Hey Guys

As you may know, we out here in socal hada dyno day yesterday, and the results wil be posted by Tony (Hydrazine) soon (he's got all the data), and i will post by dyno sheet when i get it scanned.

As some of you know, I have been using the InnovateMotorsports modular data logging suite. A very big thanks to them for sponsoring me, which allows me to present to you this great suite of data.

The car today:
5AT G35 sedan
PE TT
AEBS sleeved (3.5L)/decked block
Arias 9.3:1 pistons, Pouter rods.
PE 510cc injectors
CJM FPR correction kit, and 1:1 PE FPR (short loop fuel return, ala APS)
Stock Fuel rails (AAM ones going in this week).
TechnoSquare reflash (specially developed/custom).

My data logging setup:
I was logging DUAL BANK AFR durign these runs. I have my primary LM-1 wideband (pink trace) sensor in bank 2 (p-side) downpipe, and a second LC-1 wideband controller and sensor in bank 1 (red trace, d-side). Both use identical Bosch wideband sensors (also used in all the Audi's and VW's) calibrated at the same time under the same conditions.

Here are some results that i logged during the dyno runs:




Look at the graphs for a while, noting the legend in the upper rt. hand corner. This is the overlay screen in LogWorks. In the first overlay, I compare 8.5 to 11 psi runs, while in the second, my 9.3 ans 12.3 psi runs.

First: I made 390/405 (hp/trq) at 8.5psi (which as you see tapers down as you go up the rpm band), while at 12.x psi, I made ~428/440 (hp/trq). The first run is very much in the ballpark, but my HP and torque curves drop off pretty rapidly on the to end. When I ppost the dyno, you will see what I mean. These three pieces of evidence, the hp drop off on top, boost taper, and a relatively small power increase between 8.5 and 12 psi (i should probably have been about 20 hp/torque units higher at 12psi, but that's just an educated guess) all end support to a restrictive Stillen Exhaust being the culprit. Tony (Hydrazine) has already done some very good scientific research into the Stillen exhaust, and you can ask him to comment more on it, but it is also fairly obvious that 400-500 hp and 8-12 psi of boost (and more in the future) cannot be suppoted by an exhaust that ultimately (in the muffler can) only has a single 2.5" opening.

Second: Note the difference in the bank one (red AFR trace) and bank two (pink AFR trace) as I get to the top end, especially on my 11 and 12 psi runs. The d-side bank consistently runs a tad (some times 'as much as' 0.5 AFR units) leaner than the p-side bank. What I am getting at, as you may have guessed, is whether this leaner condition on the d-side is caused by the reallythin fuel rail crossover to the d-side fuel rail from the p-side 'main' fuel rail?

Third: An intersting observation. At 8psi I already start to reach 5V on the MAF sensor (not logged here, no room) at 4200+ rpm. You can very well see how the fuel map is responding to this. The green "DWELL" trace is the injector duty cycle (taken from the #1 injector), and, BTW, it tops out at around 77% duty, right where I want it to be (maybe another 5% more or so would be safe). You can see that at some point the duty cycle curves for all the different boost runs converge, i.e. are not higher than the other on the count of load. Here is a chart from LogWorks that shows MAF voltage with resepct to RPM and Boost on a few runs that I made before dyno day (at 8 and 11 psi settings); here you can see that the MAF voltage can be slightly different at different load (boost) and identical RPM settings, that is below the pegged 4.9x V values (the limit of the AuxBox voltage recording).

This, combined with the DWELL traces above suggests a couple of things, at least provisionally until more data is available, at least to me:
1) The injector duy cycle tops out at just under 77% AS I reach (presumably subject to logging of MAF voltage and confirming then) 5V on the MAF output, i.e. AS i 'run out' of map. It's nice to see graphically that from the point where the two green traces converge, you can see the AFR get leaner compared to lower boost traces. WHile the MAF is still 'in range' (under ~5000rpm), you can still see the AFR being fairly constant while the injector duty cycle is greater at the higher boost setting, i.e. different parts of the map are being used to maintain the AFR. Don't mind the fact that the entire AFR is too rich, it is, i just need to lower the base fuel pressure to achieve this (it's been a little high after applying phunk's correction kit). A slightly higher than 1:1 rate FPR would probably allow me to keep a more constant AFR throughout the boost settings. Looking at the CarTech FPR for this (anyone know anything about this FPR?)
2) MAF Voltage of above 4.5V ARE in fact usable for load calculations by the ECU, even though an NA car, even that which is relatively well modified, sees no more than 4.5V at full load at redline. So, we do have the range of up to 5V instead of 4.5 to tune. I'll make some more runs logging MAF for this.

Next, i will be installing a fuel pressure transducers and logging FP before and after the fuel rails to note whether there is a drop pre-post injectors (on AAM rails only, will not have the time to do it fro the stock rails, although it would make it an arguably more useful experiment). That should be interesting as well

What do you guys think?

 

Someones been busy here. Good to see data logging results off a F/I set-up so everybody can put in their 2 cents on the results.....especially the more proficient tuners on the board. Thanks Gurgen.

 

Yes Gurgen you are surely correct. From the data you would expect a much larger increase in power for the extra PSI that are being put into the motor. I would expect closer to the mid to high 400's at 12 lbs. I am curious about the proposed exhaust restriction from a 2.5" exhaust. Maybe it is the baffle material in the cannister that is the problem. Maybe switching to a straight through exhaust like a Greddy Evo 2 will work better. No exhaust path restrictions. With regard to the A/F ratio's on the different sides of the car....it is a good possibilty that the rail cross over may be a restriction. I wonder if the stock manifold with the current throttle body position has something to do with it. The air does have to do a 90 degree curve to get to those intake runners on the driver side. Maybe a central t-body design above the radiator like on the TopSecret car would work better??

 

bump

 

Gurgen,

Even with the restrictive muffler your car was about to do back flips on the dyno!

There are several problems with muffler on the Stillen sedan exhaust system.
1) Its a true dual up to the muffler where both 2.5" pipes discharge into a common chamber. This alone causes V^2/2g velocity head total loss. Pressure drop.

2) The flow of each discharge plume is directed against a baffle where it must stop and make a 90' miter turn to the inward projecting inlet of a single 2.5" flow through channel. Pressure drop.

3) The flow of the chamber (at zero velocity head) must then flow into the inward projecting single 2.5" inlet at a discharge coefficent Cd of ~0.51. This is the real killer... At this Cd the effective diameter of the single flow through is 1.784"!!! Therefore the gasflow velocity is increased from zero to some high subsonic value (possibly sonic) with tremendous pressure loss.

4) The 2.5" pipe discharges into another chamber. This causes another V^2/2g velocity head total loss.... but because the central flow area is half that of the dual 2.5" pipes the pressure drop is 4X higher than the pressure drop of the first V^2/2g loss.

5) The flow of this discharge plume is directed against another baffle where it must make 90' miter turns to the 2 inward projecting inlets of two 2.5" flow through exhaust tips. More pressure drop.

6) The flow of the 2nd chamber (at zero velocity head) must then flow into the 2 inward projecting 2.5" inlets at a discharge coefficent Cd of ~0.51. This is isn't nearly as bad as the single inward projecting inlet but still causes significant pressure drop... At this Cd the effective diameter of the 2 flow through channels is ~2.6". The gasflow velocity is once again increased from zero to some subsonic value and into the atmosphere with significant pressure loss.

I've been around and heard enough rocket engines in operation to know what I heard comming out of your muffler was blasting out with such great force it was probably choked (sonic) at some point in your muffler. It can only drag down your HP and increase the load on both the engine and your turbo.

I'll bet you gain at least 20 HP with dual 2.5" straight through muffler change outs. And all the other details you mention in your post above will probably clear up.

Becareful when you change out the muffler. You may need to retune for all the additional flow!

You've already seen this but for anyone else with the Stillen G35 sedan exhaust you might be interested in the link.
http://g35driver.com/forums/showthre...t=56629&page=2



.

 

What a great analysis Tony, thanks very much. I know that I had discussed this with you several times, but it's a treat to have it all with the specifics that you included. I do think that I will get 20+ HP at my baseline boost, and even more at higher boost setting if my hypothesis above is correct (low power gain with increased boost).

 

Humm.. Seems you are doing very well with your tools.

Please give be a call - to discuss the potential for Tony DG's Project.

I Think Tony Dg will use a Greddy EU as last I spoke with him

The InnovateMotorsports logging unit seems to be superior.

Will catch you on the next time.

Cheers Amy -

 

Hi Amy,

I wouldn't say the Greddy EU and Innovate Motorsports are comparable. They are both completely different in function.

1) The Greddy EU is a controller.
2) The Innovate Motorsports LM-1 is a monitor/data logger.

So they actually complement eachother very, very well. Let the EU do the engine tuning and let the LM-1 show you what the effects are.

In fact, I will be getting both just for the same reasons.

 

what a great analysis, Tony. THanks a lot. I know I have discussed this with you more than once, but it's nice to hear the specifics. ttys

 

These are awesome logs, I totally expected to see these cars start to run out of MAF. Usually 1 to 4.3 is the sweet spot for MAF based vehicle. This is where these cars have a strong ability to count the air going into them. Once you are at the higher maf ranges I am pretty sure the MAF slope looks similar to this:



As you can see the top of the slop is very verticle. What that means is that when you are at the limit of the sensor it doesn't have a linear response. So that means you are putting more air into the car but the ecu is not making big enough changes to the injector on times to compensate. If for at one moment you peg the MAF 4.9v + and you don't have enough fuel added, you will lean out, cause the car to knock and detroy your motor. As many probably have found without even knowing it.

Our company TurboXS will have a solution! Stay Tuned!

Thanks,

Jermaine~