whats wrong with that power curve?

[Austrian Dodge]

friend of mine hit the dyno yesterday...company he bought the tune-up from promised 350 crank ponies

car is an audi s2 avant
2.2 liter 5 cylinder turbo
turbo is a late style watercooled KKK K26 (modified by MTM germany)
24PSI overboost @ 3500rpm then goes down to 15PSI

now check out what he got yesterday


290 crank HP and 480NM torque
i don't know exactly if the overboost setup is causing the rapid falling torque curve or a too small turbine housing.
how about the "waves" in the HP curve? maybe a little bit of detonation were timing is beeing pulled??

what do you think?

[tryingbe]

Take it back to the tune company and say you want those 60 hp back.

[Austrian Dodge]

thats what he's going to do anyways (has an appointment there in 2 weeks), but we still want to know what happened here

[tryingbe]

Without the a/f ratio, it's really hard to tell.

[Frank]

Without the a/f ratio, it's really hard to tell.

My thoughts exactly... or other dyno sheets from the same place.... it might be a fluid thing.

[Ondonti]

the general shape of the power curve seems awesome.
250+hp from 3500rpms to 7000 is a FUN car.

The k26 obviously is pushing EVERYTHING ITS GOT and is dropping of very badly.
k27 ftw 49lb/min woot.

BTW, its impossible for a k26 to make 350 crank hp.
The turbo is pushing maybe 25-30lb/min @ 3500 rpms, and it still only pushes that much air @ 6000-7000 rpms so the boost MUST drop.
The dyno you posted seems to be "crank" power estimates.

290 is about the max...and thats where your friend is at. If they think it can make 350hp then its no suprise to me that they cant manage to get the power curve very flat either . those dips dont seem to be very serious but Im sure the tune isnt "perfect" since in a mail order. As was said, other info is needed to break down these small 5hp differences.
But that company...they are either a bit clueless or they are liars.

[8valves]

Was this a chassis dyno or engine dyno?

If it's a chassis dyno, so these numbers were recorded to the wheels, then I see nothing wrong with their claims really. They're not that far off if any at all. I'll make the assumption it's an AWD car, and those trannies can suck a lot of power dependent upon design. The stock WRX tranny takes up 27%!! An STi is around 18%.

If it's an engine dyno, well... whoops.

[Ondonti]

If its a chassis dyno then those numbers are inflated.
k26 is a 26lb/hr turbo and @ 26lb/hr and the boost numbers he was running you are looking at 50% compressor efficiency. Not gonna make 290whp when you are pushing 26lb/min @ 50% efficiency.

In fact the sheet says "corrected" but I dont see a correction factor.

[8valves]

The only maps I see have a decimal for the airflow and I didn't look long enough to do a conversion. Did you find one out there or do a conversion?

I'd have to look at what's capable with different setups I've seen... I know there are some that are way right but still pound out some numbers, even if they could be way better on a nicer matched compressor.

Sounds like you've more than likely spotted the error though.

[Aries_Turbo]

some thoughts.....

the turbo was said to be "modified" by MTM.... it could possibly have a larger compressor wheel in it... something that should be found out.

so say the K26 has a larger compressor wheel in it and it is able to make 350 chp then the problem seems to lie in the boost control. ive seen my own car spike to a high number then fall to a lower one and hold that to redline cause of a controllers inability to hold the pressure back or something like a weak wastegate.

the dips could be the wastegate regulating pressure at that 15psi level and wavering a bit.

boost datalogs and a/f logs would paint a better picture.

good luck.

Brian

[Ondonti]

you guys can go play with Franks turbo calculator because he has that map or just look online. The KKK series turbo maps Ive seen show both m^3/sec on top and kg/sec on bottom. there is no real point using the calculator to "map" because we already know when boost is hitting and that the compressor is pretty much on the brink of its flow potencial. Just makes the conversions easy.
26lb/min = .179 m^3/sec which is at the edge of the 50% line on the compressor @ 2.0 bar.

Also, .179 m^3/sec (or 26lb/min) @24psi boost is WAY off the compressor chart. That turbo is OVERSPINNING like crazy.

There are two trims on the internet. The larger one (which i didnt look at before) has a bit more high boost flow ability (without overspining) but I think 2.6 bar and .179 m^3/sec is still going to be a bad idea as its still not quite in the compressor map and its way beyond 120k rpms which is the last rpm curve they give.

@ 2.0 bar (where this engine seems to be when it hits redline) the two versions of the k26 flow the same.

[Austrian Dodge]

it's indeed a chassis dyno and the car is AWD. dyno spits out crank HP.

the real problem is, that we don't know exactly what kind K26 this is (sure it's not stock, but MTM knows if its k26/k24/k27 or whatever )
wastegate doesn't leak - it's an external audi and the diaphram has been changed last week

thanks for that input guys, we'll do some testing today and i'll report later

[Rattlesnake]

I had a similar problem, and I find out that if I kept the boost steady at my desired level it was going to meet my spectations. So what I'm trying to say is that if he can hold the boost at 25psi and keep it there throughout the entire RPM range I bet you he'll pass the 350hp mark.

Reinaldo Moloon

[8valves]

it's indeed a chassis dyno and the car is AWD. dyno spits out crank HP.

Wierd! That would mean you would need to input a drivetrain loss, which you can't know for sure unless you dyno'd that car bone stock engine wise, and it would be on the same brakes, wheels, tires, etc as it was when it was stock.

Good luck with it.

[Aries_Turbo]

you dont get it.... what I was saying is that maybe they did something similar to stuffing a 50 trim T4 compressor in a T3 housing... I know for a fact that they make some garrett/KKK hybrids for Porsche 944's. I know a local guy that has one.

but if the boost controller is holding the pressure from the non leaking wastegate it does look like the turbo is too small. but if the turbo has been "played with" and is large enough to support 25psi to redline, then the fault might be in the boost control.

id remove the line from the wastegate and do a short pull to see where the boost will level out at. keep a firm eye on the gauge of course.

Brian

you guys can go play with Franks turbo calculator because he has that map or just look online. The KKK series turbo maps Ive seen show both m^3/sec on top and kg/sec on bottom. there is no real point using the calculator to "map" because we already know when boost is hitting and that the compressor is pretty much on the brink of its flow potencial. Just makes the conversions easy.
26lb/min = .179 m^3/sec which is at the edge of the 50% line on the compressor @ 2.0 bar.

Also, .179 m^3/sec (or 26lb/min) @24psi boost is WAY off the compressor chart. That turbo is OVERSPINNING like crazy.

There are two trims on the internet. The larger one (which i didnt look at before) has a bit more high boost flow ability (without overspining) but I think 2.6 bar and .179 m^3/sec is still going to be a bad idea as its still not quite in the compressor map and its way beyond 120k rpms which is the last rpm curve they give.

@ 2.0 bar (where this engine seems to be when it hits redline) the two versions of the k26 flow the same.

[Vigo]

if compressor efficiency is really at or below 50%, id fix that before unplugging the wastegate.

[Austrian Dodge]

i didn't understand why they did that overboost thing, i hate overboost

the main problem is the cal (plus the turbo, which is on its limit) imho...we didn't do anything to the car today, so guess we'll have to wait till after the meeting with the tuning company (which is a very respected company over here btw)