He's not relating compressor flow to turbine flow. he's listing max flow for each compressor and turbine separately. I believe the point is that the mass flow of the components is a more valid comparison than their physical dimensions.

 

What does that mean to post max flow for compressor and turbine separately?

If you are going to post efficiency numbers on the compressor side for comparative purposes, which are based on shaft speed(see graph below), and then post comparative turbine flow numbers, those turbine flow numbers should be based on the same shaft speed. Otherwise you are comparing apples to oranges. Again, because that information is not published you only have a fraction of the information required and have no way to make real comparisons amongst these different turbochargers.

 

Which is why testing would be awesome, none of us know what the outcome will be until that happens. I all but referenced your earlier comparisons between EFR7670 and GTX3076R when people were saying the GTX35 76R is more comparable to the EFR btw.

At the end of the day my bet is the EFR is going to be the better unit, winning out especially in transient response though actual dyno figures for the GTX3076R will be more comparable than people stuck on the 10mm OD difference might expect in terms of peak power numbers.

 

240z, I only brought efficiency into this for the compressor, simply so it provides a common limit that can be compared between both. Just as the EFR maps go out passed 60%, the GTX wheels will do the same, even if it's not on paper. I just put a quantitative limit on efficiency for comparison purposes.

Agreed, the match between compressor and turbine is very important and you really can't figure it out based on the available data.

If you look at the physical dimensions though, the 10mm smaller Garrett wheel seems like there is no way it could keep up with the EFR wheel. Reality though, at 3:1 and higher pressure ratios, it's within 1 lb/min of the EFR despite being 10mm smaller.

While BW went to high end materials to minimize inertia, Garrett seems to have gone after maximizing flow for a given wheel size, there by reducing inertia by allow the use of a smaller wheel. Now, yes, that means there is the potentially for efficiency gains with the EFR due to blade tip speed match, higher expansion efficiency due to the flow being in contact with the blades for longer, etc.

As a whole, the larger diameter TiAl wheel should be the winner on overall system efficiency. But efficiency doesn't always win the race...

 

It is just a friendly debate and nothing more. See how fun it can be to bench race these turbochargers.

 

bingo. we need to test it. If you have data on the 3071, 3076, and 3576 compared to the EFR... you can find out where the EFR stands.

Now that we're all in agreement that we need to test these turbos. Any inputs on what additional sensors (backpressure) might be very telling of the data? Any suggestion as to what boost pressures should be tested? etc.

 

I would like to see the RPM sensors for all the turbos but adding that to the GT turbos would not be easy.

Backpressure would be nice to see along with EGT, should not change much in theory but still might be interesting. I assume that IAT's are already a given, a second IAT sensor pre-intercooler would make for some possibly interesting data as well.

I would like to see at least 2 boost pressures. The first around ~18psi and then other at ~30psi. A last test with them in "kill mode" would be interesting as well.

As has been pointed out though, the biggest thing to test is the transition response. This could be done via datalog but I think you might also get some interesting data from a dynojet in Time mode.

Pick a few RPM's, coast down to them and then floor it and see how the power looks as a factor of time vs the normal RPM or speed. If the dyno had the boost logger add-on that would be even better. This would net the most interesting results IMHO and also be directly comparable to the different turbos in a visual format.

Really hard to know what things to test until you start the testing, you never know what you might find until you do it.

I really want to see the 8374 vs the 6266 and the 9180 vs the 6466 or 6766.

 

The bigger turbo's full song would be more than taxing on my 2.2L... they'd probably put a big hole in it. I could definitely run the 8374 vs. the 82-84mm competitors, but you wouldn't see me push them past 650whp... which probably isn't interesting to the folks who want to buy these turbos. I could however do transient response testing... which would be interesting to see the trend line form with regards to transient response of the Ti wheels vs. diameter against the steel alloy wheels vs. diameter. I think the spread will widen as the wheel diameter gets larger in favor of the EFR. And if Brock at BW is right about it (which because he's already tested it... he probably is) that WILL be the case.

 

If you have a way to record something that would show surge during or right after spoolup, I'd be very interested in that.
Reason being, I've flogged that Garrett Boost Advisor thing several times trying to get it to recommend a GTX3071 for my moderate setup, and it's not doing it very enthusiastically. Usually my "Mid Range" point falls to the left of the surge line. The GTX compressor maps are narrow, especially at high PR's like 3 or more. The BW's are much better on this.
Anyway, if there is a way to show whether this is really happening - basically, surge in the high torque area around 4000 - 5000 rpm. Maybe this would only show up in a high gear where the acceleration rate is less so there is more time for the surge to develop.

 

I could just drive the car in 5th gear on the freeway and it'll do it. My OG Fp Red, and HKS turbo surged.

 

Most turbochargers that spool well usually end up with surge lugging them in 5th gear. It's the nature of the beast. I haven’t heard of guys reporting surge problems on the GTX3071R, but there aren’t a ton out there and those that are might have various reasons to not mention surge issues (shop owner types).

The EFR wheels definitely have an advantage there though on the compressor maps.

 

I kind of agree. Just a little.

 

Yeah I guess that's what it would boil down to. If you can arrive at a comparison between the different turbos on this test, some kind of rating or ranking, that would be super cool.

That's for sure. Here in these 2 pics are the results I got from Garrett boost advisor, asking for 600hp at the crankshaft. "Mid range" rpm is 5000, "Max power" rpm is 7500, race gas, cool western Washington weather conditions, GTX3071. The left red dot (mid range point) is quite off the map. The map is zoomed which is why you can't see the vert or horiz scales. But the pressure ratio and airflow numbers are given in the accompanying chart.
If you put these same 2 points on a EFR7670 map, the "mid" point is very comfortably on the map at better than 70% efficiency.

[IMG][/IMG]


[IMG][/IMG]

 

ha... funny that garrett's GTX wheels now have BW extended tip geometry.
also funny that garrett has a "boost advisor" to keep up with BW's "matchbot"

just wanted to point that out :/

 

Yep it's kinda funny.
I guess my thought about the charts I posted above is, my "mid" point at 5000 rpm that's in surge territory - fer crum sake this is not exactly lugging the engine. This is not exactly what most people mean when they talk about surging in 5th gear when accelerating from normal highway speeds. It's a bugger that a turbo significantly smaller than the EFR7670 should be in surge at a point that should be a very useful part of the rev band.