I suggested that but R/T Ernie is very particular about going from decel to WOT.

Btw I think hydra needs the height of the wheel for his calculations as well (base plate to root of exducer). You didn't provide it in you other post.

 

Now could you kindly measure the overall height of the turbine wheel as well?

The nut at the end of the exhaust wheel is only 15.5mm from tip to tip but the rounded part beneath it is 17.75mm, there's only 8mm from the face of the nut to the exducer blade root anyway

I assume you measured the blades at several spots like I did?

 

Gentlemen,
I measured blade thickness at 3-4 points along the length of the blades, going from inducer to exducer and ended up with something like 0.625mm +/- 0.025mm, difficult to be more exact than that. I think its reasonable to assume design similitude again in this case, i.e. that the blade thickness profile between both wheels is going to be similar. These are both turbine wheels with similar force/stress distributions, its not like we're comparing a donkey with a jellyfish here

With each passing post I'm more and more convinced that while the EFR wheel MAY be a hair LIGHTER (and even then don't hold your breath), I'm certain it will have greater rotational inertia than the GT30 turbine (which it dwarfs in size in comparison). Again, inertia is proportional to mass and to radius SQUARED so size is the dominant factor here...

Besides, If the EFR really did have less inertia, how would you explain these results:



Same car same dyno same everything similar hotsides. Note that the GTX3076 has a 1mm bigger compressor inducer as well which should work to the EFR's advantage...

Also look at how much more the EFR improves with hot headers compared to the GT30, surely this is due to its higher inertia

 

That isn't transient response.
10% throttle when in boost is already and under heavy load if held for a few seconds and is no measure at all of rate of change from closed throttle to open throttle.
If anyone wants to measure an engines recovery rate then they need to measure against time not rpm and using a very repeatable load not a fixed sweep rate.

 

These 2 attachments show what i'm referring to. This is logged evo data from a circuit race. The coloured lines are a gt35 and the black lines are a twin scroll gt40. You can see the rpm, throttle and boost pressures clearly as i've left the vertical cursor in the key areas. Driver has closed throttle then reapplied. I've aligned the 2 throttle traces each time. You can clearly see the twin scroll has a much faster recovery time. The gt40 is actually starting from very slightly lower rpm each time and is a bigger turbo with a heavier compressor wheel yet recovers its boost much quicker. So don't get hung up on turbine wheel moment of inertia because that is only a part of the picture when looking at transient response.

Edit: The time frame from close throttle to full boost is about 1 second give or take a bit in both plots.

 

What you just posted is a comparison of a smaller SS setup with a larger TS setup. We don't even know what hotsides were used, and we don't care either since its not a valid comparison, since the aim of this thread isn't to compare SS and TS. The graph I posted above compares two turbos on the same engine with the same turbokit, with similar hotsides and measured on the same dyno by the same person. Doesn't get much better than that as far as comparisons go. And while I realize that rotational inertia is only one of several variables that determine transient response, my whole point was to debunk the idea that the supposedly "lightweight" nature of the EFR wheels reduces rotational inertia compared to a similarly flowing GT wheel... Now this would definitely hold true if you could compare the 7670 with a GT35, or a 7064 with a GT30 but unfortunately this is not the case... Makes you wonder how MHI's TD05HRA turbos compare so favorably to the plain TD05HRs...

 

As the very first sentence of my post clearly states, i'm attempting to explain transient data from the racetrack where one setup has a clear transient advatange over the other. I did this because some replies to be misunderstanding the subject or even the reason for transient data. If you just want to compare dyno numbers then carry on as you were, it makes no difference to me. I was merely trying to offer some assistance.

 

Thats also internal wastegate vs. external wastegate. Not a fair comparison.

 

You are smarter than this. The gt's and gtx's are optimized for single scroll and the efr's are optimized for twinscroll. The turbine wheels dont typically do well at both.

 

You're right about the IWG vs EWG thing, although I don't think it matters all that much in this case (look at the shape of the respective curves before 15psi where both WGs are slammed shut) And you're absolutely right about the EFRs being more suited to TS and the "mismatched" GT turbines being optimised for SS...

Once again, the point I'm trying to make is that the larger gamma-Ti turbines don't have the claimed inertia advantage over the inconel GT turbines, and that any spool advantage they might have would likely be a result of a) them being better suited to twin scroll when used in that configuration , and b) higher turbine efficiency, especially at low expansion ratios due to the superback design turbine wheel. How much this is worth in practice I do not know...

 

FWIW,

EFR 58mm turbine in 0.65 A/R T2 is comparable to a Garrett GT25 in 0.64 T2
EFR 64mm turbine in 0.83 A/R T3 is comparable to a Garrett GT28 in 0.86 T2
EFR 70mm turbine in 0.83 A/R T3 is comparable to a Garrett GT30 in 0.82 T3
EFR 74mm turbine in 0.83 A/R T3 is comparable to a Garrett GT35 in 0.82 T3
EFR 80mm turbine in 1.05 A/R T4 is comparable to a Garrett GT37 in 1.11 T4

By comparable I mean not more than a 1lb/min difference in maximum corrected turbine flow

 

This is what I would want, if I understand correctly, that you could be at light throttle while in hold mode, then go WOT, and the whole thing would be recorded as the "run".

What I wouldn't want for my own use, is to be forced to be at WOT during hold mode so that the run starts from a stabilized WOT condition from whatever rpm. That is what I meant in the post above.

The other method, which sounds like starting the run from a decel - trailing throttle condition - I appreciate that this would be the easiest way to get repeatability if you don't have the luxury of logging turbo shaft speed. So it is a good way. It's just that I don't normally drive that way on the street. My normal idea of throttle response in street driving is response from a cruise condition to WOT. Anyway, sure, I'd be open to either of those 2 ways!
A video would be cool, I'd watch it for sure.

 

This is what I meant when I asked if the blades were constant thickness. I wasn't talking about variability between multiple examples of the "same" wheel.
So then I guess you are saying the blades are constant thickness, until you get close to the hub where they fillet into the hub.

 

The fact that there is this much debate on how best to compare the response of two setups indicates to me that the comparison would be so close it'd be hard to tell the difference in the real world.

 

Hydra if you read that entire post by Jeff Perrin you would've noticed the part where the EFR's transient response dominated the GTX, even though the GTX spooled up faster on the static dyno pull.

Boost versus throttle position. EFR7670 is Red, GTX3076 is green. You can see the EFR goes WOT well after the GTX, but responds MUCH faster.

without looking at the raw data it looks like the GTX takes ~2.20 seconds to respond and it takes the EFR ~1.6 seconds.

OH YEAH... AND THE EFR HERE IS A SINGLE SCROLL WITH AN INTERNAL GATE. The twin scroll EFR's transient response will be quite a bit better than what's shown here.