Engine technology thread.

[Dimman]

Exage, remember when we were discussing the hypothetical de-stroked FB motors?

What did you say about the deck height for the rod:stroke ratio to be good? Needed changing, but then that might affect cooling because of water jacket volume changing, right?

Now think about this. We 'designed' (thanks to Old Greg for pointing it out I think...) the 72mm stroke FB motors that gave more performance oriented 1.6L (84mm bore) and 2.0L (94mm bore) motors with bores that were close to a couple of notable Toyota products (2ZZGE is 82mm and 2GRFSE is 94mm) that could have their combustion chamber/port designs transferred over to give us some pretty epic NA power plants (FB16R 180 bhp@8550 rpm, FB20R 205 bhp@7500 rpm).

Now with the rumour of Subaru developing a 1.6L for the next WRX solidifying, I was thinking about what Moto-P said in the interview...

Assume that maybe Moto-P mis-interpreted the EJ20 thing. Maybe only 'boxer' was used to reference the new motor and the 'EJ20' was used only in the mounts/swap context, and that this IS a new Subaru performance motor.

What if they had the same 72mm stroke idea (saves some money on components by using existing FB architecture) and came to the same deck height/rod:stroke ratio problem you pointed out. Solution? Re-cast the block with appropriate dimensions and coolant passages. And Moto mentions that being done.

Here's my fantasy: The Subaru and Toyota models should have different motors.

The Subaru gets a short-stroke 1.6L with a small turbo, and base FB heads (turbo needs enough flow, boost, efficiency and charge cooling for a 45% density increase) making 215 hp @ 7500 rpm.

The Toyota gets a short-stroke 2.0L NA based on the 2GR, with a bit more performance enhancement by Yamaha making 215 bhp @ 7800 rpm. Torque I'm not that great at predicting, but I think both would be around 150ish lb-ft, with. So both motors are similar in numbers, but the character would be quite different.

Maybe in Europe or Japan there is a Toyota 1.6L NA option, the above mentioned 180 bhp @ 8600 rpm.

Up the scale, the new WRX gets the short-stroke 1.6L with Yamaha headwork (I'm basing it on the 2ZZGE) and a bit more boost (55% more density) it makes 265 bhp @ 8600 rpm. This could also fit in the Suabaru 216a as the 'STI' version. They may swap more boost for less revs in the name of reliability, but as I have it here, it would be awesome. Japanese market would love it with their high revving turbo EJ20 history.

Finally the new STI gets a turbo version of the short stroke 2.0L with the same boost, but maybe higher-flowing turbo to match the extra displacement (also 55% more density) it makes about 315 bhp @ 7500 rpm.

All with Dual VVTL-i/Dual AVCS/i-ALVS, D4-S (and Twinscroll or VGT turbos for the boosted ones).


Dreaming while waiting for December specs...

[Exage]

Those are some seriously screaming engines!

I think they would go with the reduced rpm, lower compression, high boost application because it's a more simple build and less problematic (play it safe). I would definitely be interested in a high rpm turbo application if I had the money and time. The current twin scroll EJ207 in the JDM STI is 308ps@6400rpm on a 8:1 compression... But we can dream racing engines haha.

I must admit though, the fact that Toyota might be using a design similar to the EJ series throws me a kind of loop for the prediction for Subaru and their engines for their FT-86 project/WRX models. My guess is that it's between a new design or direct injection EJ series turbo update for their performance models.

Originally Posted by Moto-P

The engine? It's still EJ20 based shape. But Toyota re-casted everything, even down to the engine block with a new design, and with the Yamaha acoustics assigned to make every pulse, beat, and soul of air and vibration to become something to spice performance and tactile feel. Not much is exactly shared with the Subaru engine except the flat four configurations. However, the external mounting points remain the same as the Subaru EJ20, shared for very obvious reasons. That is so YOU can use a Subaru engine to swap if you PREFER the Subaru engine for racing. Why bother with complex fabrication, if Subaru has something you can use from 11 years of WRC?

It's thought out for the aftermarket builders, racing operations, and not without YOU as enthusiast, the grassroots racers, custom builders, show car builders, and parking lot meet goers.

My breakdown of this information:

Moto-P stated on the foundation is that:
A. It's the shape of the EJ20
B. The mounting points are the same as the EJ20

Which means Toyota is make it much more appealing to perform boxer swaps into the car. It's very possible that a EJ turbo might fit (infact he pretty much suggests this), or even shoehorn a EZ Flat-6 with some fabrication depending on the room.

That fact still remains that both of us have been somewhat accurate in predicting what engine internal specs are needed for Toyotas' goal of 200bhp for naturally aspirated boxer engine. Perhaps it should have occurred that they could've used an updated redesign of the dinosaur EJ block and not the FB. It is still possible to see incorporation of the components of the FB block design (or Toyota's knowledge) or that it's the FB 94mm bore redesign that you and I have suggested (,but perhaps less likely now).

An engine that caught my eye was the EJ204 with AVCS produced 190hp, unfortunately there is little information on the particular engine (or I didn't try hard enough).

Regardless of the base design of the engine block. I'm extremely pleased that Toyota has gotten in there examined it, and redesigned it. A member here asked months ago "does it made sense to create a entirely new engine just for the FT-86". Perhaps I should have responded with: If they really give a damn then yes.

I have yet to see something in the Toyota-Yamaha collaboration engine roster that would be a disappointment. Yamaha has proven that they can design heads and engine components that rival some of the most prestigious auto manufactures and make one hell of a noise while at it.

The Big T and Triple Tuning Forks undoubtedly have an excellent engine for the FR-S/FT-86. I can't wait to see the cards they hold and the ones up their sleeves.

[serialk11r]

Are you sure about the EJ204? I looked around on the internet, it seems like it was on the JDM Legacy, which isn't exactly a performance application, and some guy selling it on Alibaba claims 150hp. Seeing how the other EJ20s seem to be around that ballpark I think it's more likely. Only Wikipedia says 190hp, or so I have seen...

And over here I'm still hoping for Valvematic With the shorter stroke they just might be able to fit in the intermediate rockers, and afterall, the FT is somewhat wide...if they throw it on this car, it would be the first US model Toyota to have it, which would make this car even more special. Due to the huge part load efficiency increase, they could probably give it more aggressive cams for even more power up top without fuel economy penalty. And then adding turbo wouldn't hurt fuel economy much either. It would be perfect

If that doesn't happen, well it'll have to come sooner or later since there's no way car companies can hit CAFE requirements without more advanced valve control tech.

[Exage]

Originally Posted by serialk11r

Are you sure about the EJ204? I looked around on the internet, it seems like it was on the JDM Legacy, which isn't exactly a performance application, and some guy selling it on Alibaba claims 150hp. Seeing how the other EJ20s seem to be around that ballpark I think it's more likely. Only Wikipedia says 190hp, or so I have seen...

The EJ204 without AVCS is 150hp. Ichi is a Subaru nut and he seems to think it exists!

Originally Posted by ichitaka05

EZ30D runs 245hp/219tq vs Porsche Boxer 3.2L VarioCam Plus runs 280hp (VarioCam only rums 252hp)

FYI EJ25 don't run AVCS, only EJ204 NA DOHC AVCS runs 190hp. While Porsche haven't ran Flat-4 for long time (since 70's).

As for Turbo, 22B STi is still hands down powerful H4 engine Subie have made. 2.2L 350hp/267tq

Next closest ones are S series (S201~S204)... but they're 2L w 315hp.

http://www.ft86club.com/forums/showthread.php?p=40022

[Dimman]

Originally Posted by serialk11r

Are you sure about the EJ204? I looked around on the internet, it seems like it was on the JDM Legacy, which isn't exactly a performance application, and some guy selling it on Alibaba claims 150hp. Seeing how the other EJ20s seem to be around that ballpark I think it's more likely. Only Wikipedia says 190hp, or so I have seen...

And over here I'm still hoping for Valvematic With the shorter stroke they just might be able to fit in the intermediate rockers, and afterall, the FT is somewhat wide...if they throw it on this car, it would be the first US model Toyota to have it, which would make this car even more special. Due to the huge part load efficiency increase, they could probably give it more aggressive cams for even more power up top without fuel economy penalty. And then adding turbo wouldn't hurt fuel economy much either. It would be perfect

If that doesn't happen, well it'll have to come sooner or later since there's no way car companies can hit CAFE requirements without more advanced valve control tech.

The 190 hp version came in some special Touring Legacy or something. If you want more info on it Aussie Subaru forums are probably your best bet for English info on obscure JDM models.

I don't want Valvematic on this yet, as it will be a huge new learning curve for tuning and mods. Bring it out on the new Camry, Corolla or Yaris so that there will be a billion cars with it. That way picking one up at a wrecker to experiment with will be cheap. Then bring it out on the second generation FT86.

[serialk11r]

I guess that's a good point
They need to start ushering the FAE ZR engines in for the US market sooner! It's not fair that the Europeans have had it for like 3 years already When you look at the picture, it seems like the US must be the place where they're dumping all their stock of old engines or something. Sucks.

Oh now that you mention tuning, can someone explain what exactly that is? I know engines have mass airflow sensors and O2 sensors in exhaust and whatever, and I know the ECU will give different amounts of fuel under different circumstances, can anyone give a comprehensive explanation? The only difference Valvematic should bring really is reduced intake temperatures at part load, and it wouldn't affect full power that much, but if the intake temperature is lower then you have even less to worry about right?

[Dimman]

Originally Posted by serialk11r

I guess that's a good point
They need to start ushering the FAE ZR engines in for the US market sooner! It's not fair that the Europeans have had it for like 3 years already When you look at the picture, it seems like the US must be the place where they're dumping all their stock of old engines or something. Sucks.

Oh now that you mention tuning, can someone explain what exactly that is? I know engines have mass airflow sensors and O2 sensors in exhaust and whatever, and I know the ECU will give different amounts of fuel under different circumstances, can anyone give a comprehensive explanation? The only difference Valvematic should bring really is reduced intake temperatures at part load, and it wouldn't affect full power that much, but if the intake temperature is lower then you have even less to worry about right?

Strictly speaking, tuning is optimizing the engine for performance. (I was talking about 'modding'.) The ECU is set to be all things to all drivers. But as we know that means a bunch of compromises. So tuning would basically adjust all the parameters to favour performance (basically power and response) at the expense of maybe a bit of economy or emissions cleanliness. An example of where some ECU tuning is needed, is on a lot of cars with 'laggy' e-throttles. They do this for economy, but performance demands response, so the e-throttle is tuned (usually made more linear). Ignition timing and AFR's can be tweaked if you know you are only going to be running better gas. Some ECUs don't adjust beyond a certain limit, so mods that increase airflow (intake, headers, exhaust...) may need the ECU's fueling tweaked to take advantage of them. Stuff like that. This is what a lot of the piggy-back fuel controllers can do. They intercept the stock signals and allow you to adjust more or less (up to a limit) at different rpm points.

On the more complex end it is basically programming the ECU to what the car needs. This is what someone who's job title is a 'tuner' does. Usually they are the ones programming (sometimes from scratch, sometimes from 'base' maps) stand-alone aftermarket ECUs of heavily modded cars on the dyno.

[serialk11r]

So since I was rather bored I went and looked up more specifics on valve control systems, and I realized something...Valvematic actually does not add much height over a VVTL-i engine because Toyota does away with VVTL-i on Valvematic engines for simplicity! So the locking pin and rollers are replaced with a secondary shaft and rollers/followers. So there isn't actually a packaging issue with Valvematic.

I realized there's one issue with Valvematic however. Because it only has one cam profile that is high lift, the duration at low rpm will need to be limited to reduce the lift. With VVTi, they can advance the intake significantly then cut duration so some of the bottom of the intake stroke is used to expand the charge, reducing the pressure and helping the fuel evaporate (for the record, there are minimal pumping losses associated with this since the piston travels back up and regains the energy in the compression stroke). However this comes at the cost of reducing the amount of charge pulled in. With VVTL-i (and i-VTEC for that matter), the low rpm torque is pretty well optimized because the cam is a full duration cam with a low lift profile. Of course with less charge pulled in we have less pressure blown out the exhaust, but I think when Valvematic engines hit American shores we will notice that they have a slightly upward sloping torque curve. Unless of course Toyota figured out a different way to promote fuel atomization.

On the side, I noticed that wikipedia says all the 1ZR and 2ZR engines (the only valvematic engines so far besides the 3ZR-FAE) are built in Tianjin, China, in FAW factories, which might be why we don't see much of them here in the US. I'm guessing they're waiting to retool their factories elsewhere for more A engines since it'd mean throwing out everything they use to make the VVTL-i system. But then the 2ZZ-GE is at the end of its lifespan, so they might be starting already

[Dimman]

Originally Posted by serialk11r

So since I was rather bored I went and looked up more specifics on valve control systems, and I realized something...Valvematic actually does not add much height over a VVTL-i engine because Toyota does away with VVTL-i on Valvematic engines for simplicity! So the locking pin and rollers are replaced with a secondary shaft and rollers/followers. So there isn't actually a packaging issue with Valvematic.

I realized there's one issue with Valvematic however. Because it only has one cam profile that is high lift, the duration at low rpm will need to be limited to reduce the lift. With VVTi, they can advance the intake significantly then cut duration so some of the bottom of the intake stroke is used to expand the charge, reducing the pressure and helping the fuel evaporate (for the record, there are minimal pumping losses associated with this since the piston travels back up and regains the energy in the compression stroke). However this comes at the cost of reducing the amount of charge pulled in. With VVTL-i (and i-VTEC for that matter), the low rpm torque is pretty well optimized because the cam is a full duration cam with a low lift profile. Of course with less charge pulled in we have less pressure blown out the exhaust, but I think when Valvematic engines hit American shores we will notice that they have a slightly upward sloping torque curve. Unless of course Toyota figured out a different way to promote fuel atomization.

On the side, I noticed that wikipedia says all the 1ZR and 2ZR engines (the only valvematic engines so far besides the 3ZR-FAE) are built in Tianjin, China, in FAW factories, which might be why we don't see much of them here in the US. I'm guessing they're waiting to retool their factories elsewhere for more A engines since it'd mean throwing out everything they use to make the VVTL-i system. But then the 2ZZ-GE is at the end of its lifespan, so they might be starting already

With less charge pulled in, we get less power. Boo...

Extended duration allows cars to make more power up top when the intake charge is traveling faster. The extended intake duration allows the high energy (air/fuel has inertia) charge to continue to enter the combustion chamber even as the piston starts up on the compression stroke.

Sounding like Valvematic is more of an efficiency oriented system.

PS have any Valvematic engines incorporated D4-S? This would help with the atomization.

[Exage]

Originally Posted by Dimman

PS have any Valvematic engines incorporated D4-S? This would help with the atomization.

None that I know of. IS-F 2GR-GSE is a VVT-ie, is probably the closest. Still waiting on #XX-FASE I'm afraid.

[serialk11r]

Valvematic is certainly efficiency oriented, this sort of system is probably the simplest method of near lossless throttling. But you can't really call it an efficiency only add-on, as the better efficiency all around and "infinite" cam profiles lets them put more aggressive cams for best possible high end power. Of course the high rpm cams on VVTL-i and i-VTEC engines are pretty well optimized for max power as well, so the benefit isn't that large.

However the biggest advantage to performance in my opinion is the ability to add a turbo and lose very little part load efficiency. You'd still lose some energy to spooling the turbo up, and lower compression ratio, but that's much much better than losing energy on the exhaust stroke pushing gas through a turbo that subsequently pushes air through a partially closed throttle. As an example, the BMW 335 can get pretty similar mpg to the 328. The turbo likely even helps efficiency under acceleration. Of course the BMW engine's displacement is still rather high which hurts efficiency but it's pretty impressive. Similarly the 370Z's engine does alright in the fuel efficiency department too, although the even bigger displacement doesn't help.

Direct injection I'm guessing might have some issues with the most power optimized of valve control systems, Valvematic and VVTL-i, since the extra equipment in the head might crowd out the injector? You guys know anything about this?

@Dimman - I completely forgot that at high rpm the intake is still open after the piston starts coming back up! Depending on how they shape the cam then we might be able to have nearly full duration at low rpm
However I think the reduced duration might not be an issue at all, since earlier we were discussing combustion efficiency, and I guess it was established that combustion efficiency is a major issue, even with efforts like VVTL-i which have low lift at low rpm, plus the swirl whatever it's called stuff.. If the intake valve opens as the air starts coming in, but then closes a little early, say...80% through the intake stroke, when the piston hits the bottom our air/fuel mix is now at something like 0.8 atm pressure (assuming adiabatic expansion, 1.3ish heat ratio) and all of a sudden our fuel is above boiling point This can produce a nearly homogeneous charge! Considering the ideal thermodynamic cycle the 20% reduced fuel charge translates into most likely a less than 10% loss (very rough estimates here sorry) in overall mechanical work, but in real life the 100% duration charge would burn worse and the 80% duration charge would burn better and bring it even closer.

This is certainly exciting stuff, a #XX-FASE (or G) with VVT-iE + Valvematic would be getting close to a perfect gasoline engine, it would just need variable exhaust duration, which is a bit more difficult to do and possibly not worth the cost (although Fiat Multiair can probably accomplish that). Well, with an electrically coupled variable geometry turbo (but I suppose that's debatable).

[Snaps]

Originally Posted by serialk11r

Well, with an electrically coupled variable geometry turbo

What do you mean an electrically coupled turbo? Electrically assisted to remove spool lag?

[serialk11r]

Electrically coupled = turbine => generator => battery, motor => compressor. This way compressor speed is independent of turbine speed, so the turbine can be running closer to optimal speed and the compressor can have no lag. Electrically assisted might actually be better though, since at high speed the motors would have to be rather large to handle the current. Fully independent turbine would be better for efficiency as it could be just a generator type thing instead of needing to produce boost. But then again that's not really an issue as long as we have variable duration, and a blow off that vents back into the turbine instead of wasting the energy.

[bigbcraig]

Electric turbos have repeatedly failed in execution though. I understand electric motor technology is getting better, but (in my experience) people fail to realize just how great the forces acting on a turbo are.

A small turbo may run as high as 200,000rpm, and though there are electric motors that may be able to spin that quickly at no load, they have very little torque available at that speed. Thus, you have a motor/compressor that on paper can free-flow enough air but fails to actually build boost.

In a few years maybe, but compressors are better mechanical for now IMO (traditional supercharger or turbocharger)

On the other hand, things like electric water pumps are overall more efficient than their mechanical counterparts, for the reasons serialk11r stated: electrically driven, the pump will only have to operate in a narrow rpm range, rather than across a wide range as a function of engine speed. Thus, the compressor wheel can be more finely tuned. There will be a very narrow RPM band where the purely mechanical system is more efficient, but in general operation electric is better.

In a couple years, once DI has been pretty well sorted out, I'll begin to get hopeful again for a true revolution in valvetrain / engine breathing. The variable systems today are nifty but likely a system other than cams actuating poppet valves would be much more efficient, and I believe that will be the next bottleneck in engine performance.
But, my hopes won't be too high because people have claimed we were on the brink of a valvetrain revolution (be it rotary or otherwise) for the last 30 years.


Does F1 mandate poppet valves? I know they have ran some unconventional actuation systems (pneumatic, etc) but don't think there have been any F1 engines that tried to use rotary valves or otherwise. Formula 1 has traditionally been a great technology development platform, its sad if traditional setups are mandated.