Single Vs Twins

[ali-turko]

ok thx heaps guys iv made up me mind and im going with a 35r

[Sydneykid]

Well I have 35r and a mate has 2530's, I can promise you my 35r (.82) comes on earlier and harder. It also went 400rwkw with afr's at 10.9 running 24psi.

My experience is exactly the reverse. I have had a GT35R (0.82 A/R turbine) on an RB30 and 2530's on an RB26 and the 2530's on the RB26 were much more responsive, even with the 17% less capacity. Both were tuned pretty much the same, on the same dyno, by the same tuner, with the same dump back exhaust, intercooler etc. But the 2530's were noticeably superior in throttle response every time. Personally there is no way I would use a GT35R on a 2.6 litre, the pooor response would pee me off day 1.

Cheers

Gary

[Cubes]

Of which comes down to the old...

The greater the mass of the comp and turbine wheels the less responsive it is going to be when accelerating from x rpm to y rpm.

[Lithium]

My experience is exactly the reverse. I have had a GT35R (0.82 A/R turbine) on an RB30 and 2530's on an RB26 and the 2530's on the RB26 were much more responsive, even with the 17% less capacity. Both were tuned pretty much the same, on the same dyno, by the same tuner, with the same dump back exhaust, intercooler etc. But the 2530's were noticeably superior in throttle response every time. Personally there is no way I would use a GT35R on a 2.6 litre, the pooor response would pee me off day 1.

I have disagreed with you before and you've made out like I didn't know what I was talking about, and I realise you have a LOT of experience with Skylines but I don't know what the full story is with what you are claiming above - maybe something was wrong with the RB30DET setup or maybe something else!? Either way, there is no way in hell a decent RB30DET setup with a .82 GT3582R turbo on it is laggier or less responsive than an RB26 with GT2530s.

I've been in and/or driven all the setups you describe including an R32 GTR with a .82 GT3582R and the GT3582R on the RB30 monsters a GT2530 turbo'd RB26 everywhere. If you rolled both setups side by side from 2500rpm in 3rd gear the RB30DET / GT3582R car would be gone by the time the RB26/GT2530s was starting to think about getting serious. Its been a while, but the way I remember it the GT2530s did seem to "windmill" a little happier than the GT35R on an RB26 at driving around town speeds but as soon as you started getting into it the thing would charge into life and get out of dodge. I swear than the GT35R on the R32 GTR I have driven is the least laggy near 350-400wkw setup I have driven overall.

This isn't the best comparison but it will give an idea - the plot from two cars tune on hub dynos, the one with the GT3582R (the one the dyno plot is actually for) is running .82 a/r turbine housing and the run was started at 3600rpm. 4WD was used for the run, and the car was running 20psi with Tomei 260deg cams.

The blue line I overlapped from another mate's GTR which was running more mild (possibly stock?) cams on 22psi with the fuse pulled - running straight RWD for the dyno run. The GT2530's setup was started at 2000rpm, giving a little more time to get the turbo up to speed. Probably worth a mention as both the GT3582R and the GT2530 runs were done with quite quick "ramp rates".

[Lithium]

The greater the mass of the comp and turbine wheels the less responsive it is going to be when accelerating from x rpm to y rpm.

I'm fairly sure a pair of GT2530s will have more mass to accelerate than a single GT3582R.

[Cubes]

Ahh Mr Lith but 2500rpm is far from measuring response.

You don't combine the 2 turbo's mass in order to determine their over all acceleration. Sure there is slight loss due to additional friction (turbine housing etc) but the neg's do not out weigh the positives.

Response is the time taken to spool once at an rpm that is able to provide full boost.

Consider a GT35r at 4000rpm and a pair of twin gt2530's at 4000rpm on the same RB26 with very similar mods making similar power.

Second gear. Mash the throttle; the twins will reach their peak boost quicker and get moving sooner than the gt35 with its higher inertia.

No take the two cars to the track where your almost always at an rpm where you can bring the turbo to full song; the twins excel and offer improved acceleration that is more tractable out of corners.

The big singles tend to lag then bang come on hard due to the whole inertia thing.

Thats my understanding and experience with a few different settings anyway.

It sucks that a RB26 with a pair of 2510's has the same 'response' as my gt30 .82 on my rb30. YET mine spools a hell of a lot earlier and has a shit load more under the curve; his doesn't really get moving until ~4k.

[Lithium]

Consider a GT35r at 4000rpm and a pair of twin gt2530's at 4000rpm on the same RB26 with very similar mods making similar power.

Second gear. Mash the throttle; the twins will reach their peak boost quicker and get moving sooner than the gt35 with its higher inertia.

No take the two cars to the track where your almost always at an rpm where you can bring the turbo to full song; the twins excel and offer improved acceleration that is more tractable out of corners.

The big singles tend to lag then bang come on hard due to the whole inertia thing.

Thats my understanding of the whole shebang anyway.

I understand that what you say above is consistant with the popular belief, though 99.9% of people that say this stuff has never actually driven a GTR with a GT3582R, and wouldn't be surprised if a good percentage who repeat that rule haven't even driven a car with GT2530s to realise exactly how laggy they really are. For what its worth, the car with the GT2530s has to pretty much clip the 7500rpm limiter it has at the drags to launch without bogging - the car with the GT35R will smoke all four tires if it launches over 7000rpm, only needs 6000rpm to get a clean take off. The GTR running the GT3582R on the road feels better from 4000rpm than the GT2530s I have driven, I believe it has better overall response.

I still maintain that rule of thumb comes from people comparing big singles which outflow the pants off a pair of GT2530s with GT2530s. Maybe I need to sort out a camera and make a response clip between a twin GT2530 and a single GT3582R and see if observers of the clip can pick which car is running which turbos

[SimonR32]

funny about 2530's i also consider them on the laggy side

i have seen dynos of 2 cars, one with 2.6l T04Z and one with HKS 2.8l with 2530's... same dyno, same tuner, same ecu, similar supporting mods and similar intake temps

low and mid range there is no difference, top end Z is in heaven!!!

surely a gt3582R is more responsive than a T04Z?!?

[Lithium]

surely a gt3582R is more responsive than a T04Z?!?

GT3582Rs are WAY more responsive than T04Zs. Cubes, that response you talk about is going to be hard to improve on. No matter what turbo you have it has to accelerate up to speed. The fact yours spools way earlier and has the same kind of response indicates to me your car is WAY punchier.

Strictly speaking up to a point you may not get a lot more response beyond a certain point, and clearly there is around a 1000rpm area where yours has to be more responsive as it can make more boost in that rev range? Also while I have you, you know how your car spools the GT30R - and that GT35Rs are the next logical upgrade (laggier, but not hugely so), and you are just running a basic internally gated setup on a stock exhaust manifold - do you not find it strange that someone has claimed that GT2530s on an RB26 is more responsive than a GT35R on an RB30?

Edited July 3, 2008 by Lithium

[Sydneykid]

Maybe we are talking about 2 different things, coming from a circuit racing perspective my observations are based on the speed of the response from the engine when I move the throttle. Looking at a dyno graph tells me nothing about throttle response. Stick it in 4th gear, load it up against the dyno and even the laggiest turbo looks OK, because it's not measuring throttle response. There is no way I can do the same lap times with a GT325R on a 2.6 litre as I do with it when it's 2530 equipped. I can't drive it on the throttle because the response to my inputs is simply too slow. That's why an RB26 has multiple throttle bodies close to the engine, that's why it has twin turbos mounted close to the exhaust ports, it's all about the maximising the speed of the response that's required for winning in circuit racing.

A simple example of one piece of a complex puzzle, looking at a dyno graph doesn't tell you how long the exhaust gas takes to get from the exhaust valve to the turbine blades. Forget the relative sizing of the turbines, forget the 6 cylinders versus 3 cylinders. Think about 1 cylinder having a 2530 turbine that is ~30 mm away from its exhaust port and how much faster it is going to respond than a GT35R turbine that is 300 mm away from the exhaust port.

Cheers

Gary

[GeeTR]

its kinda hard to knock the king off its post.. u might make more power, but can u lap the tsukuba in less than a min???

Garage Saurus's 800hp dinosaur BNR32 w/ OSG 3.0L and T88 does a 0'56"641. The power in this case, makes up for the lag it seems.

Interestingly, Tsukuba is a larger track then those, which the mines car seems to dominate. Mines has done a 0'57"840

times for comparison. Edited July 4, 2008 by GeeTR

[archie01]

Maybe we are talking about 2 different things, coming from a circuit racing perspective my observations are based on the speed of the response from the engine when I move the throttle. Looking at a dyno graph tells me nothing about throttle response. Stick it in 4th gear, load it up against the dyno and even the laggiest turbo looks OK, because it's not measuring throttle response. There is no way I can do the same lap times with a GT325R on a 2.6 litre as I do with it when it's 2530 equipped. I can't drive it on the throttle because the response to my inputs is simply too slow. That's why an RB26 has multiple throttle bodies close to the engine, that's why it has twin turbos mounted close to the exhaust ports, it's all about the maximising the speed of the response that's required for winning in circuit racing.

So true.

From memory, when i think about my To4z graph to my -5's graph, down low there isn't a hell of a lot between them.

On the track however out of slower corners it is VERY noticeable

[Lithium]

Maybe we are talking about 2 different things, coming from a circuit racing perspective my observations are based on the speed of the response from the engine when I move the throttle. Looking at a dyno graph tells me nothing about throttle response. Stick it in 4th gear, load it up against the dyno and even the laggiest turbo looks OK, because it's not measuring throttle response. There is no way I can do the same lap times with a GT325R on a 2.6 litre as I do with it when it's 2530 equipped. I can't drive it on the throttle because the response to my inputs is simply too slow. That's why an RB26 has multiple throttle bodies close to the engine, that's why it has twin turbos mounted close to the exhaust ports, it's all about the maximising the speed of the response that's required for winning in circuit racing.

I realise this, but I can't show you a graph to demonstrate as there is no easy way of doing so. I can and will try and endevour to get some evidence of how a GT2530/GT2860-5/GT3582R car all compare as I get sick of speculating on this topic - and hearing what are often unfounded opinions supporting one or the other way of doing things. For what its worth some of the fastest track cars in NZ are EVOs running GT3582Rs on 2litre engines, I've been in one myself which with full interior and pump gas temporarily (until a full competition car at the same event) took out the course record on a tight hill climb event. If a 2litre engine running high 300awkw on pump gas on one of these turbos can be highly competitive on a circuit I don't think the response is going to harm a 2.6litre.

As I've said before and you seem to ignore, the GT3582R GTR I have driven comes across as overall more responsive than GT2530 cars I have driven - Cube's example of GT2530s from 4000rpm assumed they take off from those revs on an RB26. They don't, if you do that from 4000rpm in 2nd on an RB26 it fluffs around for a bit before it does much. A GT3582R while it doesn't leap straight onto full boost either will give you a pretty decisive push in the back and then charge up the road, I can't argue it TOO hard but I really remember feeling more impressed at how the GT3582R responded than how disappointed with the GT2530s I felt. The reason I focus on 4000rpm is if you keep the GTR I talk about on the boil (ie, race track revs) then whenever you floor it the thing takes off like something with its arse on fire.

[DCIEVE]

So true.

From memory, when i think about my To4z graph to my -5's graph, down low there isn't a hell of a lot between them.

On the track however out of slower corners it is VERY noticeable

which might be expected when you compare a turbo happy in the 450+rwkw range with a pair that work in the 360rwkw range

plenty of talk of inertia but no numbers, if someone provides compressor and exh wheel diameters and weights we can very roughly (very) approximate the relative moments of inertia and get a better idea of the actual difference..

Edited July 4, 2008 by DCIEVE

[Lithium]

which might be expected when you compare a turbo happy in the 450+rwkw range with a pair that work in the 360rwkw range

plenty of talk of inertia but no numbers, if someone provides compressor and exh wheel diameters and weights we can very roughly (very) approximate the relative moments of inertia and get a better idea of the actual difference..

Thats pretty easy:

HKS GT2530

Compressor: 47.7mm inducer, 60.1mm exducer

Turbine: 53.8mm Inducer, 47.0mm exducer

If you were to double the area of the wheels, the inducer and exducer equivalents would be:

Compressor: 67.5mm inducer, 85mm exducer

Turbine: 76.1mm inducer, 66.5 exducer

Garrett GT3582R

Compressor: 61.4mm inducer, 82.0mm exducer

Turbine: 68mm inducer, 62.3mm exducer

The trick is with the GT2530 it has double the amount of material for the equivalent surface area, seeing as it has two wheels - that is weight and also space not taken up by air. I'd imagine that will (aside from fluid dynamics which are beyond me but must come into effect when two compressors are trying to pump in the same path) compromise some efficiency. There are definitely advantages to the twins, but I'm yet to be sold on them making them better than a matched single.

I'm still learning stuff, but can't help but think that the tight housings they put on GT2530s etc must strangle them a bit compare to the relative large housings you get on GT3582Rs. A GT3582R I am very sure has a LOT more housing volume than a pair of GT2530s, which is maybe part of the reason why they aren't as punchy as you'd imagine compared to the GT2530s - but also why they can push enough air to warrant a conversation about them?

Edited July 4, 2008 by Lithium

[Sydneykid]

I realise this, but I can't show you a graph to demonstrate as there is no easy way of doing so. I can and will try and endevour to get some evidence of how a GT2530/GT2860-5/GT3582R car all compare as I get sick of speculating on this topic - and hearing what are often unfounded opinions supporting one or the other way of doing things. For what its worth some of the fastest track cars in NZ are EVOs running GT3582Rs on 2litre engines, I've been in one myself which with full interior and pump gas temporarily (until a full competition car at the same event) took out the course record on a tight hill climb event. If a 2litre engine running high 300awkw on pump gas on one of these turbos can be highly competitive on a circuit I don't think the response is going to harm a 2.6litre.

As I've said before and you seem to ignore, the GT3582R GTR I have driven comes across as overall more responsive than GT2530 cars I have driven - Cube's example of GT2530s from 4000rpm assumed they take off from those revs on an RB26. They don't, if you do that from 4000rpm in 2nd on an RB26 it fluffs around for a bit before it does much. A GT3582R while it doesn't leap straight onto full boost either will give you a pretty decisive push in the back and then charge up the road, I can't argue it TOO hard but I really remember feeling more impressed at how the GT3582R responded than how disappointed with the GT2530s I felt. The reason I focus on 4000rpm is if you keep the GTR I talk about on the boil (ie, race track revs) then whenever you floor it the thing takes off like something with its arse on fire.

I am not ignoring what you said, I am simply comparing it with my own different experiences and coming up with a different result. Don't get me wrong I am not saying that a GT35R is a laggy pile of crap, the comparison is closer than that. Perhaps we are seeing different results because I am comparing the same car, same tuner, same race track etc and we do spend a lot of time and effort searching for minute improvements in response. We even sacrifice horsepower for improved throttle control if that's the best answer to faster race pace. So I'm comparing the same car tuned for maximum response with 2 different turbos, don't worry so much about a handfull of horsepower, don't care about the shove in the back, the data log is my judge, jury and executioner.

Hillclimbs reward traction in low gears, they don't give a meaningful comparison with floating a car throught a high G force turn in 3rd or 4th gear, using the throttle to balance the chassis. That's when that poofteenth of faster response pays off. You can see it in the data log, the throttle sensor shows the driver input and the long G shows the accceleration as a result of that input, any delay is measurable and comparable.

Cheers

Gary

[discopotato03]

Purely speculating (and in my own opinion) I'm not so sure that the polar moment of innertia between a GT35 turbine or a pair of GT28 NS111's makes a whole lot of difference .

Where I think the real world differences are is (IMO) the short 3 branch exhaust manifolds (less volume to fill/less heat loss) and that a GT2530R cartridges wheels are a more even match than a GT3582R . I tend to think their integral waste gates stand a better chance of controlling ratating group speed because the compressor and it flow capacity matches fairly closely the performance of teir turbine/housing size .

You can say here we go again all you like but you cannot escape the fact that parallel twins or properly set up TS singles will reduce turbine inlet pressure in relation to compressor flow with similar or better response - if they're speced properly . I'd say in this case the extra complexity of two turbochargers pays off over the effort to develop a really flying manifold / gate system for a TS single .

The turbines styles are very different animals too , the GT28 NS111's are the small version of a race turbine so are very light and have less (9) blades and more open ones . Their trim at 76 is probably more appropriate than a GT35's 84 trim size .

BTW GT30 and GT35 UHP turbines have 10 blades and a larger heavier hub .

Cheers A .

[Lithium]

The turbines styles are very different animals too , the GT28 NS111's are the small version of a race turbine so are very light and have less (9) blades and more open ones . Their trim at 76 is probably more appropriate than a GT35's 84 trim size .

BTW GT30 and GT35 UHP turbines have 10 blades and a larger heavier hub .

That is a fairly good point in the case, cheers DP - I can accept that as an advantage of the twins.... though in that instance more to do with the actual twins than the difference between twins and a single. Something like an RX6R would be nifty to get a hold of and see what it can do.

[Butters]

common debate this. "what turbo is more responsive" or "what turbo is less laggy"

Now key thing here is lag is a measure of TIME !

By now most people should relise that dyno graphs showing rpm(or km/h) vs hp made do not tell you how laggy a motor is.

As there is no time scale what so ever.

Seat of the pants always sucks too. I remember a BMI where they tested an EVO(i think) which was 7kg lighter in the roof area.

They crapped on how it "felt" faster and you could notice the 7kg difference. Then dropped a 7kg magnet on the roof and said it

felt slow and handled worse. The lap times came up and the 7kg heavier(with magnet) car was consistantly faster and the top level drivers really had to eat there words !!

Would be nice if we could come up with a rock solid way of measuring lag so that we could compare these setups.

Not being a dyno operator I don't know what can be done.

I am guessing though we would need logs of throttle % vs time vs tourqe. What you guys think ??

Wonder if DD could come up with set test so that we could compare cars.

[discopotato03]

I think the factory twin system makes it a bit easier to obtain parts that can be made to work properly , may not necessarily be cheap or simple but very workable . As has been flogged to death the single TS system can work really well but you HAVE to have a suitable manifold/gate/hairdryer and these are expensive - maybe 7.5 G's worth plus .

For real power IMO the head has to come off to be reworked and better than factory cams tossed in . Its no us having 700+ hp turbos on a 500 Hp head and claiming "its the turbo wots f*kt" .

I also think different people see different things in the engines power delivery , somewhere between the almighty rush and a Kw figure lies one theory and the datta dobber method the other ...

A .

Also BTW the larger 60mm version of that NS111 turbine (in 73 or 76 trim) is what many of the WRC terrors were using in the not too distant past . It helped the teams make something like 650 Nm of torque from a 2L four that was changed up at 5500 odd rpm and breathed through a (I think) 32mm restrictor .

The compressors were (again from memory) ~ 69 or 76mm and 5/10 bladed .

You could look at them as competition versions of the GT3071R/GT3076R .