Put me in the "not impressed" column. Reducing mid-range torque in order to ward off detonation is the definition of throwing the baby out with the bathwater.

I agree somewhat with that point, though the ends may justify the means, or at least the intent. Look at it another way maybe. If one can shift the VE peak upward with the shorter runners, and compensate with boost at the 'old' lower RPM VE peak, why not?

The part I think is cool is the hybrid fuel system. Offhand I do not remember anyone here using a stock ECU to control stock-injector-based first-stage fueling (and using stock timing), and avoiding knock with second-stage ethanol fueling (via injectors, not a normal WI nozzle) using a separate ECU to control them. Logistics of ethanol aside, the approach has appeal. And, like he said, he can get ethanol easily and cheaply.

Why use shorter runners to reduce midrange VE, instead of just running lower boost in the midrange? The engine should make about the same max knock-limited torque...

However, if one's max torque in the midrange is knock-limited, the shorter runners would improve the top-end if it's near the limits of the turbine...

Oh, good lord...
Do you have any manner in which to measure cylinder pressure in relation to crank angle?

Not without expensive engine test cell equipment. I do not mean stuff that will be at your typical dyno shop.

I've been a huge proponent of electronic wastegate control for this exact reason. Set your boost target lower in the midrange and boost the crap out of it as RPMs climb and VE drops off. Linear torque curve, and best of all just like building the bionic man, we have the technology! It won't cost you a 1000 dollar piece of equipment(unless you count an Adaptronic :)) to do it either.

I figured maybe he has access to that type of equipment if he's doing 'knock studies'. I could be wrong though... :dunno:
The reason I ask is because it could reveal 'lost' power. Ie; the rods being compressed before or very shortly after tdc.

The dyno does the same thing with torque output anyway. Besides cylinder pressure is a non-existant load compared to inertial loading from RPM. And, having a cylinder pressure monitoring device might tell you when knock is likely, but when it actually occurs you just see a spike compared to a normal burn. Its unlikely the sensor would do anything for you necessarily better then a 30 dollar knock sensor.

I think you missed that the original "rods bending to adjust CR" comment was a joke about our stock rods being weak.

I had a looooooong discussion with a very knowledgeable buddy of mine that is in the gas engine R&D business. I wanted to try this very same thing for the same reason, and since I had an Adaptronic and an EBC, I had everything in place to try it. Just add dyno time.

He pointed out that a flat torque curve does not necessarily indicate constant cylinder pressure since other forces such as friction losses are increasing greatly at higher RPMs. Therefore, if you flatten your torque curve at high RPM with boost, you could actually have increasing cylinder pressures. What he suggested was using RPM, AFR, and injector duty cycle (all logged data) to approximate VE as RPM increases. This is assuming the boost curve is fairly flat at higher rpm, which mine was.

I thought this was a great idea and started data crunching. Everything I needed to know was in my data logs. When I ran the numbers, I found that the torque dropoff was not directly proportional to VE dropoff. In other words, the VE loss percentage was significantly less than the torque loss percentage. So, if I had added boost to flatten the torque curve, I could have actually had higher cylinder pressures at high RPM since the VE would have actually increased to offset the aforementioned friction and other losses.

Now the knock tolerance should be better since the piston speeds are higher at high RPM. There are about a dozen other things going on as well. So YMMV. If you want to do this, check your logs closely. I was going to write a post on this but never got around to it. So there it is.

Cool. BTDT, calculating VE from datalogs. Yup, it goes down more slowly than torque.

Definitely understood on the point of VE and RPM. From what I've seen knock tolerance is EXTREMELY high at the upper RPM ranges though. You might not be able to get a flat torque curve, but out past 5000 RPM I can see several more PSI of boost being added with no ill effects.

Its one of the reasons why i think big turbos actually "add" a lot of power. When the boost doesn't come on until 4500, and even then its a low amount of boost, you're not actually pushing cylinder pressures until piston speed is relatively very high! So people with massive turbos run massive amounts of boost with reasonable reliability.

Also from what I understand the knock indexes of ethanol fuels are more detonation resistant as RPM increases. So everyone knows that ethanol is better for performance, but given the knock threshold at 4000 RPM on ethanol and gasoline at 7000 RPM the ethanol fuel will drastically proportionally scale with better detonation resistance at the higher RPM.

An example:

at 4000 RPM on gasoline we run 10 psi of boost and 18 deg of timing

at 7000 RPM we can run 14psi of boost on gasoline at 18 deg of timing

Ratio of boost threshold as a function of RPM with gasoline: 1.4

at 4000 RPM On ethanol we run 16 psi of boost and 18 deg of timing

at 7000 RPM we can run 25psi of boost on ethanol at 18 deg of timing

Ratio of boost threshold as a function of RPM with etoh: 1.56

These are fictitious numbers, but you get the point.

This evidently has to do with the MON and RON numbers as well as the ability of ethanol to cool charges more quickly ( the dE in the phase change of ethanol is better placed to reduce knock then in gasoline)

As we make the switch to ethanol fuels I would think we will see more electronic waste-gate controllers because of this.

So you think the extra torque loss percentage was caused by friction? what else?

Also, can please someone explain why we can run more timing and boost before detonation at higher RPM- what other reasons are there than VE?

Actually I was just thinking along the lines of power left on the table moreso than detno. Y'know, the idea that the fuel burnt that's creating the higher pressures around tdc could be a little more useful a bit after tdc when we have a more favorable crank-angle.

Also, I take it that the 'speed of the burn' remains relatively constant regarding low vs high rpm? And is that why the fuels seem to be less knock-prone at higher crankspeeds? Or is something else going on?

The big thing is based around what Rattle said.

Flamefront, the actual burn edge expansion just like in a large forest fire(slower ofcourse!) is a fixed speed. As your piston moves faster you have to start the burn earlier to put the cylinder pressure at the right place so that you make maximum power.

I would think the variables in order that most influence power/detonation is:

Constant velocity flame front creating different cylinder pressures and temperatures as piston speed changes

The pinch (the area at the very edges of the piston that is covered by the head) creates different air mixing as piston speed changes. At higher piston speed turbulence goes up so I think you will see more homogenous cylinder mixtures that are probably more efficiently burning.

Volumetric changes as inertial tuning and cycle time change the way the cylinder is filled.

Cylinder charge cooling through the amount of time that is available to evaporate, and when the evaporation is occurring to cool the charge in the cylinder as cylinder pressure and temperature is going up.

Engines are more detonation resistant as RPM increases because there is less time for heat to transfer from the burned/burning gas region to the unburned gas region.

I thought it was simply, the longer the time the unburned gas sits, the greater the chance of spontaneous combustion.

Yea, its more a probability function I would think. You just need the right combination of AFR and heat. The longer it sits, the better it gets... Well sorta.. :X

ftfy

Adjusting boost pressures in relation to RPM the way you describe (TravisR), sounds a lot like a Rotrex...

*flame suit*