On an engine dyno like Porsche use, with proper intercooler cooling, my engine made a maximum of 522PS DIN which is about 515HP SAE (I use PS and HP interchangeably although 1 PS = about 0.987HP). . It was over 500PS from 4750rpm to 6850rpm. Following a recent service I decided to get a chassis dyno reading from a local shop. I had a chassis dyno run when the engine was stock in ’97 (atmospheric conditions were almost identical)
The number which it would seem most 993tt Rennlisters are interested in is 463RWHP.
Now before everyone starts adding 20% or whatever, remember it is producing around 520HP at the Flywheel under DIN conditions.
The dyno in question, a Bosch FLA-203 is used by lots of tuners in Europe and gives corrected Flywheel figures which are reliable on Nat Asp cars.
The corrected flywheel PS it gave me was 507.
For comparison, the stock figures from ’97 were 356 RWHP and 407HP corrected Why “only” 507HP ? –well I reckon it is down to the fallibility of testing 993 turbo cars on chassis dynos. I have an intake temperature guage which whilst not very fast acting showed a max of 97deg F during the run with an ambient temp of around 58 deg F. I have spent plenty of time watching the guage on the road (yes its sad !) and typically on a 58 deg day it would not go higher than 80deg F on a full boost run. Herein lies the problem since the chassis dyno only corrects for ambient temp and thinks the engine is sucking in 58degF air, not 97+. It corrects the RWHP figure by a factor based on the DIN standard which is 68degF, so on this day it probably reduces the corrected figure. Apart from the dyno not "appreciating" the high intake temp, the Motronic will be doing its thing with the timing and boost to further complicate a meanigful reading.
Something else I noticed in the “correction” process is that whilst the RWHP figure may drop at higher revs, the FWHP figure may increase eg if at 5000rpm the RWHP fig is 373hp and the corrected fig is 403HP, at 5990rpm the RWHP fig is only 356HP but the corrected is 407HP –so RWHP advocates (and I know where they are coming from generally) may not realize that their motor is actually producing more power further up the rev band which is being robbed from the wheels by a combination of heat and friction.
I am interested in any solutions or compromises anyone has come up with for maintaining post turbo intake temperatures on our cars at realistic “real road” levels whilst on chassis dynos ?

 

I would not get to hung up on it. As you know from driving your car performance does vary with conditions. Also don't forget the dyno has a fudge factor because all variables cannot be accounted for.

Some things that will affect a chassis dyno are, software calibration, tire pressure, wheel aligment, tire size, gear lube type & temp etc...

The chassis dyno is best used for evaluating performance upgrades when compaired to previous runs on the same dyno not for trying to obtain accuate flywheel HP.

 

Madness
I wonder just how good chassis dynos are at evaluating upgrades or rather the effect of the upgrade on the real road performance because so much of the power variations are down to the varying temperatures (particularly that inlet temp). An obvious example for me is my Gruppe M cold air inlet which must help in getting cooler (than stock airbox) air into the motor yet will not show any improvement on the CD. How about turbos, we have had discussions on K16 hybrids v K24s - if the k16 hybrids are running outside their design envelope at higher revs on the dyno compared with a cooler running K24 at the same revs, will the extra heat skew the figures further 'cos of Motronic intervention and hotter inlet temps ? the same may happen on the road but the degree of Motronic intervention may be pro rata less (vis a vis the K24 set up) due to better cooling - difficult to measure !?

 

I have had great success using them. I will always dyno the car in the AM and discuss with the operator if they have had it recalibrated since the last time I was there. Temps don't vary much around here so its not hard to keep test conditions within a resonable range. I also check tire pressures and adjust to specs to help remove that variable. Instead of focusing in on the peak numbers I look at how soon does power fall off? How does boost climb? Where is peak torque. Then I decide if there is room for improvement.

I don't get hung up on a few HP here or there.

Although not ideal the chassis dyno is the best tool at our disposal. The dyno does not play favorates.

 

It would be interesting to see what happens to the intake air temp on the chassis dyno when a water mist is sprayed onto the intercooler...

 

Better yet is to inject water into the intake track. This has proved very successful on detonation prone turbo cars running pump gas. You can revover 3-5 oct points using a properly designed water injection system. I happen to like the aquamist 2D system. It uses accumulaters, 100 psi pump and uses the fuel injector pulse signal to pulse the water nozzles so to keep ratios even.

 

K24madness, you have a PM.

 

madness
A friend of mine who was involved in developing a 993 GT2 EVO race car (RS Tuning, Motronic controlled) over 600HP attempted to use the aquamist system, had quite a lot of manufacturer help as I recall, it did absolutely ziltch for the power - the aquamist people were baffled - but it didn't work ! May be something to do with the Motronic, or maybe it was the chassis dyno they were measuring it on
phelix
You would think that water mist sprays (like on the Imprezas) would help on the dyno as they do on the road. I think it was Kevin (turbo expert) who was talking about the CO2 mist sprays If you look back through the archives, cjv was experimenting (on his 996tt) with a metal bulb filled with coolant CO2 or some other gas, which was in the inlet path in the throttle housing. I think it actually robbed power on the dyno, but very ingenious

 

I was coming at the water mist from the point of view of trying to provide an IAT on the dyno that's seen on the road. Adding water mist to the intake air is another can of worms. And a CO2 bulb in the intake stream is yet another completely different can of worms. I wonder what that does to air flow? You would think that it would need a fair surface area to have any effect.

If I'm reading my Bosch Automotive Handbook right every 5 degrees C of intake air temp above 20 is a power loss of approximately 1%. This of course assumes the IAT isn't high enough for the ECU to cut back on boost or spark advance. So adding 3% to your 507 brings it to 522 (!!).

 

phelix
thanks for the formula, I was trying to find it - I seemed to remember reading somewhere that every 2deg f increase cost 1 hp this works out the same 521.5.
I suppose the IAT at which the Motronic will interven on spark advance and boost will vary according to the programming.
Just to throw some more facts and figures on the subject, I have an RS engine dyno readout which shows that at an ambient temp of 30degC my engine actually measured at 486 flywheel hp which was corrected for DIN to 522 Flywheel PS - I am not certain of exactly what volume of
air is driven through the intercoolers so the above whlist interesting is somewhat meaningless

 

There are a couple of things I'd like to point out here. Standard conditions are Atmospheric pressure at seal level of 14.7psi and Temperature at sea level of 60 degrees F. The correction factor the dyno provides is to correct the actual room air temp and pressure to that of standard conditions. The correction has nothing to do with inlet temps or pressures. An example is that an engine tested at sea level will produce more power than an engine at 1 mile in elevation because of the amount of oxygen in the air from temp and pressure is less than at sea level.

Air density is a formula of 2.7 * pressure / temp.

So, density at sea level in absolute conditions is 2.7(14.7)/(460+60) = .076lbs of air per cubic foot.

As temp increase and/or pressure decreases, air density decreases which means there is less oxygen to mix with the fuel and combust, so power is down.

I use chassis dynos almost exclusively and have found that no matter how well I cool the intercoolers, fan, and air intake, they always run richer on the dyno than in real world conditions. This means to me that they are making more power in real world conditions than on the dyno because of temp issues.

 

I'll also add that I have found that for every 10 degrees C of temp change above 20 degrees C I need 2-3% less fuel to maintain the same Air Fuel Ratio. This is critical when putting together an Air Temp compensation table for a turbo car.

 

Your engine dyno showed 518hp and your rear wheel chassis dyno showed 507hp. I don't see where your concern is as to accuracy of the dynos??? The difference you are looking at is 11hp or ~2% margin of error. Even the dyno itself could account for that small variance.

Perhaps I have read something wrong in the original note or it was tongue in cheek? I am actually quite surprised you got the two results so close! I only use chassis dynos for comparative purposes and I don't care about +/- 10hp because turbo cars are so variable in that way anyways.

 

Something interesting to add, here are some 944 Turbo S dyno plots that I did on a marathon dyno session. We did so many runs, I was completely drained at the end of the day. We would see a 10hp difference from one run to the next with no changes in the car.

http://www.novustelecom.net/~fkassam/dyno/

 

fc
I suppose you are right I am looking at getting more power. 30 more PS is looking like costing me around US$18K equivalent - I guess in this light I am a bit more concerned than others about "missing" HP