219 replies to this topic

[acres]

Just thought I would share what I have been up to over the past couple of weeks.
Rather than travelling down the tried and trusted path of binning the twins and strapping on a massive single I thought I might try and see just how far I can take the twin setup. This is just the first step and involves the 2 new turbos and intake pipe. Next step will be fitting a dual core vertical FMIC, bigger IC valve and I have an idea for a set of headers that I would like to try.



This is the 2 OEM turbos on top and the new hybrids below - secondary's on the left, primaries on the right. OEM is VF18/19 and the hybrids are TD04-13T/VF25 primary and TD04-13T/VF14 secondary. I have kept the wastegate on the secondary for future use. (might be necessary after the headers are done)



VF18 on the left, hybrid on the right. This clearly shows the difference in compressor size between the two.



Again, VF18 on the left, hybrid on the right. You can see the turbine wheel is much larger so flow should improve and it will be capable of driving the larger 13T comp wheel. The turbine wheel has been replaced with a TD04 wheel rather than the TD04L wheel which is used from factory. The 2 wheels share almost identical exducer's and inducer's however the standard TD04 wheel has a slightly shorter tip height. This reduces weight of the wheel and flow capabilites which should help reduce lag on spool up, particularly at changeover.



This is the 2 primaries again. Bit hard to tell but you can see that VF25 turbine housing on the hybrid is much smaller than the VF18. This reduced AR ratio should help with spool up as the TD04 core is journal bearing as opposed to the VF18's ball bearing core, also the physically larger wheels will now be much harder to drive too.



New intake pipe. I have made it from 2.5 inch mild steel and 63mm radiator hose and coated with a layer of high temp engine enamel. This is a total pain in the arse to fit as there is only just enough room for the standard plastic pipe as it is. I have had to delete plumback section for the BOV just before the primary, and also the crankcase breather that plumbs back into the OEM pipe just under the TB. This will insted be rerouted to a tee piece between the secondary bypass valve and intake pipe. PCV valve will also be deleted as I can't be bothered rerouting those hoses too (there just isn't enough room to work with). The 3 smaller pieces up top are 2 inch mild steel and are used from the compressor outlets -> TMIC.

The maths I have done suggest that airflow from these 2 puppies should be ample for a good power figure but I am in no way going to make any predictions yet. Just going to see if it actually works first - the turbos might respond poorly and it will just be a real pig to drive

P.S. apologies about the quality of pics, apparently the card in the camera was corrupted so I have just used my phone.

Ta
Aaron

[Alex]

Nice work.

Some people rattle on about reinventing the TT wheel, and you are one of few who have actually put in enough effort to see some positive result.

Most chuck on VF33/32, bang on with some half arsed shit, and fail miserably. (hai Jim!)


Anyways.

Your hybrids look interesting. Wouldn't it have been more beneficial to pursue Garrett parts (namely BB cores) instead of the Mitsubishi gear? I understand the Mitsubishi stuff is more cost effective (especially the TD04 stuff which is nearly throw away in the aftermarket world).

Common theme though, is that every TT modder I've met, are possibly too focused on cost effectiveness...

So how will you run this shebang in terms of ECU?

[acres]

Benefit to the TD04 is that the core is the same diameter as the VF series which are found on the TT's. The only machining that is required is to bore out the housing to fit the larger turbine wheels. Once this is machined out to suit, the turbine housing simply slides onto the TD core like it was meant to be there.
The Garret cores is something that I looked at but to be honest I couldn't get a straight answer from anybody whether they could actually modify the turbos with garret cores. I must have spoken to a dozen different turbo specialists and they all gave me they same doe eyed look of "that sounds like hard work for me". In the end I gave up and just went with the TD's.
Just using of the shelf Garretts wont work either as any turbo that has a large enough compressor to be an upgrade, also has a much larger turbine housing. Being just a 2L (or 2 x 1L in twin mode) exhaust gas to drive the turbines is at a premium so the original housings have to be retained.
Cost is a big thing as I refuse to spend a cent of credit on a car, rather only cash that I have spare, but that's not why I went for the MHI.

Will see how the stock ecu holds up first, then I will probably just go piggyback depending on whether the seq points can be modified.

[RX25SE]

Nice to see someone thinking outside the square and taking the road less traveled. Props to you.

I have to admit, if I ever bought a TT car, I could see myself fiddling with the twins.

Keep us posted!

[acres]

First post edited. It was the PCV valve on the intake manifold that I am deleting, not the ECV. This will still be a sequential setup and not parallel.

Thanks mate. I like the theory behind the TT setup so trying to make it better is an interesting and enjoyable challenge, albeit frustrating at times.

[Trent]

Keen to see the outcome. Hope it all goes well.

[Soop]

Very interesting.
Good to see something different.

I can see this having good results if you continue to run the same boost pressure. Less heat ftw.

[natho]

as someone who was considering doing just this, i suggest you look into the difference between the subaru TT setup and the 2JZ TT setup and attempt to apply some of what toyota did - it should help improve the VOD, which i fear may be an issue with bigger turbos, depending on exactly how you set them up.

[BigWillieStyles]

Looks great, what other mods are you running along side the hybrids? Are you going to up the boost? Making power with the sequential systme is easy, making it linear seems to be the challenge.

In my opinion, Subaru could have made a sequential system that didnt have VOD, just as was created for the Mazda Rx7, Supra, Porsche 959. But for some reason they didnt. They were obviously restricted by fincical costs trying to keep the overall price of the car down, while also trying to deliver a car that was economical and met local power and emission regulations. The Australian B4 was just a modified version of what they got in Japan, and as such, it would have been too expensive to redo the whole system.

[Soop]

How different is the TT setup to the JDM version?
I was under the impression they worked more or less the same, with a tune being altered to suit our crap fuel

[acres]

The systems are exactly the same between adm and jdm, just the tunes for differing fuel standards. The reason for VOD on the Subaru TT comes down to one word - capacity. We are essentially working with 2 x 1L 2 cylinder motors from 4000 rpm onwards. This gives minimal exhaust gas and pulses to drive the turbines which is why it takes so long to bring the secondary online. I believe this can be combated somewhat with a set of headers that will even out the exhaust pulses, ie. L, R, L, R as opposed to the current L, L, R, R setup. Also by lifting the point at which the ecv is fully opened and the ic valve is opened by another 500 - 1000 rpm should assist with spool up of the secondary. This is not practical to do with the oem turbos as the VF18 simply doesn't have the airflow capabilities to safely feed the engine by itself past 4000rpm but with the larger 13T comp wheel from the TD04 it should have no trouble running out to 5000rpm by itself.

Boost won't be touched until after the fmic is done and the ecu is sorted (still not sure on whether piggyback or full ecu will be best). But I anticipate that after losses through the I/C I should be able to comfortably run close to 1.5 bar. Airflow from these compressors at this boost level at 7500rpm at ideal temps with a relatively high VE should max out the capabilities of the 2L.

[BigWillieStyles]

Boost won't be touched until after the fmic is done and the ecu is sorted (still not sure on whether piggyback or full ecu will be best). But I anticipate that after losses through the I/C I should be able to comfortably run close to 1.5 bar. Airflow from these compressors at this boost level at 7500rpm at ideal temps with a relatively high VE should max out the capabilities of the 2L.

Thats some healthy amount of boost, nearly 22psi, VOD will be greatly reduced.

How are you going to get the ECV to open earlier? Will a zerosports sequential controlled be able to help with this?

[acres]

Thats some healthy amount of boost, nearly 22psi, VOD will be greatly reduced.

How are you going to get the ECV to open earlier? Will a zerosports sequential controlled be able to help with this?

Idea is to have it fully open at a higher rpm not earlier. Try to capitalise on the extra gases and pulses from the higher engine speed. Also keen on improving pre-spool by eliminating the use of the primary wastegate until after the ecv is fully open and using the ecv itself as the wastegate while it is operating on primary only. Also plan on delaying the I/C valve from opening until a higher rpm too. The I/C valve will be placed downstream from the FMIC (dual core setup) as close to the TB as possible to help elimante any "lag" caused by the pipework/FMIC. Also plan on placing a second valve within the crossover pipe in the headers which will close when the ecv opens to allow smoother and even exhaust flow - hence adding the extra wastegate to the secondary turbo as each turbo will then nedd to control its own boost.

How will I control this? Not sure to be honest, I have already trialled re-routing some of the hoses and solenoids in the black box to try and simulate these effects with some degree of success but I will most likely have to use something like the zerosports controller and/or some electronic boost controllers or solenoids. Unless somebody can suggest an ecu with enough outputs to be able to control the TT steup?

None of this can happen though until I have the FMIC and preferably the headers too, so for now I'm just keen on getting this reassembled with the standard TMIC , headers and ecu and seeing just how they respond.

Have you used the zerosports controller at all? Do you know if it can control teh degree of pre-spool too or does it just open and close the ecv?

Thanks for your help
Aaron

[acres]

as someone who was considering doing just this, i suggest you look into the difference between the subaru TT setup and the 2JZ TT setup and attempt to apply some of what toyota did - it should help improve the VOD, which i fear may be an issue with bigger turbos, depending on exactly how you set them up.

In my opinion, Subaru could have made a sequential system that didnt have VOD, just as was created for the Mazda Rx7, Supra, Porsche 959. But for some reason they didnt.

I have honestly never looked to see how these other systems work. Do have any info or links on how they differ that you would be keen to share? I am interested in having a squiz.
Ta

[Alex]

I think I still have my zerosport TT controller in my garage somewhere if you're interested....

I'll have to check tomorrow morning

[acres]

Did you ever use it before you went single Alex? Was it able to control pre-spool too?

[Soop]

Have you considered re-ground cam shafts?
More lift/duration, more air/fuel bigger bang more exhaust gass less lag?

[Alex]

Did you ever use it before you went single Alex? Was it able to control pre-spool too?

Define "pre-spool"...

It basically could change when the changeover occurred.

memory is hazy about its workings.

[acres]

Yep thought about bigger cams but from what I understand the gtb/v3 Sti heads flow very well from factory so I am unsure as to how big a difference it would make with these compressors. What is your opinion? Motor is an ej20r from 96/97 gtb.
With the td04 I always have the option to upgrade the comp wheels for more flow too - 15g, 16t, 18t, 19t are all available.

By pre-spool I mean when the solenoid in the black box begins to operate the positive pressure side of the actuator so as to feed the secondary a small amount of exhaust just to get it rotating before the valve is fully opened. I'm positive this occurs at around 3500rpm, up until this point the ecv is fully closed and boost is controlled via the wastegate. I would like to eliminate the use of the wastegate until the ecv is fully opened and instead activate the pre-spool at a much lower rpm

[Soop]

You'd have to find out what the exact cam specs are for what you already have. Most people leave them standard. Bit boring