The Split Pulse Turbocharger Thread

[Dale FZ1]

lol, talk about making a mountain out of a molehill. you guys have taken the massive leap of believing that split pulse turbine housings make significantly more power, or are significantly more efficient than an open housing. there certainly is a difference, but it is by no means significant.

my opinion is that split pulse will always be better (all else being equal) as you shield the interference pulses from each other a bit more effectively than on an open collector.

the downside is the angles involved in the merging gas streams are usually a bit more acute, which is not good. but shielding the pulses so that any given exhaust valve never sees a huge spike in back pressure from another valve is PROBABLY going to outweigh the gains given by a slightly more obtuse entry angle. can't say for certain.

what i can say is that the difference between two optimised designs of split pulse and open collector (all else being perfectly equal) will not be hugely significant. by simply porting the head, or using a custom manifold or larger exhaust size, you will easily outstrip the power and efficiency gains to be had from using an optimised split pulse exhaust side.

essentially if you're looking for a bolt-on turbo for an RB25 manifold, you are totally negating the benefits of using a split pulse turbo anway, since you are using the shitty stock manifold to bolt it to. the gains would not be measurable, and certainly not as cost effective as fitting a better intercooler or exhaust manifold.

Check the rationale for the thread. As stated, I'm very satisfied with my particular setup.

Not many of us are rocket scientists, though there is the odd claim to that qualification from time to time. I'm just interested to see what is out there and/or could be set up ECONOMICALLY to yield the benefits claimed for the system.

Frankly, I'm not yet convinced that it can be done, but hey it's worth a look and a yarn. If something came up and someone with the $$ and motivation to try it, then we might all learn something more. For example, Mafia showed what could be achieved with a 3076, a mechanically stock RB25 and water/meth. Just a step sideways to try to take a step forwards.

[paulr33]

im with stockmystock all the way (macgyver) i dont reckon its worth the cost involved and efforts

especially given you keep the stock exhaust manifold

[StockyMcStock]

the rationale for the thread, i believe, was to explore the idea that a twin scroll turbo bolted onto an RB25 manifold could be better than all the current turbos people are using with open collectors.

power level is irrelevant, the theory behind the different housings remains the same: shield as many of the exhaust pulses from the other cylinders as possible to avoid any one cylinder being hit with a very high pressure spike from another cylinder. thus, you will gain maximum scavenging from all cylinders.

the answer is theoretically yes, it will help, but it is not a significant difference. by using a more efficient and modern turbocharger with improved wheels/scroll angles/internal leakage etc you will far outstrip the potential gains to be had from using an older twin scroll turbo. i don't know of many dedicated modern twin scroll turbos rated at less than 800hp, but then again i haven't looked for one since i bought mine.

[Adriano]

Has anyone seen the wastegate setup on an evo4-9 turbo, they are split pulse, but i dont recall the wastegate setup. PS and the turbo's are quite cheap second hand and can be highflowed to reasonable power levels.

[discopotato03]

Hi Stocky , the rational is not just about foreign pressure waves beyond an opening exhaust valve . Its also about the cylinder exhausting into an area of lower pressure so the sudden rapid pressure rise means the gas can accelerate more easily and lose less of its velocity between the exhaust valve and the turbine nozzle . To a degree the exhaust pulse/shot/slug moving into a volume of lower pressure is still expanding and this expansion will assist exhaust gas velocity into the path of least resistance . If the chosen path has higher than desirable pressure the expansion rate will be reduced as will the exhaust gas velocity AND the merged gas pulses have a cumulative temperature rise effect on the gas . When its possible to use a larger A/R divided turbine housing than would be the case with a single its all win .

I wouldn't be too harsh on the std RB25DET exhaust manifold , obviously its possible to be improved on but to be able to get 300 od Kw worth of gas flow through the thing means its hardly woeful . You don't lose the divided nature of these manifolds because the angles are a little sharp and the runners all different lengths . That moves the exhaust pulse timing around a bit at the nozzles but thats the compromise you get from not using a properly developed competition grade exhaust manifold . By this I don't mean you have to have a manifold like Geoff Raicers to get some benefits but they seem to take the most advantage of of whats going on .

Stocky yes there are a few sub 800Hp divided turbos out there but the gap between say 350 and 700 is pretty big and right where we can't take advantage of it . Some of the later IHI VF series , Garretts GT3267/GT3271/GT3576 (plain and BB) GT4082 and probably many more .

Just a bit extra from the Grp N/WRC rally world .

These classes make the cars run an inlet air restrictor in front of their turbos to basically cap the maximum airflow the engine has available . It really is a clever way of enforcing a maximum power limit amongst competitors . The works teams had all sorts of dramas trying to sort out how to make enough torque over a wide enough engine speed range before the restrictor said no more . They needed to have high torque starting down low and short shift to stay on the boil . They also had smart boost controllers taper the boost off as the revs rose so as to get an acceptable rev ceiling even if at low boost to not suffer massive compressor surge . Anti Lag Systems (ALS) helps but its very hard temperature wise on the system . Close ratio gearboxes make it possible to stay in the enforced narrow rev range and thats why we are seeing six or seven etc speed boxes in the competition cars and nowdays filtering down to road cars . Mitsubishi got on the twin scroll wagon fairly early and Subaru (no doubt the others as well) were not too far behind .

It may be news to some but a few years back they were getting 650 Nm of torque out of a 2 litre four that was changed up ratios at about 5500 rpm . These are serious machines basically cross country F1 tech vehicles and they use twin scroll systems to help get them there . Production car manufacturers are using it as well , didn't someone mention in Mafias 300Kw adventure thread that the later Evo Lancer road car is making GTR34 torque but with a 2L four cylinder ? Surely we stand a chance of getting a large portion of that 650 Nm (possibly 3/4 so 487 odd) with a 25% larger inline six . This is 67 more than the JDM Legacy makes with an EJ25 and its quiet and complies with their emission regs - and has a warranty .

The current Wheels mag mentions the MY08 Rex which is supposed to land here next year , toned down to 400Nm and a bit over 200Kw . JDM I think 430Nm and 234 Kw ? If these cars are not too porky they are going to have some serious squirt and be pulling strongly under 4000 revs . The common factor seems to be that they make strong torque over a wide range and without massive revs , the turbos are not huge either .

Anyhow let the debate continue , cheers A .

[AndrewD]

Just on the VATN turbos I don't think they're a free ride either , Variable Area Turbine Nozzle I think equates to variable turbine inlet pressure turbine housing as well so its not without its compromises .

Cheers A .

No not VATN, VNT! The variable vane type ofcourse. Re New 911 Porche turbos.

Or seeing as you know your garretts try these small ones, gt15v, gt17v...

[AndrewD]

These classes make the cars run an inlet air restrictor in front of their turbos to basically cap the maximum airflow the engine has available . It really is a clever way of enforcing a maximum power limit amongst competitors . The works teams had all sorts of dramas trying to sort out how to make enough torque over a wide enough engine speed range before the restrictor said no more .

Close ratio gearboxes make it possible to stay in the enforced narrow rev range and thats why we are seeing six or seven etc speed boxes in the competition cars and nowdays filtering down to road cars .are not huge either .

Anyhow let the debate continue , cheers A .

Actually the inlet restrictor is not a smart way. Its an easy, cheap and MEASURABLE way. If it was smart try justifying that to any engineer by asking him of what he thinks about the vacuum created infront of the inducer, and then try convincing him that the severe pressure fluctuations as the nozzle approaches mach conditions is of benefit! Then if your game ask about the extra friction created by the necesscity to keep lubrication oil from being drawn in due to the vacuum and what harm this has to turbo speed response.

Maximum power can also be debated considering that teams like SUBARU run a massive intake volume between the outlet of the compressor and throttle bodies / ports. Although this is mostly for transient on/off throttle and not sustainable during steady state.

Also the pure claim that a narrow power band is the result of said restrictor that is not true either. Ive seen a restricted 600 cc turbo make peak power across a 4krpm range!

The main reason teams would run a narrow ratio box with increased gears is that they have found increased acceleration. This wouldnt be possible with the old manual shift boxes, but considering all teams now run automated hydraulic or electric paddle shift which actuate within 100ms increasing the number of gears aids acceleration without the wasted shift time. Again this is found in street cars, but primarily as an emissions / fuel economy perspective.

[discopotato03]

Well if you went to those same engineers and said find a better way of putting a measurable and policable devise on those class cars to create a level playing field they'd probably agree that its a neat way to achieve it . The whole turbo concept with competition engines I think came to a head in the F1 era . It was getting harder and harder to to police the rules with fuels amongst other things and power was climbing at a huge rate . It seemed to end up in the too hard basket and Grp A on tarmack only lived on another what 5 years ? Even then out here we had the bullshit "Cams Grp A" not FIA Grp A because the poor 2wd taxis with glorified truck engines couldn't keep up ...

Do you think anyone cried about Ford having to remove the silly TA34 turbos off their Cosworth YBD powered cars because the power range was unacceptably narrow with them ? Nope , it forced them to find a better way to live with the limited air flow allowed . The reason its smart is because it makes the engineers think harder about how to make an acceptable power range without increasing revs/boost/turbo size . Actually its a great leg up because as I mentioned the methods flow on to production cars and that can spell increased sales to the public .

Trust me the engineers don't give a rats arse if a poorly sized turbo surges in that application , they are far more concerned about finding things that do work rather than those that don't . If anyone followed the development of the Evolution Series Lancers they'd see that the only thing that bears any resembelance to the first one is possibly vaguely the engine block . They evolved right through the Grp A era (some major upgrading every year or two for homoligation purposes) and they sold heaps of road cars which after all help pay the bills . WRC is different to Grp A in that the manufacturers don't have to sell road going cars to make up homoligation quotas , however Mitsubishi Subaru etc sell lots of Evos Rexes etc and the development dollar is paying off .

With the gearboxes the ratios still have to be suitable , alternate means of moving the dog rings is fine but the ratios (and the gaps betwen them) have to suit the engines useful rev range . Also the sections they run over these days vary so much speed wise that the ratio spread lowest to highest has to cover a broad speed range . The only practical way to achieve this is to have more ratios with acceptable gaps in between .

In class racing there is no fair , there is just a set of rules the cars have to comply with . If you think the knuckle draggers had something to winge about with the Grp A GTR's there would have been some very much more red faces and bent egos had they not had their restrictors .

Cheers A .

[AndrewD]

Trust me the engineers don't give a rats arse if a poorly sized turbo surges in that application , they are far more concerned about finding things that do work rather than those that don't . If anyone followed the development of the Evolution Series Lancers they'd see that the only thing that bears any resembelance to the first one is possibly vaguely the engine block .

With the gearboxes the ratios still have to be suitable , alternate means of moving the dog rings is fine but the ratios (and the gaps betwen them) have to suit the engines useful rev range . Also the sections they run over these days vary so much speed wise that the ratio spread lowest to highest has to cover a broad speed range . The only practical way to achieve this is to have more ratios with acceptable gaps in between .

In class racing there is no fair , there is just a set of rules the cars have to comply with . If you think the knuckle draggers had something to winge about with the Grp A GTR's there would have been some very much more red faces and bent egos had they not had their restrictors .

Cheers A .

If they were my engineers they would be fired! Lucky for me non of mine are short minded like you suggest!

Obviously you are not awake to the idea that motorsport is all about seconds. It is not an alternative of moving the dogs, they are still actuated by the same mechanism within the gearbox (assuming sequential) its the human interaction which is of importance! On tracks where there are hundreds of shifts and because the hydraulic / electrically actuated mechanisms can save easily 200ms per shift over a manual this time adds up to a few seconds. Try tell me or any engineer that it isnt worth the complexity of the mechanism nor the extra gears especially when races such as WRC and F1 are won or lost because of only hundreths of a second!

The speed variance of the tracks has changed little over the years,because of the powercap of restrictors. I think either your missing the point or avoiding it!

Another reason why the powerbands may be so narrow is that most 'RACECAR DRIVERS' are expected to be able to drive with peaky narrow powerbands, if they cant they arent very good! Average drivers are not expected to handle such engines hence the fact that all road based productions cars are designed and tuned for wide powerbands.

I would be willing to bet that 'motorsport' is used primarily as a MARKETING hype rather than purely develpment for production based cars. Considering the extra attention a constructor recieves from success in these categories and the fact that the production cars hardly represent the race companion by any stretch of the imagination except for some similarities in bodywork and four wheels!

[STATUS]

lol, talk about making a mountain out of a molehill. you guys have taken the massive leap of believing that split pulse turbine housings make significantly more power, or are significantly more efficient than an open housing. there certainly is a difference, but it is by no means significant.

My thoughts exactly... i spent tens of 1000's on playing with differrent turbos and the biggest decider (after all the practical experience is said and done) for me now is the relation ship between wheels/trims/housing size not whether or not it is split pulse... in fact i dont even give split pulse a second thought.

Edited May 29, 2007 by URAS

[Dale FZ1]

Electro-hydraulically actuated gearchanges aren’t part of the discussion guys, but provide a bit of interesting discussion. I’m interested in a road-car turbo setup that doesn’t require expensive gearbox or final drive ratio changes.

im with stockmystock all the way (macgyver) i dont reckon its worth the cost involved and efforts

especially given you keep the stock exhaust manifold

the rationale for the thread, i believe, was to explore the idea that a twin scroll turbo bolted onto an RB25 manifold could be better than all the current turbos people are using with open collectors.

power level is irrelevant, the theory behind the different housings remains the same: shield as many of the exhaust pulses from the other cylinders as possible to avoid any one cylinder being hit with a very high pressure spike from another cylinder. thus, you will gain maximum scavenging from all cylinders.

the answer is theoretically yes, it will help, but it is not a significant difference. by using a more efficient and modern turbocharger with improved wheels/scroll angles/internal leakage etc you will far outstrip the potential gains to be had from using an older twin scroll turbo. i don't know of many dedicated modern twin scroll turbos rated at less than 800hp, but then again i haven't looked for one since i bought mine.

The rationale was to explore the system as a whole, within a defined power range, and with budget in mind. We are pretty well versed on what is common upgrade product, and the idea was to see exactly what we could get for the money and the desired performance. If discussion/experience showed that the stock manifold is a workable compromise, then good. If not, then that works against the cost factor. We really had to discuss at some level the dynamics of exhaust pulsing just to establish a platform. Whether everyone’s able to visualize the process is anyone’s guess but worth a go.

I would suspect that the stock manifold is technically compromised, but workable in practice – within the defined power output as discopotato03 has said. I happen to believe that the power level is a key aspect to that point, because it’d be possible to live with the compromise where the discussion is focused, but probably not if you were aiming for (arbitrarily) 400rwkW or more. That is some pretty serious power, and having all systems working TOGETHER in a complementary manner becomes increasingly important. Laws of diminishing returns do apply.

As Adriano has said – have a look at the MHI TDO5 units being used on the Evo from the ‘90s to current. They are current spec split pulse. To me it is significant that there is technology sharing between MHI and Holset, and that’s why I’ve been musing about both their products, and what is being used in service. I’ll back it in that Cummins (which owns Holset) is NOT putting dinosaur spec turbos on its engines.

Check the compressor map for a HX30E Holset – very comparable to a GT37 Garrett in terms of size and maximum flow capacity. My opinion is that the HX30 will easily feed a petrol engine with 480hp or more worth of air. Basic specs search points to it having an impeller of around 76mm, and with 8 blades. Whether it proves to have a sudden delivery when applied to a small capacity petrol engine is another matter, but evidently they produce different trim sizes and impeller designs. No doubt they can be specified to better match the application.

i spent tens of 1000's on playing with differrent turbos and the biggest decider (after all the practical experience is said and done) for me now is the relation ship between wheels/trims/housing size not whether or not it is split pulse... in fact i dont even give split pulse a second thought.

Can you add constructively to the discussion URAS, especially with respect to split pulse? This is not a wind-up. I do agree that impeller size and trim is a big consideration. With that amount of money spent, I want to hear what back to back split pulse vs single scroll experience you’ve had, and outline why split pulse just wasn’t worth it. Again: NOT A WIND-UP. If you’ve been and done it, then share the experiences please, not just your conclusion. I'm interested in the story.

[StockyMcStock]

okay somehow people started talking about rally car inlet restrictors there but essentially there is nothing to it, the split pulse exhaust housing will not make significantly more power.

Given that there are ?no? T3 flanged split pulse turbo upgrades available within the boundaries you specify, the average punter will need to either make up a custom made spacer plate or use a custom manifold. if that means one hundred dollars or one thousand dollars of extra cost, the loss far exceeds any potential gain. Spending that hundred bucks extra on an intercooler or good quality cold air box will be much more effective.

Oh ps: you're right about the Evo turbos - split pulse all the way. it just becomes cost-inhibitive to put a split pulse turbo on a stock skyline manifold due to all the other things you need to change (oil, water fittings, flanges etc)

[discopotato03]

okay somehow people started talking about rally car inlet restrictors there but essentially there is nothing to it, the split pulse exhaust housing will not make significantly more power.

Given that there are ?no? T3 flanged split pulse turbo upgrades available within the boundaries you specify, the average punter will need to either make up a custom made spacer plate or use a custom manifold. if that means one hundred dollars or one thousand dollars of extra cost, the loss far exceeds any potential gain. Spending that hundred bucks extra on an intercooler or good quality cold air box will be much more effective.

Oh ps: you're right about the Evo turbos - split pulse all the way. it just becomes cost-inhibitive to put a split pulse turbo on a stock skyline manifold due to all the other things you need to change (oil, water fittings, flanges etc)

Yes , exactly what you'd have to do with any non std turbo . The difficult issue is generally a non factory style wastegate .

Cheers .

[Dale FZ1]

Given that there are ?no? T3 flanged split pulse turbo upgrades available within the boundaries you specify, the average punter will need to either make up a custom made spacer plate or use a custom manifold. if that means one hundred dollars or one thousand dollars of extra cost, the loss far exceeds any potential gain. Spending that hundred bucks extra on an intercooler or good quality cold air box will be much more effective.

FWIW, the HX30 Holset model seems to be a reasonable starting point, IF that technical info can be obtained. Compressor flow range 45-50lb/hr, high efficiency up to PR 3.1, and turbine options available (IW, various A/R, and those titanium impellers). AND they have a T3 flange.

As per earlier info posted up, these things DO come with a T3 flange.

My understanding is that depending on spec, the turbine outlet is VERY similar in appearance to the MHI TD-06 with a bolt-up flange (trying to relocate a pic I found), or with a 4 inch V band fitment.

Couldn't agree more on making it cost effective, especially if someone was wanting to bolt up to a stock manifold.

FWIW, even the GT30 IW Garrett requires a different dump flange, so as soon as you go away from a high-flow of stock housings, you need to spend to get that one right.

[Warpspeed]

O/k guys, split pulse housings can sometimes help, and sometimes make no difference.

Where a split pulse housing can really help is where there are very few cylinders. You get an exhaust pop, then nothing happens for quite a while, then you get another exhaust pop. Take an extreme example of a twin cylinder motorcycle engine where both pistons go up and down together, and there is one power stroke every 360 crank degrees.

Now a turbine with two inlets, each with an a/r of say 0.3 is going to work a lot better than a single entry turbine with an a/r of 0.6 The whole thing will still have the same total flow capacity of the larger housing, but a split pulse half housing will really help it at low rpm where every exhaust pop goes into a much smaller turbine scroll of only half the total flow area. High gas velocity is maintained and the turbine gets a really good good kick every exhaust pop.

Now think of the other extreme, a V12 engine. Each bank will have six cylinders, and there are so many overlapping exhaust pulses that the exhaust flow is almost continuous by the time it reaches the turbo. A split housing will be zero advantage.

So you could say that a four cylinder engine will probably gain from having a split pulse housing, and a V8 definitely will not. A six cylinder engine is right smack in the middle. A six cylinder engine fires three times per revolution, and each group of three cylinders fires three times every two revolutions, or every 240 crank degrees. If the exhaust cam has 240 degrees duration, then there is no exhaust ovelap between cylinders in each group of three. But neither is there any pause between exhaust pops in each three cylinder group. So a split pulse housing on a six cylinder offers no real advantage. If it did, you can be sure Nissan would have built it that way.

A much better idea is to use a single entry housing with six individual runners correctly aimed into the turbine. This has been proved time after time as being the optimum design for an inline six.

Don't bother dreaming about the variable vane turbines, they just don't work very well on petrol engines. These things have been around now for twenty years, Nissan COULD have used them on the original R32 GTR if the idea worked, the technology was available back then.

The biggest problem is that ALL the exhaust has to go through the turbine all of the time. This creates massive exhaust back pressure. The vanes can certainly reduce turbine rpm, but the cannot increase the physical flow area inside the turbine housing. I am speaking from experience, I have fitted a Garrett VNT turbo to my engine, it was very responsive, strong mid range, and absolutely hopeless top end power. Stick with the Garrett GT series ball bearing turbos. If VNT turbos worked on petrol engines, Garrett would be advertising them as such. They only work properly on diesels.

Porsche are the only manufacturer to use them, and they also run wastegates, and the whole thing is controlled by the engine management unit.

I have messed around with VNT turbos myself for three years, and so have many others. Where are all the fast drag cars running VNT turbos ? Almost everyone has tried it and given up. If you want to do a whole lot better than a GT ball bearing turbo, consider twincharging. I have tried that too, and can thoroughly recommend it.

[paulr33]

you guys are so smart, seriously

so much knowledge in FI from you guys, well done, keeping us all educated on some of the more complex stuff

[Adriano]

Warpspeed, i would be very interested in knowing some more about the VNT turbo's that you have played with. I spoke to garrett australia about these a year ago, and they told me that they had not yet reached the development level for high performnce petrol engines, as they could not get the VNT to be reliable with the ex temps. Why is there a problem with also runnung a wastegate if necessary, or alternatively as a diesel runs, just use a tuned size exhaust housing, and then use the vnt to spool it up nice and early. Have you considered the possibility that the overall size of the ex area was too small and thats why the top end sucked? I havent looked into these 100% yet, due to the lack of range of availability, but i can say for sure the vehicles i have driven with VNT turbo's(BMW X5 and Merc 4wd diesel thing)i have been incredably impressed with given their small capacity diesel motor. I believe these have totally revoloutinised diesels along with single rail direct injection. I believe the ideal turbo for a petrol engined car, would have VNT, possibly a wastegate if necessary, and definately anti surge ports to reduce the surge caused by being able to deliver such high pressure ratios at small flow rates. Imagine a t04z that could spool up as though it had a 0.4 exhaust housing, and then open up to a 1.2 exhaust housing progressively. The only electronic control that i believe would possibly be wanted is to open up the housing when under a certian throttle to reduce ex backpressure when at light/mid throttle, to improve drivability and fuel economy. Maybe VNT isnt the answer, but maybe the sliding partition steup that Holset are also diong would work just as well with better life expectancy. I believe that the reason these turbos aren't on drag cars is because they arent needed, it it easy to spool up enormous turbos these days with electronics, anti lag, high stalls and launch control, but how drivable is a 9 second drag car in terms of lag and rpm/boost threshold, who cares, thats not what its designed for. As for the technology being around for 20 years, not that i dont believe you, but 20 years ago, manufacturers couldnt keep ex housings in 1 piece on pertol cars let alone something this complex, and when they work so well on diesels, why have they only been fitted for the last 3 years? These really are questions i would like answers to, im not trying to be a smart arse, and if you could point me in the direction of where to find imfo about this it would be much appreciated.

end rant

Edited May 30, 2007 by Adriano

[h2k]

I found a reasonable RB25DET manifold from OBX Racing with pulse splitting design:

It only costs about 300 USD wich is a bit more expensive then XS-Power cheap-o-manifolds but again alot more reasonable then the more expensive HKS and TOMEI stuff.

They look alot better then XS-Power stuff on the pictures, have anybody had any experience with this brand OBX Racing and these headers?

Edited May 30, 2007 by h2k

[discopotato03]

Hi all , whilst looking around at NASOIC (Sp ?) North American Subaru site I found this . Still too illiterate to post links but search this thread - New Exhaust Manifold by Full-Race for Subies !

The first page sets the pace then on the third Geoff Raicers goes into detail of what materials he uses and the kind of welding techniques to make his manifolds long term reliable . I've not read the complete string yet but it looks very interesting so far . The one thing that really made me sit up and take notice was where he mentioned that measuring manifold runner lengths is not the way to go about assessing its worth . He mentions that its the pressure drop from the port to the nozzle (turbine housing inducer inlet) that tells the story , so even pressure drops rather than pipe lengths . Subarus flat fours ie EJ20/22/25 are a prime example of uneven length manifolds with the turbo usually high behind the offside head and the manifold runners from the nearside head passing round the front of the sump .

This makes me wonder how well manufacturers have thought about this with std manifolds - possibly the lengths are all over the place but with cross sectional area variations and a turbine housing fitted for testing there may be method in madness .

BTW I'm going to try and get Geoff to read some of these like threads and have some input to the whole twin scroll system , he has that much test data from applying this system to lots of different engines in a multitude of cylinder numbers and capacities that its difficult to dispute it . Before I do I really hope the mood does not get caustic because people are going to believe what their comfortable with and to maintain some decorum sometimes we'll have to agree to differ .

Such is life , cheers Adrian .

[Dale FZ1]

O/k guys, split pulse housings can sometimes help, and sometimes make no difference.

A much better idea is to use a single entry housing with six individual runners correctly aimed into the turbine. This has been proved time after time as being the optimum design for an inline six.

You could be right - I've maintained the "systems" view. It would surely be dependent on how the manifold/turbine work TOGETHER, rather than chasing some technical advantage in one component. If there is some design/specification/manufacturing aspect in one or the other that works against the other, or is not geared to meet the performance requirements (ie: response/power balance) of the vehicle then the good idea may come to nothing.

I found a reasonable RB25DET manifold from OBX Racing with pulse splitting design:

It only costs about 300 USD wich is a bit more expensive then XS-Power cheap-o-manifolds but again alot more reasonable then the more expensive HKS and TOMEI stuff.

They look alot better then XS-Power stuff on the pictures, have anybody had any experience with this brand OBX Racing and these headers?

They sure LOOK ok! Got a link? I can't find anything from them to suit an RB25 like that with a Google search.

This makes me wonder how well manufacturers have thought about this with std manifolds - possibly the lengths are all over the place but with cross sectional area variations and a turbine housing fitted for testing there may be method in madness .

BTW I'm going to try and get Geoff to read some of these like threads and have some input to the whole twin scroll system , he has that much test data from applying this system to lots of different engines in a multitude of cylinder numbers and capacities that its difficult to dispute it . Before I do I really hope the mood does not get caustic because people are going to believe what their comfortable with and to maintain some decorum sometimes we'll have to agree to differ .

The idea of varying diameter as a function of length to maintain consistent pressure is not at all bad. I'd say hydraulic engineers working out specs for stormwater drains etc would probably play with this sort of stuff pretty regularly. Varying the cross sectional area would possibly affect the speed of the pressure wave exiting the exhaust port, and might keep the timing of the pulse at the critical junctions nice and even.

BUT I'd think there is a heap of mathmatics and probably simulation work required to get it right, or risk a lot of wasted time and $$ in trial and error.

I'd say yes, please ask Geoff to comment here. He's one bloke who's working with the design and should be listened to.

I don't see any need to keep this thread as anything other than a discussion that we can LEARN from.