RomRaider

Yeah, that's correct.

Here's the formula I derived from the ideal gas law for doing this.

Let:
CFM be volumetric flow in cubic feet per minute;
MAF be mass air flow in grams per second;
IAT be intake air temperature in degrees Fahrenheit;
Baro be ambient air pressure in PSI;

then:
CFM=(MAF*((IAT*.37139)+170.7163))/(Baro*7.56)

I applied it to the example in the quoted link and it yielded 459.266 CFM. The value determined in the quoted link was 459.44 CFM, so I guess they're really the same formulas.

Hi, is the VF38 map available?

I just noticed a few discrepancies in the spreadsheet, at the top.

CID is given as 152.5 but Subaru says the displacement of the Legacy 2.5 GT engine is 2457cc / 149.94 ci. Also, "displacement" ignores combustion chamber volume and is simply the swept volume. The Legacy's 8.4 compression ratio implies each combustion chamber's volume at TDC is 83 cc. That means the total volume inside all four cylinders at BDC is 2789 cc / 170.2 ci. I think this is the better value to use as the basis for calculating VE.

I also see that Patm is given as 14.5 PSI whereas 1 atmosphere is usually given as 14.7 PSI.

Thanks for the true CID of the 2.5 liter engine. I couldn't find it so I put in 2500 cc. I figured everyone would put in their own numbers. I was more interested in setting up the scales onto the maps to make them easier to read.

The term, Patm, is actually the ambient pressure and not meant to be equal to 1 standard atmosphere. I used 14.5 as an arbitrary value that most people will see. Since the graphs are all plotted as a function of PR, it shouldn't change the shape but the boost pressure will be different.


EDIT: Yes, you are correct about using absolute volume in calculating VE. I'm going to have to update that. I'll probably have to use my own datalogs which would include the affect of modified AVCS curves and headers.

There's a reason the scale based on "atmospheres" was invented.
http://en.wikipedia.org/wiki/Earth's_atmosphere#Pressure_and_thickness

I've always seen volumetric efficiency calculated in reference to only the displacement volume of the engine, not including combustion chamber.

A Graham Bell in Forced Induction Performance, pages 90-91, uses engine displacement to calculate VE. Corky Bell in Maximum Boost, page 27.

I think that is the standard way of doing it. It's not critical. You could put in 5 liters for your 2.5L engine and it would just skew your VE numbers. But I think it is smarter to stick with the standard already in place.

I am not sure I understand your point. I know what 1 standard atmosphere is. I changed the term to Pambient in the spreadsheet so that people know that they need to type in their own pressure.

It won't change the shape of the plot, just the boost pressure values in the spreadsheet.

That's really cool.

Man... I need to get more sleep.

I thought about this on my way home and realized what you said... the VE and CID are used to relate back to the projected CFM if we had used the data from the datalog. We can arbitrarily pick a CID; the VE would be based upon that CID and the CFM data would be unchanged in the spreadsheet. So, as long as we use our own data and our own CID, it won't make a difference.

BTW, I updated the spreadsheet to include TD05H-16G Small and I moved the graphs around so they line up better when enlarged to 150%. EXCEL s*cks in that the visualization seems to float around a bit based upon the enlargement percentage.

imo, you need to use swept volume in your VE calculations, and swept volume amounts to displacement.

I added some Garrett compressor maps from http://www.turbobygarrett.com website.

A little bit OT... Garrett publishes turbine maps. Anyone know how to interpret them? I understand how A/R affects flow qualitatively but I'm not sure how to read an actual turbine map. The examples in Garrett's website show much lower peak exhaust CFM than the compressor flow. I don't understand how the corrected gas turbine flow (lb/min) can be so much lower than the intake flow (lb/min).



http://www.turbobygarrett.com/turbobyga ... 102.html#b

Once the turbine flow tops out, the rest should be flowing through the wastegate.

Thanks. I didn't realize how much flow comes through the wastegate.

I'm still not sure how to relate the compressor map to the turbine map. There doesn't seem to be enough data ... like wheel rpm.

Here's an example turbine map from Borg-Warner's website.



It looks like the Garrett maps are just approximations. The real turbine map would have rpm and efficiency curves.

I just started really looking at the numbers and have a question.

It appears that on a 2.5l engine, the RHF55 (VF39) can support a pretty high pressure ratio at pretty low RPMs. Staying under the curve I think I see that I could still get:

2.8 PR (26.5 psi) until 2250 RPM
2.6 PR (23.5 psi) until 3750 RPM
2.4 PR (20.6 psi) until 5000 RPM
2.2 PR (17.6 psi) until 5500 RPM
2.0 PR (14.7 psi) until 6000 RPM

I understand this is just the theoretical limit of what the turbo should be able to push, but the numbers are still higher than I've ever seen in use before. Can it really handle 26.5 psi and spool that fast?

Given the correct numbers for timing and everything else, would it be possible to use the above targets on just pump gas? Is there a glass ceiling where flow levels off even if pressure increases?

Justin