Been experimentin'.

I spent a lot of time last year poring over all of the various NACA papers on the subject of WI that came out of the WWII-era research on piston aircraft engines. In a lot of those cases, the researchers found the optimum water:fuel ratio for best power to occur at what we might consider to be silly ratios, in the vicinity of 1:1.

To put this into perspective, a 1:1 mixture on my engine would be ~ 22 GPH at peak load.

Obviously it's hard to draw direct parallels between massive-displacement, low-speed, low-compression engines like those and what we've got, but in general I've just been playing around with different nozzles making subjective comparisons. 5 GPH is what happened to be on at the time.


It actually makes mine seem aggressive.

http://img37.imagefra.me/img/img37/6...5m_2e0c9ac.gif

That's the map I ended up with. Only the 184 & 212 rows got touched on the rollers, I wound up putting about 5 or 6 degrees into it overall. With this map, I got only one knock event, at a time when my intercooler was pretty well & truly heatsoaked. (The fan they had wasn't working very well.)



One thing that tends to bug me here is the apparently rather loose relationship between a lot of these dyno charts I see and basic physics. Allow me to explain.

We take an engine which, when it was brand new, was rated for 115 BHP. Throw a guess at drivetrain loss and call it 100 WHP even.

Now, make no internal mods, but double the manifold pressure. Power output should increase by a factor of 2, not a factor of 2.5.

Ok, so you can claim we're optimizing certain conditions, like running higher octane fuel. But on the other hand, we're placing a hell of a restriction directly on the exhaust ports. Goodbye, VE.

I think some of these correction factors that dyno operators apply might be a tad... optimistic. The chart I posted is raw data.



It's all about where you take the signal from. See how flat my boost curve is? That's with an MBC. I'd bet you a dollar that when Ben did the MBC pull in that chart, he was taking his wastegate signal from a point prior to the intercooler. That decrease in boost represents the pressure drop across the intercooler, which increases with rate of airflow.




:laugh:

A respectable amount of power - though I too would have expected a bit more - i.e. my dyno with a peak of 11.5 during spool up but holding 10.5 to redline gave 206 whp.. The big difference though is torque - mine was in the 175 range. Big difference there. The thing that throws me off is that this is your raw plot as in no correction factor - this is exactly what makes it difficult to compare dyno results with other guys, we need to all go to the same dyno mfg. and use the same correction factor ;) same elevation while we are at it too...
I suspect that using a correction factor roughly equivilant to what others used to create the plots they've gotten would yeild slightly higher numbers.

Actually, I seem to remember recently seeing a picture of your engine bay and it looked like the manifold was a bottom mount design. Is that still on the car? Certaintly not the Greddy manifold...
-Ryan

You sure you were looking at my engine?

http://img28.imagefra.me/img/img28/2...rm_2920f4b.jpg

Pretty sure that's a top-mount.

Yowza, that's a lot of timing. Have you checked if you have any hardware latency issues as RPM increases?

Yes, I re-checked that after I built my 9924-based VR decoder. Latency is very nearly immeasurable, which I attribute to having almost no capacitive filtering on the new circuit. I think I have it set to about 20μs or so.

I actually forgot to do this initially after installing the new circuit, and it took me a day or so to figure out why I was getting light knock all of a sudden every time I went into boost. I'd previously calibrated it for the old circuit, which was pretty laggy.

Sort of the two halves of my pulley deflecting as a function of RPM (yet remaining perfectly aligned when measured statically) I'm pretty sure that map is reflective of reality.

The miracle of WI, I guess.

I just realized something- those runs were made in third gear. Remember when I said that the chart was uncorrected? That includes uncorrected for drivetrain loss.

Slaps self on forehead...

on a dynojet, gear doesn't matter, since you are just pushing a giant concrete roller and it calculates gearing through the software based on the rate of acceleration. I've run back to back in 3rd and 4th and seen no difference in output.

jesus .... do you have thirty fog lights mounted to the nose?


-Zach

Joe you have the same Rad as me...so that makes you Rad.

In 3rd, there should in theory be some frictional losses between the main and driven gears on the input shaft and countershaft, and between the third gearset on the countershaft and mainshaft. In fourth, the input shaft and main shafts are simply coupled, so no power is transmitted through any meshed gears. I'll bet you a dollar that you could see a difference between back-to-back runs.



I don't have any fog lights or other aux lights at all. Why do you ask?



It makes us both chepaskates is what it does. :D

All the loom running over the core support. Aside from the airbag sensor I don't know what the hell you'd have going on up there.

-Zach

Ah, lemme see if I can remember it all.

There's a wire that picks up +12 from the blue aux connector to drive some relays on the other side, a pair of wires for EBC that never got implemented, a pair of wires to drive the solenoid valve that bypasses the MBC if WI isn't on, and... I think there's something else but I can't recall off the top of my head.