John Speake

There's no doubt that some more power and smoother running can be achieved with SharkTuning a stock spec car. Louie did some work on such a car some while ago, and I think the gains were about 10HP for the LH tune and another 10 or for the EZK tune.

An X pipe fitment also would benefit, but I don't have any figures for before and after.

Fitting headers definitely requires a SharkTune to get the best out of the potential of that setup.. in my experience. They seem to lean out the top end.

Louie928

Peter,
Judging from what I have seen a GT with X pipe will make about 315 rwhp. Sharktune and you should see about 10 more or around 325. Sharktuning will also help the low and mid range a lot so the help is felt more in torque in the RPM range normally used than in absolute top end numbers. These are just generalities since there are difference in engines, cam timing, and MAF aging that will affect it.

When I put the headers on this car and tried to Sharktune, I had a really hard time of it and saw no gain from the headers. From that, you could safely say that a Sharktune for an X pipe isn't correct for headers. I should add that the problem I had wasn't due to the Sharktuner, but other things. One was that the WOT switch quit working. The other was that I could not get much correlation between the dyno O2 sensor (mounted in the tailpipe) and the wide band (mounted in the X) that fed the Sharktuner. I didn't know which to believe. The car has an adjustable FPR and I thought for a while that it was changing fuel pressure on us. Finally, I ran out of time and gave up. Later, while doing some other work on this car, I noticed that the exhaust joint between the outlet of the X and the rear exhaust wasn't tight. I think that was admitting some air into the exhaust and messing up my tailpipe O2 readings. I've become quite sceptical of mixture readings taken at the tailpipe. The owner removed the intake and replaced the throttle switch as well as the idle valve and generally refurbished everything. This time, I only relied on the Sharktuner O2 sensor readings and the tuning session went really well.

The Sharktuner Autotune feature is helpful for me when going to larger injectors, but otherwise I don't use it much. I find that on Autotune, the map will have a large (either + or-) right next to small numbers. I don't think that is right for final tune. The LH seems to have some averaging function that takes the value of adjacent cell numbers into account to calculate the final pulse width. This would be especially important when operation is near a border between cells. What I'm trying to say is that you could get a map to work with a big number next to a small number because of the averaging function. However, if you manually smooth out the map numbers, you'll see less of the wild fluctuations in O2 % correction as you move around the map. I have manually edited those large number changes between cells and did Autotune again and that sometimes helps. Each Autotune session usually results in different cell numbers. I'd do several autotunes and try to average out the cell numbers between sessions and you should get pretty close. Especially watch where large changes occur between cells.

I spend hours driving on the (inertia) dyno moving around the map approaching cells from low RPM to higher RPM, from lower load to higher load, and vice versa. At lower RPM and loads, I'll use the car brake and throttle to get to a particular cell, at higher rpm/load, I'll use the dyno brake and throttle. I monitor the O2 correction % and try to keep corrections to around +/-3%. Pretty soon the map begins to smooth out. The map number differences between adjacent cells are usually less than 15. The entire map will be within +/- 50. At least that is the goal. I spend a lot more tuning time at idle and up to 2500 rpm than time above that. All this work probably makes no difference at all on peak hp or tq numbers, but I think it gives a smoother running engine. This is just the way I do it. I usually don't log anything when tuning mixture and closely watch what happens in real time. I also zero out all the WOT map values and tune WOT on the main map.

Tuning timing is a whole 'nother story.

AO

Nick-
Sadly, I regret to inform you that a band of thieves broke into my house and stole the headers you were storing in my basement... This appears to be the only thing they took. The local police are involved. Should they locate your property, I will inform you (... eventually). I am truly sorry for your loss.

BTW, the CF hood is still there.

Louie928

Yes, especially GT cams. Turn that GTS motor loose!

Louie928

Probably around 10hp +/-. Better torque though so I'd say it would feel like more than 10hp.

Louie928

I'd say that if you have an X pipe and no Sharktuning, then adding (Devek) headers would get you some gain. I didn't see the gain from headers on a car that was already sharktuned for the X pipe, but not sharktuned for headers. The Sharktuning difference with the headers was more than I thought it would be.

Peter F

Thanks Louie,

not sure where I got Tom from, probably reading in the text some where

/Peter

Louie928

Could you use a curved runner? Have the runner curve inward toward the valve as it enters the port. Then as it moves away from the port curve inward toward the center toward a common plenum. Seems like the 2v motors have at least a couple more inches of height between the head and underside of the hood. Might work.

If I understand correctly, you plan on using air from under the car to cool the oil? That under car air can get really hot. Not useful for cooling.

I'm ok with the idle. I have the idle set at 1000 rpm. It's not a smooth idle like a stock 928. There is a lot of burbling and babbling going on at the exhaust if that description means much. The engine is steady and it's not loping, or lumpy. It will idle lower rpm, but then there is not much rpm reduction tolerated before it dies. Below 700 rpm it's pretty much dead. The most annoying thing is trying to drive at a low speed under 30 mph. The engine does not tolerate low throttle opening/low power very well and it'll run rough. I think it is because of too much valve overlap. With high intake vacuum, when the intake valve opens it sucks back in some exhaust through the still open exhaust valve. Too much exhaust dilution and it stumbles. That's my theory. Open the throttle just a little and it smooths right out. It runs smoother in 5th gear and 1100 rpm than in 3rd or 4th and higher rpm, and more closed throttle and higher intake vacuum. The ITBs are supposed to help with this as compared to a single throttle, and I expect that they do, but still borderline in traffic. At 1mm valve lift, there is 22 degrees of overlap. At 0.5mm lift it has 34 deg overlap. A bit too much I think. I was able to help with the low power stumble by opening up the plug gap to 0.045". That made a noticeable difference and now it is tolerable in traffic down to 20 mph. 30 - 35 mph is fine.

Louie928

Rick,
I think the reason the collectors are 3" is so the primary pipes have an end; a discontinuity to generate the pulse travelling back up the pipe. A 2.5" collector may not be large enough plus the primary pipes would have to be squashed at the end to fit in a 2.5" collector. The MSDS headers are not anything close to equal length so I don't know if the tuning matters much or even may be an impediment at some RPMs. I just don't know. Your 3" to 3.5" Y should take care of any transition from the 3" collectors. Guess you'd need a short 3" piece of tubing to connect to the Y.

gjm

I used to drive 2.5 V6 Vauxhall (GM) fitted with a Jenvey set up; it looked like this:



This was the only way to fit throttle bodies and have them fit under the standard bonnet.

Please forgive me if I've misunderstood, but could something like this work?

I'm enjoying the discussions - thanks guys. (Greg and I have exchanged some emails - he's written, I've read and learnt, and then asked more questions...)

Graham

Louie928

Well, sort of. The 928 motor is constructed quite differently in that the cyl head ports are quite far apart. However, curved runners could be used to point inward from the ports to a common plenum or box with filtered air.

Louie928

Greg,
I measured a TB length (3.5") and air horn (2.75"). That gives a total length of 6.25". The adapters add about an inch average. The port is about 4". Looks like a total of 11.25". The air horns are bell mouthed which I understand makes their effective length shorter than their physical length. I did cut the air horn for cyl #1 a quarter inch shorter than the rest because that air horn top was quite close to the air box top. The power is no different with the box top on or off so apparently that air horn isn't smothered by proximity to the top.

drnick

john, interesting observation about lean running with headers.. ive got a dyno with the whole air fuel getting richer and richer! mind you this is with the map we had previously set up for the exhaust with louis X pipe.

andrew - will your insurance cover the £1000 i paid for those headers?

does anyone know if you can get at the clutch with the devek level two headers on the car??

Louie928

Yes, no problem getting to the clutch with Devek L2 headers on.

Glenn Evans

Greg was referring to the profile of the intake trumpet, not its cross-section.

An article in Race Engine Technology magazine tested several trumpet profiles (elliptical, circular, "aerofoil" and a plain pipe end) to determine which yielded the best mass flow and pressure recovery. The elliptical bell profile was found to be the best (the cross-section of the duct and bell mount was circular).

For a duct diameter, d, the best ellipse shape had a mouth of diameter 2.13d, and length (from the lip (ie normal to the direction of the duct) to where it necked down to the duct diameter) of d (hence Greg's description of "short and fat". There also was a radius outboard of the lip (leading edge) to smooth the flow of air being ingested from the side. I don't recall how much effect that radius was found to have.

In the article, the duct used for the tests was 23mm in diameter, so the mouth of the optimum elliptical trumpet profile was 49mm (2.13d) in diameter, length 23mm and the radius outboard of the lip was (I think) 3 mm (so 0.13d if I remember correctly).