Engine technology thread.

[bambbrose]

Originally Posted by Dimman

Does an AccessPort allow you to control the AVCS?

I'm not 100%, but I would say it has to doesn't it? You can't properly tune without AVCS control. If it doesn't, or isn't tunable, that will be a severely limiting factor.

I know on the K20a motors, they dyno pull with nearly every phase configuration held constant, and then map out the rpm torque peaks into a curve.

Another interesting fact, on the K20a motors, the advance/retard is around 25 ° if I remember correctly. This is to prevent contact if a component fails. A few hardcore tuners machine out the mechanical stop within the VTC, and run as far as 45°, although in that territory, if you screw up with the tune, you kiss a valve with the piston.

[Dimman]

Originally Posted by bambbrose

I'm not 100%, but I would say it has to doesn't it? You can't properly tune without AVCS control.

I know on the K20a motors, they dyno pull with nearly every phase configuration held constant, and then map out the rpm torque peaks into a curve.

I would agree, but I've met a ton of tuners that don't like to so they say that things can't be done.

Why wouldn't they hold the rpm steady (eddy-current dyno) and then adjust the advance/retard/overlap on the fly?

[bambbrose]

Originally Posted by Dimman

Why wouldn't they hold the rpm steady (eddy-current dyno) and then adjust the advance/retard/overlap on the fly?

That may be what they do, it's been a few years since I have read up on it. I do remember that they hold most variables constant to find that baseline peak torque curve, and then fine tune from there.

[Dimman]

Originally Posted by bambbrose

That may be what they do, it's been a few years since I have read up on it. I do remember that they hold most variables constant to find that baseline peak torque curve, and then fine tune from there.

I'm probably going to have to get in touch with some local Honda tuners. They probably have the most/best experience relevant to this car (at least the cam phasing bit). Weird.

I should also clarify my earlier rather silly overstatement. I've talked to a couple of local tuners with Supra experience in the context of putting an IS300 VVT-i 2JZGE head on my old 1JZGTE Supra, with a Vipec V88 or Haltech PS2000 running it. Asking one about tuning the VVT-i and how it should help spool, I got a blank look and ridiculou$ time quote. The other guy just asked why I didn't just run more boost and how he could just get the same result with a divided turbine... I interpreted these as "We don't know anything about VVT-i/cam phasing."

[Homemade WRX]

Originally Posted by Dimman

Does an AccessPort allow you to control the AVCS?

Of course but I'm talking of the phase timing for exhaust manifold design...that and from a cam design point of view.

As the comment about tuners, only the crappy/lazy tuners don't touch the exhaust AVCS, which is sadly and all to common occurrence. On EJ motors, when you get some real cams however, you quickly run out of AVCS room. This is one reason that for customers who want to keep it, I end up staking the AVCS cam gear to mechanically limit it's travel. You'll find this practice done in the Honda world too.

In the Subaru world, guys running big cams for lots of top end power (drag, standing mile or destroked builds), just remove AVCS all together as it's more or less dead weight and a liability (also a leach on the oil system).

[Dimman]

Originally Posted by Homemade WRX

Of course but I'm talking of the phase timing for exhaust manifold design...that and from a cam design point of view.

As the comment about tuners, only the crappy/lazy tuners don't touch the exhaust AVCS, which is sadly and all to common occurrence. On EJ motors, when you get some real cams however, you quickly run out of AVCS room. This is one reason that for customers who want to keep it, I end up staking the AVCS cam gear to mechanically limit it's travel. You'll find this practice done in the Honda world too.

So am I. What I'm looking at, is that with a bunch of bolt-ons, but stock cams, if we change the header tube lengths (plus whether we have a step, keep it tri-Y, go 4-1, etc...), we change when the wave returns are and possibly their shape and lengths as well. So to take full advantage of the headers, a re-tune including AVCS adjustments may (will probably) be necessary.

[Homemade WRX]

that's a given...however from a design point of view, the AVCS will play a factor in the cam design...not the other way around.

[Lonewolf]

I probably missed it, but is the FR-S/FT86 motor chain-driven (as opposed to belt)? I believe that it is chain-driven, but I want to be sure...

[Homemade WRX]

yes, it's chain

[Ryephile]

Originally Posted by Dimman

Dimman, according to the latest Top Gear magazine article:

Originally Posted by Ollie Marriage

I was shown a chart earlier that showed how the direct injection helped deliver almost as much torque from 2,800 to 3,200 RPM, as it does at the top end. Trouble is, in-between these points, the torque curve sags...

It appears the torque curve is very close to your Fig.2 graph. This means the VE curve of the engine is a bit funky where they chose to stretch out the torque "table" so far that they created a dip in torque output in the mid-range. We know the engine doesn't use variable lift, so perhaps the combination of manifold design, cam specs, and cam phasing is giving the engine this mid-range RPM torque dip?

[Dimman]

I'm thinking it could be the intake design. I haven't heard of any intake trickery (like ACIS) that is used to reduce flat spots. From what I'm looking into a 2-stage intake system is used to basically 'jump' over the rpm range where a flat spot would occur while still getting a significant power boost. With their emphasis on low-rpm driveability, they probably were unconcerned with the flat spot so long as they could get bigger gains down low and not lose any up top.

So instead of using the intake to boost the 'natural' intake-speed related torque peak, they left a flat spot there, and focused on either side of it to flatten everything out.

Which probably means with re-configured headers, intake and AVCS I can still get my 170 lb-ft peak.

[bambbrose]

Originally Posted by Dimman

I'm thinking it could be the intake design. I haven't heard of any intake trickery (like ACIS) that is used to reduce flat spots. From what I'm looking into a 2-stage intake system is used to basically 'jump' over the rpm range where a flat spot would occur while still getting a significant power boost. With their emphasis on low-rpm driveability, they probably were unconcerned with the flat spot so long as they could get bigger gains down low and not lose any up top.

So instead of using the intake to boost the 'natural' intake-speed related torque peak, they left a flat spot there, and focused on either side of it to flatten everything out.

Which probably means with re-configured headers, intake and AVCS I can still get my 170 lb-ft peak.

On the money, I agree 100%


Sometimes I wonder if stuff like this is intentional, for multiple reasons:
-to improve FE to grab a bigger market
-to support the aftermarket, by making the car "responsive to tuning"

the S2000 was the opposite of this approach. Aim for maximum top end, and you get a bad reception from the public outside of enthusiasts. Everyone says the S2000 is torqueless. The S2000 NA tuning aftermarket is basically useless without huge $$$.

I see the NA aftermarket for this motor being excellent, just like the K2X series motors from Honda.

[arghx7]

It's a lot like the 2GR-FSE [IS350 engine with D-4S)



I don't think there is a variable intake manifold system but the combined design of the intake ports, manifolds, etc probably results in the flat spot in the middle.

[Dimman]

Originally Posted by arghx7

It's a lot like the 2GR-FSE [IS350 engine with D-4S)



I don't think there is a variable intake manifold system but the combined design of the intake ports, manifolds, etc probably results in the flat spot in the middle.

The 2GR doesn't use ACIS either as far as I know, or TGVs.