using the 4GR intake there would be ACIS.
A plane would be awesome to own someday so good for him. And good for you that you have your benchmark stil.
A plane would be awesome to own someday so good for him. And good for you that you have your benchmark stil.








