except wrt the G40 comments, you have to consider response depending on the end goal. Even with the smallest A/R turbine housing you definitely don’t want a G40 turbo on an autocross course. At last not on a national competition level. So there are other numbers to consider as well, not just compressor efficiency.
.

 

The fun thing I was trying to accomplish was to drive compressor map points in Matchbot using the Garrett turbine map curve instead of a BW curve since none would apply. What I tried was to iterate Matchbot "Turbine Expansion Ratio" to generate "Turbine Corrected Flow" and "Exhaust Manifold Pressure" (converted to Pressure Ratio in excel using EMP+14.7/14.7) points that land on the Garrett Turbine map. Turbine data points don't line up to any one BW turbine curve, but they line up on the Garrett map in Excel. I did guess on turbine efficiency since a curve is not provided. What seems interesting or maybe wrong is that the data indicates that the G35 1.01a/r Turbine functions similar to a BW 74mm 1.45a/r, but at low expansion ratios performance is similar to a BW 80mm 1.05 a/r. I would have expected the opposite trend from latest gen turbine aero, but that may have been part of the tradeoff accepted for the long skinny lower inertia design.

Other sources of error could be from the numbers I used for Volumetric Efficiency, EGT, and BSFC which were guesses not derived from real world data.

 

edit: let me double check when I have more time than just a glance

 

sorry, I can’t delete a post at all or edit it after 24 hrs or so on this forum

I didn’t get into evaluating the numbers you used for the overall calculations, but just focusing on the turbine plot values you didn’t do too bad actually. So kudos on that. However, they should all be plotted on the same flow plot line if it’s the same turbine housing and that line is going to approx. be about half way between the 74mm 1.05 and 74mm 1.45 flow lines, which are 8374 housings. The 76mm reference is for the smaller S3xx Airwerks turbine housing line.

.

 

ok, so remembered that you posted the link to the Matchbot info, and I took a few minutes just to adjust the turbine expansion ratios only to bring them inline to where they should be imo. Again I didn’t assess or adjust anything else, just the turbine expansion adjustments and this is what it yields:



.

Note the plot points #2 and #5 are on top of each other relative to this program running three points at one boost setting (#s 1, 3, 5) and the other three at a different higher boost setting (#s 2, 4, 6)

So also note that at point 6 the projected hp is approx 451 @ ~60 lbs/min corrected compressor flow and the projected corrected turbine flow is estimated 29.4 lbs/min and 2.24 PR with the emap being about 3 psi higher than boost (~1.18).

Which my inital thought was when you posted them up is that the emap was a bit lower than expected and was borne out above. All these numbers translate directly to various comments that have been made across a number of threads here relating hp to a turbine flow expectation and then sizing the turbine housing A/R accordingly.

So if this was for a G40-900 rather than a G35-900 it should be obvious how even the smallest 0.84 A/R turbine housing having a peak flow potential of 35 lbs/min is going to be oversized for this power level. Such that it will likely end up with a positive emap value and of course the slower response that comes with it. Which again is a carryover from previous comments/discussions

The other point I had forgotten to make in a previous comment is that the space to install the turbo also needs consideration. While the G35-1050 and G40-900 compressor maps look nice wrt efficiencies, their larger dimensions make optimum fitment more difficult. It should also be apparent wrt my previous comment about not wanting a G40-900 at this power level for a higher response application such as an autocross.

Which relative to autocross, the 450 - 500 whp range becomes a limiting factor for traction when good response is also factored in. So if you can’t buy a tight enough turbine housing to get that response on a G40 then you have to be looking at a G35 or G30 instead.
.

 

Here's what I get for G35-900 with the 0.83a/r. Boost ~400 rpm earlier. Turbine map translates to be like a BW 74mm 1.05a/r, but a little more flow at low PR and less at higher PRs. Is there any estimated comparison available on the inertia difference between an efr 8374 and a G35-900?
https://www.borgwarner.com/go/06AMIF

 

you don’t seem to be plotting the compressor map values properly. The plot line should be horizontal or actually slightly upward slope to account for increasing backpressure as the flow values increase through the intake system, not the sloping vertical lines being indicated, based on the compressor map PR values and corrected compressor flow values shown in the calculation sheet.

#1 - 1.71 PR @ 16.9 lbs/min
#3 - 1.72 PR @ 36.8 lbs/min
#5 - 1.75 PR @ 47.2 lbs/min

#2 - 2.22 PR @ 23.9 lbs/min
#4 - 2.24 PR @ 46.7 lbs/min
#6 - 2.21 PR @ 60.1 lbs/min

which suggests that the intake filter, intercooler, etc. pressure values might be off, because the intake losses should be increasing as flow/boost increases.

plotting out the Garrett values is a bit of a manual iterative process to dial them in to match. I’ll try to find the time to tackle it this weekend.
,

 

OK, G35-900 0.83A/R with 17psi alone this time. https://www.borgwarner.com/go/XDS4GN

 

G30-900 w 1.01ar overlaps the points generated for the G35-900 .83a/r almost exactly. Couldn't find the divided 1.06 a/r map, but it's probably close enough to demonstrate the comparison.

 

G30-770 1.01/R.

 

I bought the G30-770 1.06 T4 for 20 psi on E85 due to limited in fitment space and also figuring the smaller turbine impeller will spool a little better than the G35

the other thing you might be overlooking is the G35 0.83 by Garrett is an open scroll V-band inlet and that’s going to cost some spool performance below 5000 rpm imo. Pulsar does offer a 0.85 A/R div T4 cast housing that is a direct fit though. I think they might be plating them in nickel or something else, but need to confirm.

Because I bought a Pulsar turbine housing in an A/R that Garrett doesn't have and initially thought it was stainless rather than cast because of the appearance. It’s not like any cast iron housing I ever saw before. If it’s not plated then possibly it has an alloy additive. Not sure yet.

I generally put the G30 1.06 housing at 29-30 lbs/min peak