I had previously posted a graph of VE which compared my vtgs with MBailey's A28s (click here). There we saw that VE is related to LOAD which is reflection of torque. Of course what people are interested in besides max TQ & HP is the area under the curve (AUC). So I had some free time and calculated it for these two graphs. Basically, this is done by simulating a curve (thin black line) to fit the data obtained from the Durametric. One can then integrate the curve to find the area. When this was done my AUC was 70% of Mike's in the region between 2000-6680 rpm. His car is a 5.5s 60-130 and mine is a 6.9s. Moral of the story...I need to get some more AUC--LOL.

 

Brilliant, great explanation and some pretty fancy math there. Thanks for the post.

 

Great stuff, TTdude!

 

Thanks guys. I made a revised "upgrade" to the calcs. I added in another 50 units to my vtgs (lol) since between 2000-3000 rpms, the VE was about the same. This was initially not taken into consideration for the above integration curves. So now, I'm only 75% of MBailey's setup--LOL. Also, I looked at the difference between 2000-4000 rpm because obviously that's were the greatest difference occurs. For that region, the AUC for A28s is 376 and vtgs is 191, a 50% difference! The nice thing about this methodology is that one can compare different AUCs for different set-ups using data the Durametric spits out. Eliminates different dyno-dyno comparisons/wars.

 

Nice work Dave! Any info from other turbos that are popular? Does it depend on other mods? i.e. is MBailey's result typical or related to his particular application and other mods?

 

Hey Asher, I compared Mike's because his 60-130 run was a 4th gear pull like mine and he had a full set of good data that he was generous enough to share. Mike's awesome. If someone had bigger turbos then 4th gear may not be best for the 2000-4000 range. Perhaps 3rd gear would be better, although the fastest 60-130 time on record was done in 4th only. It would be interesting to compare that AUC. The right supporting mods can make a significant difference as Mike would tell you. The trick is getting the right supporting mods to match the set-up. But this method could elucidate the contribution of various mods after installation. So we could compare your .2ICs vs stock if you had durametric data for each.

 

Great Info Dave!! Glad you are still finding uses for the data. Integration is a distant bad memory for me!!
+Reps for you!
Thanks for posting!

 

Thanks Mike! Thank God for computers. Integration has never been so easy! Here's some more comparison data with Ken's (996ttalot) 2420g Tip.

Looks like you have the edge from from 2000-5100 (AUC=667 vs 641) and still within about 10% between 2000-6100 (AUC=908 vs 981).

 

He might have the edge for the moment but not in a couple of weeks

Just finished last week new build on my car
- full engine build with upgraded rods, gaskets, pump, rods etc
- custom garrett turbos 56lbs p/m flow with huge hot size
- custom headers with tial wgs etc etc.
- lots of other goodies

Ran the car at 1/2 throttle last week at 1.25 bar - through our check point on testing was nearly 35mph faster (130 v 95) when compared with old build on 20gs at 1.5 bar

We just stripped the engine back down to check to see if everything is okay. All is fine. Will build her back starting Friday and will be running next week.

Should hit some numbers in a few weeks but expecting to be 16-18 seconds to 300kph and very very low 5s (well I really think a 4.xx) 60-130.

We will see but I will datalog.

Will definitely have more AUC!!!!

Ken

 

That sounds awesome Ken. Good luck with the build.
Sorry Mike, you have some competition for low-end torque King.

 

Anxious to see all this in action Ken! Some pics would be great too!! Did you ever get better numbers from your 20g setup?

 

TTDude... what are the mods to your 911. I would have thought that VTG's would spool just as fast as 28's.

MBailey... what kind of WHP does your 911 make on Pump gas?

 

Yeah, Mike's set-up is pretty impressive. My numbers were generated with 63.5 mm vtgs by Proto, 5" ICs, and BMC filter.

 

Excellent!
If you haven't already - take a look at IBM's SPSS Statistical software - they have numerous components for multinomial regression studies available which would aid in this type of analysis
...another one to look at is Sigma Plot.

 

LOL. Thanks. I'll leave that up to you. Anything beyond Excel is too much for me. I know the fits aren't perfect but it's quick and dirty.