We just wrapped up our highest horsepower 2GR to date and figured it's time to share what we've learned along the way.
Trevor had his MR2 turbo 2GR swapped at ATS in 2023. It was our regular setup at the time - stock engine, repurposed stock harness, Link G4X PnP ECU, Comp oilless water cooled 6265 turbo. It made an exceptional 664whp.
Trevor recently sent it back for an ATS forged motor, with some extra supporting goodies. We upgraded the turbo to a newer Comp 6871, and added Kelford cams to make more power at a lower boost. It's making 822whp at only 20psi! We upgraded to a brand new harness made by Wiregap exclusively for ATS, that plugs directly into an ECUMaster EMU Pro8, our preferred ECU for the 2GR. We swapped from FMW cast headers to the new TCS tubular headers, because they sound amazing.
It's hard to overstate how lucky we are to have the 2GR and E153.
This is a stock E153 with a Quaife differential. We have broken third gear on a few E153s above 600ftlbs, but this one should be reliable. The fact that Toyota built a transmission bolted to engines making less than 200whp, able to handle 800+, is just insane.
Adding power to anything gets expensive fast, A basic turbo kit is one of the cheapest ways to add power to any naturally aspirated engine. The 2GR gives us a ridiculously high starting point. We can reliably make 600whp on a stock 2GR. Getting a 3S based build, or just about any other engine from any manufacturer, to that level, takes a much larger investment. But of course, we can't just stop there.
How do we make more power with the 2GR?
Let's discuss what breaks. We've broken it all!
Pistons
The stock 2GR pistons are not built for this. They have tight ring clearances, tight wall clearance, and weak ringlands. Too much heat and the rings can touch or pistons can grow, causing transfer. Too much timing and a good detonation event can break the top ring land at the valve pockets. E85 remedies all of this by lowering heat and increasing octane. We have thoroughly tested these engines with prolonged abuse at 600whp on E85 and 400whp on pump gas. Above that, we recommend going to forged pistons.
Valve Springs
The valves will float on a stock 2GR around 620whp, as low as 6000 RPM. This is due to exhaust backpressure. A larger turbine housing helps the most. There is some variation between engines. There are 3 different stock 2GR valve springs I know of. We have not yet found the limit of the affordable BC made valve springs we use.
Straight wound - 620whp
Beehive with purple paint - 580whp
Beehive with no paint - 650whp
The valve float can be pushed through to some extent, but we try to avoid it. The engine will typically make another 30whp after the float starts, but no 2 dyno runs will be the same and there will be a wavy effect visible on the graph.
Let's jump back to the dyno graph above for Trevor's car. I said his engine made an exceptional amount of power - that's because it had the unpainted beehive springs. You can also notice a bit of weirdness at the end of the rev range. That's the valve float.
Connecting Rods
Stock 2GR connecting rods will bend around 550ftlbs to the wheels.
Did you notice Trevor's stock engine made 530ftlbs? This level is still safe. However, the 2GR makes about 18ftlbs of torque per pound of boost. That means he was less than 2psi away from bending a rod. We set an aggressive overboost protection on these, but there's only so fast it can catch an overboost event. This is why we now set these to 500ftlbs on a stock motor. Trevor was tripping the overboost protection a little while before his engine died in spectacular fashion. We've bent rods in 4 engines now - it's really hard to detect. The bent rod will however fatigue and break at the small end in about 100 miles.
We use BC connecting rods in all of our built 2GRs.
Rod Bearings
The 2GR is designed for 5W or 0W oil. Therefore they have a teeny tiny oil clearance at the rod bearings. In addition, the bearings are not staked. They are only held in place by friction. At high RPM, all it takes is a split second event to for the bearing to grab harder to the crank than the rod, and it will spin. We run 10w30 VR1 in all of our 2GRs, and they have proven reliable at 7500rpm. Any time we'll be revving higher than that we install ACL race bearings with an increased oil clearance.
Head Sealing
The stock 2GR head gaskets are really good. For the most part, MLS is MLS. We use Cometic because they have a more defined flame ring.
Our largest headache with upgrading the 2GR has been with head fasteners. We have lifted heads with aftermarket head bolts at 600whp. We have lifted heads with new OEM head bolts around 700whp. We have even lifted heads with ARP head studs multiple times. Up until about a year ago, we were using head studs from other vendors. Some had a smaller shank than a stock head bolt, that can't be an upgrade! Some were listing a torque spec way lower than ARP recommended for that size fastener. We have since started sourcing these ourselves, and the problems have stopped.
We offer two head stud solutions. One is an ARP2000 stud from a Tundra kit, that we break up into sets of 16. This isn't perfect, the threads on both ends are a bit short. However, the shank diameter is perfect, and torqued properly they haven't been an issue at 750whp.
The other option is an ATS custom made ARP 625+ head stud. We got really tired of rebuilding engines and ordered these. They have ideal thread engagement, an incredible tensile strength and can be tightened higher for increased clamp. Trevor's car has them and shows no sign of head gasket issues.
Something to note here. The 2GR is open deck. There is a possibility that the cylinders are actually moving around and doing what's commonly referred to as wiping, slowly wearing down the sealing coat on the surface of the gasket. There have been several times that we had a car properly lift a head. It's rather obvious on the dyno. Change in exhaust tone, dip in power, immediate overheat, cooling system purging. Wiping will become evident more gradually. There won't be an obvious moment, the car will just overheat and/or develop air pockets out of nowhere. Every car that we've had do this had under-tightened head studs, however we have not been able to identify obvious wiping on a problematic head gasket. More clamping force and more defined fire rings in the head gasket will prevent wiping. I'm confident Trevor's car will be reliable, but I think above this, CSS is a very worthwhile insurance.
Camshafts
If you can increase the flow of the motor, you can make the same power at lower boost. Trevor's car is making 822whp at 20psi. That power level would normally be about 27psi with stock cams. Lower boost lowers intake air temperatures and exhaust backpressure. However, cams will impact the bottom end. It's a trade off.
We used the Kelford 263-C cams on Trevor's car. VVT means that a really big cam can still run well. These are a bit excessive for a turbo car and hurt bottom end more than I'd like. In the future I'd recommend the 263-B cams for a turbo car.
We have confirmed all of the Kelford cams clear stock FE pistons at stock timing throughout the VVT range. The 263C cams especially are tight at full intake cam advance with only about .5mm piston to valve clearance on stock FE pistons. At high RPM, where valve float causing contact is a threat, there is significantly more clearance as the cams will not be as advanced.
Intake Manifold
A short runner intake manifold is almost a compulsory addition with cams, but only NA. We put this car on the dyno NA, as we have not done a set of Kelford cams here yet and we were curious. The car made 340whp on a setup that normally makes 298. On the dyno graph, you can very clearly see the waves of resonance from the intake manifold. The strongest at 6200, and a weaker one at 7300. With the stock rev limiter, the strongest wave at 6200 gives you the broadest powerband. However, with the addition of cams, 6200 is really too low. I suspect the car would have made 370whp with a Frankenstein intake manifold, which moves the strongest wave closer to 8000.
Turbo is a different story. We tend to err on the side of too long with intake runners. Longer runners make more torque out of boost, and spool the turbo faster. Once the turbo is spooled, we can just add more boost up top. On Trevor's car, an intake manifold would have been an expense and fitment complication in vain. It would be able to make the same power at 3psi less boost, but the fuel pump would have run out of flow at the same peak power, and drivability would have suffered.
Fuel System
We use 1300cc Bosch based injectors that we flow match in house. These will be good for well over 900whp at 4 bar fuel pressure. We normally install a Walbro 525 hellcat pump. Trevor's car has this, and is dropping pressure on E85 right around 800whp. That's a really good single pump.
From here, the fuel system gets significantly more complicated. You're looking at dual pumps, or a brushless pump. The factory fuel return line is already too small for the hellcat pump, with more flow it will definitely need to be upgraded.
Intercooler
Our standard air to air intercooler we use on these is pretty small. Luckily the 2GR barely needs any boost to make power, so it has an easy job. The standard intercooler is at its limit at 600whp.
There are two other options. First, trunk mount intercooler, the cheaper of the options. This is what Trevor has. We've been doing these for years on 5SGTE builds. We can fit a massive core in the trunk. We put 2 high flow fans on the trunk floor and open up side vent airflow to the trunk for fresh air. This works great, but keeping your trunk is one of the attractions of 2GR.
Water to air is, in my opinion, the best way to go. I will write a separate blog entry on this, there is too much to get into here and we are working on some new things in this department.
Turbo
We normally use a Comp 6265. Initially we used a .82 turbine housing. Once we understood the valve spring issue, we switched to a 1.06 turbine housing. This hurts spool a bit but allows another 20whp or so before the valves float, from the decreased backpressure.
We have pushed the 6265 to its limit, which is about 740whp. As we push the limit, backpressure increases faster, which I'm sure hasn't helped our head gasket struggles.
Trevor was previously a 6265 with .82 turbine housing. He is now 6871 with a 1.06 turbine housing. We have turbo size and cams hurting spool, and only lost about 600rpm. We gained a ton up top. I'd say that's a worthwhile trade off.
Exhaust
Pre-turbo, the TCS headers and y-pipe are our go to. These are essentially a reproduction of the old Frankenstein tubular headers, but with thicker material. The FMW units ALWAYS crack at the flanges. The TCS ones have been great so far and sound amazing.
Post turbo, Trevor still has our dual exhaust with a cutout. The cutout means it's basically just a 3 foot long 3" diameter straight pipe. I suspect a larger diameter is in our future.
Â
What's next? Will we see 1000whp in 2026?
We're very optimistic that we will see 1000whp in the coming months. We have a build in the works getting a whole bunch of upgrades. CSS, fuel system, W2A upgrades, intake manifold, porting. Stay tuned for that.
In the meantime, we'll keep cranking out 600whp stock 2GRs. If you're insane and 600whp isn't enough for you, let us know when you're ready for more!