Lithium

If you are going to run an RB30 to 8000rpm then I'd recommend going a bit bigger than an 8374 - it would be ridiculously torquey on an RB30 but the EFR9180 would guarantee no restriction at the higher rpm, and it will not be laggy. Having the headroom to make near 1000hp on E85 would make an unbelieveable all arounder.

Yeah most definitely not the way I'd go either.

Healthy! Nice work Be interesting to see a trap speed for it!

I think its not an EFR turbo, and I suspect its probably a Bulls Eye power one with that naming convention. Its probably for a different thread to be honest. EFR9180 would be a good alternative to that, and go very well on a built RB30

GT2835s make more peak torque than GT3076Rs? I'm not convinced, if all things are equal

Owner swapped from .82 GT3582 to internal gate .83 EFR8374. Gained power pretty much everywhere on same boost, pretty sure results were posted in this thread.

You seem to be picking comparisons/expectations which handicap the BW. Why would a 7670 have to outspool a GTX3071R to be worthy, when its a fair bit bigger turbo? We can't comment too much here, but I am interested to see if the impression one would get from the maps is different from actually proves to happen when the EFR7670 gets on the dyno for comparison with GTX3076R. In terms of EFR8374 vs GT3582R, the 8374s have been found to spool similar or better despite being capable of more power. If you think that they aren't a fair comparison then isn't that just stating that the GTX is out of its league? A T04Z would get destroyed in terms of spool against an 8374

I suspect a lot of it is to do with the fact its not just a new billet compressor wheel of some description shoved onto the same old housings, core and turbine wheel. Here we have a fully re-designed unit, EVERYTHING is designed with raising the bar and imho it shows at least on paper. It will be interesting to see what happens when more people get to tune and drive setups running these turbos to see how good it really is. Initially I thought Garrett were pretty cunning bringing out the GTX turbos just as the EFR hype started winding up - but it could end up backfiring if their best counter turns out to perform barely any different to existing units on the market.

The trick is these aren't a single trick pony, even if an EFR7670 flows no morebelow crazy boost levels (the S2000 with the GT3582R vs EFR8374 on 1bar still made more power everywhere with the EFR), its going to provide less exhaust resistance by having less weight and a pretty decent flowing turbine wheel - more response and less exhaust manifold pressure on an equivalent setup sounds good. Then you get into the fact that a twin scroll setup will be so easy to get right with one, having a nicely formed wastegate flow back into the main exhaust stream sorted by the experts, no need to get or place a BOV or wastegates etc all work out as a mass of +s. I await the STI results before I get TOOOO excited (truth be told I am already fizzing at the bung) but I reckon even at the boost level I am running my car at, I'd end up with for all intents and purposes a faster car going to a TS EFR7670.... if I decided to be a nutcase and push 1.8bar through it then, who knows.

People should stop perceiving the billet thing as whats responsible for the efficiencies being at higher pressure ratios, GTXRs if they get enough buyer volume will end up with cast compressor wheels - no doubt. I do feel that a lot of these things are just changing their bias to higher boost levels instead of evenly distributed flow, meaning that they're best suited for ~2litre engines - or two of them on a bigger engine. They still have an advantage over the normal compressors, its just at high boost do they really shine.

There should be results in the next day or so of an STI which first had a .82a/r GT3076R, then replaced it with a .82a/r GTX3076R, and now afaik has a .83 EFR7670 going on it Should be VERY interesting to compare!

A lad over here is putting an EFR8374 on his RB30/26 GTR sometime in the pretty near future, I will update thread with results - I can see it being a couple of months yet just so I don't get spammed with queries on when I am going to update

I'd quite happily rock one, have no intention of stock mounting it anyway so clearance should be fine

My opinion too, definitely the safest bet.

I explained it. I'll try and do it slower/clearer - even though you almost answered your own question. When anyone who has half a clue upgrades their turbo, they retune their car to suit - right? So if you are going to have a setup which alternates between a bigger or smaller turbo and there is no way the ECU knows exactly which point in the transation the flapper is at, what tune should it use? FFS, I forgot why I stopped partaking in tech threads in forums for a while. I'm not going to bother with my huge speel now - thanks for saving me some time. To make what I said make sense, read the thread - specifically the posts by Rockabilly I was referring to. Its all here. Strange how people always made their Mazdas more reliable by binning the stock turbo setups.

Considering this is a thread about fancy new turbos, there isn't enough turbo pr0n so I thought I'd save it with some scenary:

To be fair, someone has squeezed out around 900hp @ hubs with a GTX3582R on race gas....

Urk. I have a small novel forming in my head, but just a couple of points to ponder on. 1/ If a tuner tunes to cover all scenarios in this kind of situation, they have no sensor for exhaust manifold pressure - so the only way they can make it guaranteed safe for all scenarios is tuning the whole thing as though everything is flowing through a single scroll. That would be aweful, exhaust pressure would be through the roof - if it could even maintain full boost like that. There would be minimal timing at high rpm/high boost and the car may as well have a much smaller turbo. 2/ An equivalent flowing twin scroll setup tends to take noticeably more timing/make more power through the middle of the rev range than a single entry manifold. Thats comparing two efficient setups.... what happens if you turn a twin scroll into a single scroll with half the flow potential, and 3 cylinders have an exhaust flow which could either take a pair of 90 degree bends going into the turbine housing, or just go straight in - while the other side just goes straight in.

No need to apologise at all, I understand what you are saying (and its what I hoped until some disheartening results started to shot up) but I'm not sure the possilibity of better response is enough and on either sides case its just speculation until someone tries it. FYI the GT3071R vs stock turbo comparison isn't very useful, there is something very wrong with the stock turbo dyno plot - its almost like the operator is lifting back on the throttle at a couple of points in the dyno run and feeding it back in or something.

The gradual opening thing is something which worries me a little, if its tuned on a dyno there is an assumption that at any particular boost level the valve will be so far open and timing will be adjusted to optimise that. I am prone to overthink things, but I know with swingvalves there is a bit of a delay (most people with internal wastegates get a little bit of a boost spike on gearshifts) but what if you shift at high rpm with this setup? Conceivable that if the swing valve is set to 20psi, where it hits at around 4300rpm on the dyno- but rpm drop to 5000rpm on the gear shift and shoot straight up to 26psi.... there could be a very brief moment where you have 26psi of take pressure, a gazillion psi of exhaust manifold pressure and the ignition advance to suit something with perhaps have a gazillion psi of exhaust manifold pressure. Could make for a pretty unfriendly engine environment in a track car, though just pondering out loud really.

That link doesn't work so I can't comment, but you can't really compare a .63a/r GT3071R with a .82a/r GT3076R - especially in the context of a GTX3071R where the compressor will choke in the .63 housing (like the GT3076R). I have never seen a GT3071R with a factory like boost threshold, I'd be very interested to see the dyno plot you are talking about. I can say mine is a nice and fun street setup, quite a lot of people have been in/driven it and I've never had a comment knocking it. It definitely gives something away to a stock setup at low rpm but that bit goes away so fast you hardly notice it. I do know a GT3071R spools better than a GT3076R, if I knew someone with one its the kind of thing where it'd be interesting to put two cars side by side in 3rd gear at 3000rpm and give it the jandel and see what happens. My bet would be a trivial (like inches) until the GT3076R setup lets its throat clear. Where would that leave a GTX3071R which is a little lazier?

Thats pretty cool, I approve of this method - had actually been trying to come up with a good way of achieving basically the same thing but never posted anything. It still looks a bit more whorey than what I'd like to do, but I can't really knock it too hard. Would be very interesting to see videos of it in action The thing which makes me curious is what affect the back pressure has on engine efficiency and tuning to suit, if you are letting it build to 20psi before opening the second scroll then I'd imagine that back pressure is getting very high - then it will suddenly drop substantially when the second scroll opens. It'd be interesting to see the torque curve.

I am pretty sure I've mentioned at least once in this thread that I don't have full confidence in the GTX mid frame turbos at this stage. Initially I had doubts about the GTX3076R but thought the GTX3071R would be awesome, however seeing that the GTX's are laggier than the GT equivalents and there is bugger all different in lag between a GT3071R and a GT3076R it seems to me that a GTX3071R may have no advantage over a GT3076R but will definitely cost a heap more. It seems a few other people agree with my doubts on this one. If I were in your position I'd sooner go the GT3076R with the same housing if thats REALLY the way you want to go. Its well proven, and is cheaper. Where the GTX3071R should on paper come into its own is at really high boost levels, way beyond what you'll be running I expect.

Start a new thread for quick spool theories then, its well off the topic of this thread. When you start it show evidence of a twin scroll turbo with a twin scroll manifold being outperformed (to its potential) by a spool valve too. Every comparison I have seen involve with and without the spool valve with a single merge manifold (make ts redundant), or on a turbo pushed well short of its limit meaning that you don't see if the QSV is actually a restriction.

Cheers, I personally think for a street RB25 it could be unbeatable - could redefine what a good street setup is. Not having to buy a BOV (I run a stock one atm), or wastegates, potentially having full boost by 3500rpm and making >300kw on pump gas and having a psychopath E85 tune capable of over 400kw. Want. Its probably worth noting that an EFR7670 doesn't really school a GT3076R (on paper) until 20psi+