Some unexpected results, allowing the rotors to spin compared to having them fixed.

Going through some old files I came across this picture:

The first vendor claim that I tested, AWE’s S-Flo intake.

The product is a cone style air filter that was purported by AWE to have increased air flow rates through the MAF sensor by 25 grams-per-second.  That is an astonishing increase in airflow rate simply by changing from the stock airbox and flat panel filter to an open element cone type filter.  At the time of the AWE S-Flo’s introduction in the 2000 time period I hadn’t yet grasped that with the stock MAF sensor a grams/second change resulted in nearly a 1 for 1 wheel horsepower change.  That would have meant the S-Flo should have resulted in close to 25 additional wheel horsepower if it in fact did increase airflow rates by 25 g/s.

I had been reluctant to try the AWE S-Flo, a prior test performed by another Audiworld forum member on a K&N panel filter had shown no improvement over the stock air intake system.  Still, I was hopeful AWE was accurate in their claims for the product and when a Group Buy was started to give a discount price on the S-Flo I decided to give one a try.

As the buy group members began to receive their products I was slow to get around to installing mine.  Another forum member with a car setup just like mine had recorded some measurements of airflow.  An important point to remember is that back in the mid-2001 time frame all of the B5 S4’s were relatively uniform in terms of modifications, there simply was not all that much that could be done to the motor outside of a Stage 1 tune, exhaust muffler system, and intake filter.

The other forum member, “Black BiTurbo”, had his car setup just like mine, and he provided me with the VAG-COM readings from his car with the AWE S-Flo installed.  I recorded my car and compared our data.

The results shown above were a small surprise and a big disappointment with AWE Tuning.  There was no difference, using the S-Flo was no better than the stock air intake system.

I began to bring this point up with AWE representatives on the forum.  They said they would be dyno-ing an S4 with their S-Flo intake to prove the performance gains.  Months passed with no dyno test being conducted, then years went by, still nothing.  Eventually after making the same claims for over a decade AWE stopped selling the product, never having backed up their claims as they had stated they would.

If you look around you can still find other vendors peddling the AWE S-Flo with the same unsubstantiated claims, 

I ended up sending my unused AWE S-Flo intake back to the vendor for a refund.

A theory I’ve be thinking about lately is that the turbine side of the K04 hybrid turbochargers is the limiting factor in the amount of power these turbo’s can produce.

This theory has been accepted for some time, and seems quite reasonable given the results obtained with K04 and RS6 style turbochargers.  I started thinking more about the subject when it occurred to me that continuing to upsize compressor wheels, or incorporate more advanced designs, would not be a particularly worthwhile pursuit if no matter the improvement to the compressor the turbine side put a ceiling on the power capable of being developed.

The dilemma of this situation became more clearer to me when I began to compare boost profiles of FrankenTurbo F21 turbochargers against the results I had recorded with the FrankenTurbo F4H turbo.  The trend with the F21 has been to operate with boost pressure around 24-25 psi.  The F4H turbo, with a slightly smaller compressor wheel could achieve similar results, but with noticeably quicker spool up.

It seemed that the larger compressor wheel was slowing boost onset but apparently not resulting in any more top end power.  Maybe the turbine ceiling had been reached?

Flow testing turbines is not something commonly done outside of rebuilding Variable Nozzle Turbochargers (VNT) but I did find a research study done as part of a thesis at Ohio State University that incorporated a flow bench into turbine testing.

With a few ideas in mind about what I might do to check my own turbochargers I built an adapter sized like the bell-mouth entry to the downpipe to attach my BorgWarner K04 turbochargers to the flow bench.

I then fixed the compressor wheel so that it and the turbine would not spin.  I left the wastegate closed and ran the flow bench from zero up to 30 inches of H20 depression recording airflow through the turbine at several points along the way.

By itself the information shown above doesn’t say a whole lot. Shortly I’ll have my K03’s back from being rebuilt and at that time I will put them on the bench to gain a comparison dataset.  At some point in the future I may get my hands on an RS6 hotside to be able to measure how much of a difference there is at these low airflows.

This post has some follow on testing with the wastegate door open.