Not really, I just google around a bit.

 

I'd go series sequential. Turbos work on PR's not pressure abs, so if you take for example a TD04 capable of flowing air for 280bhp at a PR of 2.0 (assuming air inlet at 1.0 abs)
Take the same turbo with an air inlet of 2.0abs and still operating at a PR of 2.0 then you have some serious airflow !
The limitation comes in the form of the exhaust turbine efficiency, a std TD04 turbine would not provide the motive force required in the above example, however it would still be capable of producing more that 280bhp worth of air as it would be operating in a higher efficiency zone.

 

Eh? I don't think so. With a series setup why would you want a PR of 4 with a flow of only 360cfm on a 2.5L? Or have I totally misunderstood what your saying Andy?

I don't think the TD04 is a good candidate for a twin (compound or parallel) setup. Not enough PR for parallel, not enough flow for compound. If I was going to redesign the sequential twin-turbo I'd look at parallel GT22 52 trims - PR up to 3 and flow up to 30lb/min each. There's not many small turbos with a good enough PR to be useful in a twin setup.

What are the goals though?

 

My goal would be a very wide and very tall power band.

TD04-13G will do about 350 CFM at PR 2.5, say 1.4 bar boost, but the Subaru TD04s are actually 13T I believe which have nearly 20% more flow at PR 2.0 than the 13G, although there are no public domain compressor maps. So we might be looking at 29lb for each TD04, I would like it to be able to do about 500 BHP.

 

Not that far behind the GT22 then; PR of 2.75 vs 3. But both are a bit short on PR for 500hp.

 

I'm not suggesting a PR of 4 however if you take a td04 and run the suction at say 1 bar gauge and it will flow MUCH more than 360cfm. Remember the std cfm rating is of free air based on an atmospheric suction pressure.

Still not ideal for a 2.5 IMO but the effects/benefits of pressurising the inlet of the secondary turbo in a series sequential set up should not be overlooked.

Andy

 

Interesting. I'll take your word for it, for now

 

2.8^0.5=1.67 need to run each at PR 1.67 bar to get 2.8 bar absolute total. Trouble is the compressor map is a bit lacking at only PR 1.67.

That 535d setup looks really neat...

 

My present engine has been in just over a year and has done about 16000 miles. I think the headgaskets are gracefully letting go again which is why I've just turned the boost down to 1 bar and it should be staying there until the engine is pulled for a teardown, at least to check the gaskets. This one has been babied although has had up to 25 PSI through the midrange. At high boost for long periods it is getting the water a bit greasy and pushing the header tank a bit more down - nothing dramatic. Using a bit more oil as time goes on, and the exhaust is sometimes a bit steamy/smoky. There is no mayo in the oil. I think the bearings and pistons will be OK but we'll see.

I'm getting another daily driver anyway - used E46 M3 SMG (difficult to get a nice one I'm finding) or a new STI (which would be intending only to have 20G, remap, exhaust and fuel pump).

I've had my fun trying to run c. 450 BHP, sometimes a bit more, sometimes less, and now it is getting to the stage where I want something more reliable as my daily driver. Don't know if I'll keep the Scooby as a toy or not, probably will depend on what is found on tear down.

I still think then that 400/400 is a sensible daily driver limit if you want hardly any worries. Even then some are breaking at that, although usually EJ257 pistons I think.

 

But are they breaking because you keep pushing them? If you set it at a nice level and it runs well, couldn't you stop playing?

F

 

I could use studs and a new Cometic HP gasket. When this engine in whatever car is retired from use as a daily driver it will drink regular methanol and run 27 PSI

 

It is properly pushing the water out at only 20 PSI with a hard drive now. It pushes enough out so that it runs low on water.

So pretty sure it is another gasket unless there is anything else to notice when torn down.

 

Damn!!

 

It is OK, I've been trying to close a deal on an M3 over the last few days and plan to keep the Scooby as a toy...

I will give it a few hours to cool down and then top up the water and see if it holds it for off boost driving. Saves me hiring a car until I sort the M3.

Copied from NASIOC post to give details and history:

I've done 16000 miles on the latest EJ257/EJ207 hybrid. The heads are from an STi v5 (GC8).

I need to tear it down to have a look, but this latest setup has run only 98 RON fuel with no additives. It has had a bolt on GT30-11 for about half of it, and a rotated GT30R-12 for the other half, both usually run at about 20 PSI, but the -12 has seen up to 25 PSI in the midrange. I use the car every day

For some months it has been pushing the expansion tank level up under hard use/high boost and I've seen some bubbles, but if I run the header tank half full when cold I could keep it all contained and not have to top it up. More recently the water has become slightly dirty/oily but the oil level has not dropped apart from the usual top up required presumably from blow by with hard use. I think this contamination is from the combustion chamber under high pressure.

On this setup it never really had much det at all, ran conservative timing in the teens, AFR was low 11s. Although I changed after a few early symptoms, it did have water wetter, cooler stat, thicker rad and a weaker coolant mix.

I have previously had trouble on my previous EJ257 block with the same heads with two headgasket failures, but they have been checked for flatness on each change. I ran my EJ205 at higher specific output for longer with no such problems. I've also had studs as well as bolts to hold the heads down.

I wonder if the EJ257/STI 5 head combination is at its limit for gasket seal at the level I've been running at?

 

Who else has had failures with gaskets on the 2.5 John? Can't you wire ring them or something?

F

 

There are a couple of others, but I don't know of anyone that has run these sorts of outputs for 16000 miles with this combination without a failure.

 

FWIW, My 2.5 has just completed 24K in 11 months as a daily driver without a problem but also without the power levels, 350/350 @ 1.2 bar & 370/365 @ 1.4bar. In that time it's done maybe 10 track days.

Don't know what the head gaskets are as the build was done by API using the original STi3 heads, etc.

Gerry

 

Gerry,
Sound simalar to mine
16k miles in 12 months (13k miles running 1.3Bar 376/372) on STI3 heads and a TD05 20g. Daily driver and 6-7 launches at SSO.
Not quite the hard use John gives his

 

Mmmmmm seems that around that 1.3bar mark these ej257's look very happy.

I've done about 16k km's on mine and am runing 1.3bar which gives about 400/370, its never used any oil or water and still seems in good health.

I've looked through the data logs i've done and there works out to be about 600-700 pulls to 7k but thats dose include some normal thrashing as i have my ecu logging at anything over 3.5k.

 

You guys are all running them at the best level I think - 20G with EJ257 is really the best compromise I think, lovely response and not an engine killer.

 

At what boost John, 1.4?
If so, will that deliver a 400 x 400 engine on Optimax with NF? More interested in the torque and where in the rev range.
I'm assuming the normal 550's/Gruppe S/ decat 3'' etc.

Graham.

 

I think so Graham.

Car now retired as daily driver, M3 bought

Maybe I should start a new thread, 2.0 bar alcohol fuelled adrenaline slag

 

I am not sure if I am excited to see what you do with it now it is not your daily drive John or disappointed that you have had to retire it from service and buy some bmw rubbish (yeah okay it is not rubbish.. lol)

Best of luck anyway
Simon

 

Shame that Subaru don't make EJ22T blocks I could have some real fun with one of those

 

Make your own?

 

From an EJ20G? Been there three times IIRC But yes it might be the way to go.

I'm sure Quirt was talking about casting a proprietary block design.... now that could be VERY interesting.

 

Time to pay Quirt some money then..

 

Graham,
I think 400/400 on an EJ257 with a TD05-20g is easily possible

My figures were from Scoobyclinic when Pat mapped it with a seriously slipping clutch
I did tell him to map it conservatively as it was my daily driver, but with a more aggressive map, a bit more boost and some NF you should see 400 all round and still not be pushing it to it's limits
Mine hits 1Bar @ about 3000 revs and just keeps pulling (Should do nicely for a hill climber)

 

Surely before 3000 RPM in a high gear? That is when my GT30R hits a bar... 20G was at 2500 RPM.

 

John,

will we be seeing you at Knockhill on Sunday then

I didn't get my run in your car last time it was out.

John