Lithium

I like all this. The thing that killed my ability to resist responding was the mention of modded S58s 💦 God they're good.

No worries, please don't take it as gospel - I wouldn't I don't have direct experience with this kind of thing and it just got my curiosity, though I'm sure you can decide whether stuff I've said seems nonsense or not as you look into it more. Out of interest, the Xona/FP turbos are pretty expensive but they do work to unusually high pressure ratios. I think they have got a bit more common with "restrictor classes" in various forms of motorsport for that

Yeah I've hardly ever tried doing math for this exact thing and suspect I may be going quite wrong, or maybe matchbot DOES need some work for this kind of thing. I decided to calculate the maximum airflow through a 38mm orifice with a .9 discharge coefficient (I have no idea what was reasonable so went for something big while not sounding completely impossible) and with standard ambient temp/pressure I have ended up settling on 6psi below ambient at the compressor inlet for ~31lb/min mass airflow. With this I put the -6psi line in to Matchbot using the "Default" 2litre setup, and ended up needing to run a boost curve that looks kinda like "5000rpm = 23psi, 6000rpm = 15psi, 7000rpm = 11psi" and ending up with 300hp. This results in matchbot estimate 54lb/min corrected mass flow (so what you look up in the compressor map as opposed to the actual mass flow), and a compressor pressure ratio of 4.46:1 at 5000rpm dropping back to "only" 3:1 at 7000rpm. Kinda doesn't seem toooo shocking to me, the 71mm Garrett GT series compressor is basically "topped out" at 46lb/min so it'd not be surprising at all if any amount of "gate shutting" was going to be in vain to a degree. Now matchbot DOES estimate a pretty terrible amount of EMAP for that boost, but from what you say you're not actually seeing that. Matchbot "assumes" a given amount of power needed to drive the turbine and I wonder if it's factoring in the fact that you may not actually need as much power to drive a given pressure ratio across the compressor if the intakeP is a LOT lower than ambient. Basically a lot less resistance on the compressor, kinda a trick that was used in early F1 for antilag (a pre compressor throttle that closes with the actual throttle to bump the PR and reduce drag on the blades). At this point if I were to guess, I'd say that the compressor can't even move enough air to get you to a point where the turbine is choking the engine - if I were going to be ruthless I'd try fitting a smaller a/r housing, possibly much smaller so you can bring the drive pressure enough to spin that compressor will a relatively unimpressive amount of exhaust flow. Probably a good idea to get a compressor that is more efficient at high PR as well, lol. Caveat: I'm still bloody tired and this could be nonsense, but may or may not have said something that could prove useful!?

I'm pretty fn tired right now so fair warning I may say something really stupid without realising it - but the way I read this it basically covers what I mentioned above. I might have to run into matchbot to illustrate what I'm getting at, but basically the pressure ratio across the compressor will be going through the roof and compressor efficiency potentially getting completely rooted. What will happen there is the thing has to shut the gate to keep driving the compressor but the compressor is not moving enough air to drive the turbine any harder than it already is, which is essentially what I was getting at when I said "how hard the turbine needs to work". Even though the absolute pressure in the inlet manifold is not super high, the pressure ratio across the compressor is MUCH higher than the pressure ratio across the turbine. It almost wouldn't shock me if a tighter hotside would work better, lol. I'm going to have a fiddle with matchbot now because I'm curious.

Trick is that the airflow through an orifice loosely speaking increases by square of the diameter, so a 76mm restrictor (lol) will support *quite* a lot of power. The other issue is the pressure ratio across the compressor will go up significantly as pressure drops below ambient between the restrictor and the inducer, so that starts doing funky things with compressor efficiency and also how hard the turbine needs to work... things I assume someone using Matchbot are hoping to try and anticipate.

I think it should do a reasonable effort so long as you put the right values in and the results will only suit the conditions you are applying the numbers to suit.

Not *as* suited as the EFR8374 I suggested - and only mentioned that because I thought you'd responded to my comment though OP has mentioned pump gas so gotta clarify some things as I may change that view a bit too haha. I'd prefer Sinco myself, I would look into Artec's cast divided manifold as well though there have been posts coming up suggesting that fitment may not be great so that might be worth keeping an eye on. My turbo suggestion was also assuming ethanol, you *may* need something different depending on what octane fuel you'll be running and how the dyno reads? You may need a "bigger" turbo than the EFR8374 that I mentioned if you're looking for 600hp @ wheels on say an Aussie Mainline or Dyno Dynamics rolling road dyno on say 98RON, but if you're going to be running on a hub dyno or Dynojet then it wouldn't be a problem.

Edit: Oops, I clicked the link to the thread notification thinking this was a response to my comment and said stuff explaining why I thought the G35 wasn't as suited to this situation and it could have looked like I was having a dig - and just realised you were responding to another comment. Ignore me!

Without being absolutely ridiculous about it, this is the gist of what I'd look into: - RB28 with VCam - Nismo intake manifold - If available a good T4 divided cast manifold. Smallest volume possible that is going to support 600hp, PROPER pulse convertor style thing - EFR8374 with 1.05 T4 twin scroll hotside - Good intercooler with minimal pressure drop and unrestrictive intercooler piping - Free-est possible flowing exhaust. 4" dump pipe off turbo - Run it on ethanol I'm wondering what I'm missing with people suggesting a G35 900, if there is a result or something I've not heard about I'd love to see it - from all I'm aware of so far that'd be a solid step down from an EFR8374 in terms of overall stand up/response if everything is put together and working right. Sure, capable of more power and go well for what it was but not "peers" with an EFR8374 in terms of responsiveness from anything I've seen so far. A G30 770 would be more comparable I'd have thought, and if I had seen such a setup I'd consider that as a possible viable alternative to the EFR8374 and would love to see how it would go - but I'm just suggesting unbiased and proven combos. At this point I'm not confident there is anything else that would compare that will safely hit 600hp and be as responsive.

Not to question this data at all, obviously very good input and context been given but I thought I'd add stuff from some data I've got for a car run in a hub dyno (these numbers are from a roller dyno that reads more like a Mainline/dyno dynamics roller than a Mainline hub) where we have pretty high confidence of no leaks etc. It's not for a G30 770 but should add some perspective. A mate's SR20VE with a Garrett Gen2 GTX3076R made 597hp @ hubs on E85 at 23psi at a bit of altitude, with 131,000rpm turbine speed. These don't have as efficient a compressor as the G30 770 and have the same max turbine rpm (around 145,000). Also gives a better lock on conversion of hub HP vs flow on E85 for a rwd on a Mainline prohub. If the are no leaks, the Pulsar G30 770 flows about the same as the Garrett version etc it seems like they shouldn't be at a point they are exploding at this power and I might go checking now, but I could swear others have made this power area fine already with them.

If you're not using throttle control for torque management probably not sooo much of a thing, but ignition cuts are pretty hard on stuff.... not necessarily something I'd totally expect to cause this kind of failure though. Cool yeah not familiar with the GPR so was just saying it in case it might be - and yeah with the GTR package there are ways of adjusting the boost targetting as well to avoid massive silly things from happening.

Haven't used the GPRP just something just occurred to me if you're using a package which supports torque management - I tuned an R35 using the M1 GTR package and one of the things that quickly got my heckles up was seeing the throttle pressure ratio data when trying to manage torque etc. Some competition level (Emtron as well) engine management systems that do stuff like this to ensure that you can control power delivery carefully but also have the turbo speed maintained so that when you want the torque "back again" without waiting for the turbo to respool can actually put a massive amount of load on the turbo if you're using e-throttle as part of the torque management. Like, huge pressure ratios and turbine speed - I'm not saying this is a thing with yours but its a thing that just popped into my head when @Komdotkom , mentioned "nifty stuff". If you are using e-throttle and torque management it's very much a thing to bare in mind, the conventional turbo matching stuff goes straight out the window when you start doing stuff like that as the transient situations can potentially end up operating the turbo in really strange places. Strictly speaking that can even be a thing if you go half throttle with a conventional throttle where the turbo is still targetting "full boost" and the boost control setup is able to maintain close to target, the turbo has to try and make up for the pressure ratio across the throttle as well and can end up working super hard - this is actually another reason why it can be sensible to reduce boost targets at part throttle. A small hint of this phenom is where some cars can go into surge conditions at part throttle. Pressure ratios vs flow get super odd.

Same. Do you have I2/I2 Pro? How familiar with the software are you?

Ah sorry, from some things you'd mentioned I assumed you were more comfortable with reading compressor maps. Nah, you're right against the edge. The red line I've drawn through the map is a rough estimate of where your engine may be operating in and it "stops" basically just at the max-rpm line on the right side of the compressor map. Touching that line doesn't mean instant fail by any means, but it does assume there are no outside factors meaning it has to spin harder and obviously I'm also just taking a punt so there is no guarantee that's accurate. Nice computer Maybe not necessarily quite a straightforward to work out the airflow the ECU reckons the engine/turbo are moving though it's definitely capable with I2 etc.

They've been out for a few years and actually a "joint" effort with (or FP say basically is a FP turbo that Mitsi liked so much they've sold as an OEM option) Forced Performance. They seem like excellent performers and are loosely FP Greenish from memory. They also happen to be "standard rotation" as they are an Evo X option

This is what I get when I have a quick go at Matchbot, let me know if it seems like I'm missing something. I'm "only" running it to about 7700rpm though as that's all your dyno plot goes to. Plausible that running it to 8500 could go further but it looks like your VE is going to start plummetting after 7700rpm so I'd not necessarily expect the compressor to need to do heaps more. What ECU are you running? You could look at logs to get a gauge of the estimated airflow. Here's my attempt at matchbot: https://www.borgwarner.com/go/N0YW5D

Ugh actually. Please hold, I'll just actually run the math properly and see where you're likely to actually be instead of guessing. Bbs

Ohh. I may have made a bad assumption with my previous post - I kinda glanced at the power and assumed your were running ethanol (which kinda assumes a certain airflow vs power conversion) which would make that power seem perfectly within the G30 770 capacity but actually, looking at the rpm and boost you're right... all things being "typical" your engine is likely to be breathing a heap more than that turbo would want to give at that boost. What fuel are you running on?

Argh also, I missed a point above - it seems that especially with the Pulsar turbos third party dealers seem to sometimes mess up orders. I've seen it (not directly experienced it, but definitely seen it happen to other people) where someone will order something like a GTX3584RS and end up with a GTX3582R, or a "G30-770" turns out to be a GTX3071R etc. If you had a GTX3071R or even GTX3076R or G30 660 and tried to run it like a G30 770 then there could easily be carnage. I've seen people run near 600hp @ hubs on the Pulsar G30 660 on ethanol so it can be done, but probably not super wise.

Seeing as it's a little ambiguous as to why you might be upgrading, I'll give some bullet point answers based off the two reasons you may be: Upgrading because you're assuming the failure is overspeeding: A G30 770, either Garrett or Pulsar version should not be overspeeding at this power level. It's leaning on them a little but should still be in a relatively safe place Any details on how the turbo looked/what broke and when? First time I've heard of a catastrophic failure with a Pulsar, can never rule out a manufacture issue with them (I know people who've had this with Garrett, Borg, Precision etc so Pulsar probably won't be immune) You can also overspeed the turbo without actually trying to make more power than it can support, if you have a restrictive intake or intercooler, or some kind of boost leak then there is the possibility that either the turbo is seeing a much higher pressure ratio than you'd guess based off when you see at the intake manifold - or alternately the turbo could be moving a heap more air than the engine is seeing, both causing it to overspeed while actually giving the engine less air than the turbo can provide if everything is working efficiently. Make sure there is no other issue with your setup that could possibly cause a premature failure, like oil feed blockage or something to that effect ^ Especially #3 and #4, make a seriously good effort in ruling out or identifying any possible cause there because you don't want to spend >2x the cost on a replacement turbo to end up with the exact same thing happening. Upgrading because you're levelling up: How much more power are you after? If you're happy with the power and response you had already, 590hp @ hubs isn't a heap short of what the compressor supports with those so going up an a/r may result in an underwhelming increase in power vs how much extra lag you get from it - imho the .83 seems like a good balance on the 770 even if it may hold it back a little at the top end of it's capabilities. Check out the 6466/G35 900 thread, someone has gone a 1.06a/r T4 Garrett G35 900 on their VCam RB28 and it's a lot laggier than this. There's no magic, if you have a bigger compressor, bigger exhaust wheel, big tubular manifold and larger exhaust housing then twin scroll is not going to offset the response at all.... but it will be capable of an extra 100kw @ wheels safely. If you are after an incremental increase maybe consider the 1.01 T3 G30 900, or if they have it a 1.06 T4 G30 900 - I'd wager that has a good chance of being a sensible in-between if you are concerned about response. It's unlikely that you'll be able to go up in power potential without some cost to response but that is one way that's likely to manage the impact somewhat. Just my thoughts.

PS part of why I did the above is so you can see your setup is likely reaching as high as PR 3.3

K, I've thrown some numbers loosely based off your dyno plot and what you've said into BW Matchbot. I've selected the EFR8474 compressor map because it happens to be an excellent match for what you're actually doing and also funnily enough Borg Warner doesn't have the Garrett data I used turbine flow data for the G35 900 with 1.01 hotside (as the 1.06 T4 isn't listen) to loosely match the turbine data. Very loose but should give a general idea of what EMAP and pressure ratio etc might be doing. https://www.borgwarner.com/go/3I1H4G

2.3PR is around 19psi boost at sea level if you have zero intake restriction - you sure you aren't missing a PR there?

Just to clarify, has yours failed or are you worrying about how risky you're playing it?

If you hate Precisions but like some of the "insane power vs turbo size" things maybe worth having a nosy at ApexTurbo, be super interesting to see something like one of their 64mm or so type units on an RB30. They seem to be getting results that seem impossible, even when you consider the latest Precision offerings... most significant Evo/Honda turbo records have already been picked up by things running ApexTurbos and one of the Honda guys "downgraded" from a bigger Precision turbo to a 62mm Apex to try and make it easier to get off the line and ended up making significantly more power at the same time, like they cracked 1000whp which is mad even on a US dyno without nitrous etc.