Effect of 2010 LGT header vs "regular" header/UP on turbo

I've read a lot of speculation in other threads on what a shorter "evo length header" or 2010 legacy GT style header would add to turbo performance vs a "comparable" EL header in the Spec C, or a APS or Perrin header. Some say the longer runners we have make almost no difference in turbo response or pushing the curve to the right. Others say there would be a noticeable increase in turbo response, and the curve could even get pushed to the left enough to be "worth it". Does anyone have any real world or calculated information on what the difference would be. Any threads out there that cover this already (I have not found any meaningful ones)?

2010

http://efilogics.com/customer/4%20-%20Subaru%20Customers/AlexM_2010LGT/AlexM_2010LGT_10.JPG

One way of doing current version in EL.

http://www.scoobytuner.com/graphics/invidia_header.jpg

well, this would be a question to ask a company that has done it before... Full Race has done 2-3 of these systems well before the 2010 LGT system came out. pretty much across the board the response, boost threshold, and overall delivery is well above what our stock location pieces of crap can do. also being able to use a true turbo instead of using the hacked design that we have in its self is an upgrade.

granted I myself have no real world numbers, but its simple physics. hotter temps (being closer to the motor), shorter outlet lengths (when compared to a FMIC at stock lcoation), and proper turbo configuration will make more power over all the curve... why do you think the 2.0 evo can compete with the 2.5 STI? location and turbo design.

T

well, this would be a question to ask a company that has done it before... Full Race has done 2-3 of these systems well before the 2010 LGT system came out. pretty much across the board the response, boost threshold, and overall delivery is well above what our stock location pieces of crap can do. also being able to use a true turbo instead of using the hacked design that we have in its self is an upgrade.

 

granted I myself have no real world numbers, but its simple physics. hotter temps (being closer to the motor), shorter outlet lengths (when compared to a FMIC at stock lcoation), and proper turbo configuration will make more power over all the curve... why do you think the 2.0 evo can compete with the 2.5 STI? location and turbo design.

 

T

thank you for your post. I've been annoying Geoff with PMs anyways (poor guy) so I will add this to the list

I've been pricing out a turbo in this proposed location.

Scavenge pump and possibly a small sump, and cut oil pan and additional remote oil basin to maintain the 4-5 quarts are the issues. Headers and plumbing and mounting won't be too bad. Need to wrap it and shield it to avoid cracking the turbo in the rain.

Question is whether all this expense will be rewarded with anything other than the fun of it.

Links to FR builds like this?

Scavenge pump and possibly a small sump, and cut oil pan and additional remote oil basin to maintain the 4-5 quarts are the issues.

I wonder if you could use parts from a 2010 LGT to solve those problems.

Check out the attached file.

If there is any way I can be of help in this discussion with part numbers, diagrams, etc., please let me know.

Best,

Rick

p6-11 2010 Mech & Function.pdf

I cant remember what the backhousing design style is.... but yeah, you COULD run the stock LGT header for it, and build a turbo for its location... before my WRX ate it, I was tinkering with the idea (ran outta funds to make it happen) and its overall fairly simple... just need the manifold made, use your desired fitting (T4 maybe?), desired turbo (I was leaning to the BW S2xx), oil hold box (no idea what the actual name is, but need that drain box after the turbo to make the scavenging pump work properly), then the correct coolant and oil lines to feed and re-leave the turbo.

this would be ideal for a FMIC application, just literally pipe the inlet up the radiator in to the engine bay to where teh stock filter currently is, and for the outlet, just across the back of the radiator, U bend to the front of the radiator, into the FMIC core, and then the outlet side of the FMIC is just like all the current kits. Exhaust is cake, just copy the LGT (or any NA version) path into the midpipe, done.

The ONLY piece of the puzzle that I was not able to figure out was what to do with the water system.... being that low, will it still be able to get water and send it back? people keep talking about convection, but I dont see it working with that far of a distance.... but that is a bit out of my realm of study.

T

Thanks Rick. that actually proves two of my previous theories... the oil routing is the exact same as previous, just longer runs, and now pump driven.

Granted its not the best image for the mechanics, but I bet you could use the new scavenging pump as well, but would need the pump, valve cover, and maybe the cam its self.....probably easier to do the electric that you mentioned earlier.

Rick: Do you have any diagrams that show the water system for the turbo? or could it be that it is oil only now?

Cooling Circuit.pdf

Cooling - General Description.pdf

Lubrication - General Description.pdf

p1-2 Intake - General Description.pdf

p3-4 Intake - General Description.pdf

I believe that the Impreza platform will switch to this style of turbo in time too. Better for emissions.

There is no question that all things being equal having a turbo closer to the cylinder heads is better. Just look at the torque the 2010 LGTS make when tuned!!!!

Unfortunately it may be a royal PITA to make something work until the aftermarket steps in, and I doubt they will til it affects the STI, maybe 2013ish?

IIRC I was the first to talk to Geoff at FR about this well over a year ago. but if I where him I wouldn't make anything for the LGT unless he knows the new Impreza platform will be the same, wouldn't make any money off of it!

How about you just sell me your 850's

Maybe.

Check out the attached file.

 

If there is any way I can be of help in this discussion with part numbers, diagrams, etc., please let me know.

Best,

Rick

Thanks!!! You guys are the best!

I cant remember what the backhousing design style is.... but yeah, you COULD run the stock LGT header for it, and build a turbo for its location... before my WRX ate it, I was tinkering with the idea (ran outta funds to make it happen) and its overall fairly simple... just need the manifold made, use your desired fitting (T4 maybe?), desired turbo (I was leaning to the BW S2xx), oil hold box (no idea what the actual name is, but need that drain box after the turbo to make the scavenging pump work properly), then the correct coolant and oil lines to feed and re-leave the turbo.

 

this would be ideal for a FMIC application, just literally pipe the inlet up the radiator in to the engine bay to where teh stock filter currently is, and for the outlet, just across the back of the radiator, U bend to the front of the radiator, into the FMIC core, and then the outlet side of the FMIC is just like all the current kits. Exhaust is cake, just copy the LGT (or any NA version) path into the midpipe, done.

 

The ONLY piece of the puzzle that I was not able to figure out was what to do with the water system.... being that low, will it still be able to get water and send it back? people keep talking about convection, but I dont see it working with that far of a distance.... but that is a bit out of my realm of study.

 

T

Not quite. I took a good look under my car, and I see no way to keep the stock oil pan and still keep the headers equal length without dropping the headers closer to the ground.

I'm looking at a Borg Warner S200SX with T4 divided flange. I've researched it decently and understand it alright, and it appears to be the ideal turbo for my goals over what else I've seen. I do not have a built block, so this turbo allows me to get some good purchase at lower boost levels. Also my trans is a 5eat, and that was a huge part of my turbo decision.

I would have my compressor side on driver side, and hot side on passenger side. Turbine outlet would follow the path of the UP location, over the frame rail/axle/tie rod, and follow the old DP path. 3" all the way. Perhaps 3.5", not sure, don't know much about sizing exhausts. But that's easy stuff to do either way, point is I only see that one way to easily run the exhaust post-turbo. Air filter and compressor inlet can really be run anywhere. driver side is fine. Lastly, I would flip the intake manifold. FMIC hot side on driver side, cold side on passenger side. All short runs this way. Make sense?

I believe that the Impreza platform will switch to this style of turbo in time too. Better for emissions.

 

There is no question that all things being equal having a turbo closer to the cylinder heads is better. Just look at the torque the 2010 LGTS make when tuned!!!!

 

Unfortunately it may be a royal PITA to make something work until the aftermarket steps in, and I doubt they will til it affects the STI, maybe 2013ish?

 

I'd rather pay one-off DIY prices than pay for a retrofit kit from someone, which would have to start at least around $5000. It's either my pain or theirs, and I don't want to pay for their pain.

 

IIRC I was the first to talk to Geoff at FR about this well over a year ago. but if I where him I wouldn't make anything for the LGT unless he knows the new Impreza platform will be the same, wouldn't make any money off of it!

Going to ask him to weigh in on this, since he would know better than most.

I dunno who it was but they did a couple one offs... but then again, if they did that then in theory they should already have a Jig for it right? anyhow, making these small parts is fairly inexpensive. especially when shortening the distance. only thing that would be some money wold be the oil and water lines.

Jig is very easy to make. Easy to see where an engine or oil pan or radiator is in the way. I'm looking forward to that part.

Note: All of this is pending me or someone else actually doing it. If someone else has done it, would love to see any tricks you picked up along the way. Don't hold your breath though. I can't do this now, just interested in learning and planning for now.

BTW I saw one like this on NASIOC, but the turbo appeared to be in front of the radiator, or in place of the radiator, so not really an under-engine mount. Something though.

Locating the turbo closer to the exhaust ports reduces cat lightoff times. This emissions control strategy (moving the cat closer) is being used universally across manufacturers of 4 cylinder engines.

1. Toyota 1ZZ/2ZZ engines have exhaust manifold on rear of engine, versus front on the previous 4AGE

2. Honda K series engines have exhaust manifold on rear of engine, versus front on the previous B/D series

3. Mitsu 4B11 engines have exhaust manifold on rear of engine, versus front on the previous 4G6x engines

So that's the emission side.

The chief advantage performance-wise of the 2010 LGT comes from the use of a twin scroll header. Twin scroll headers separate the exhaust pulses according to the firing order, so that (in the case of a 4 cylinder boxer) each bank's exhaust pulses never interfere with each other. Because they never interfere with each other there is less back pressure for the engine to work against:

http://legacygt.com/forums/attachment.php?attachmentid=85404&stc=1&d=1274938860

http://legacygt.com/forums/attachment.php?attachmentid=85405&stc=1&d=1274938860

http://legacygt.com/forums/attachment.php?attachmentid=85406&stc=1&d=1274938860

http://legacygt.com/forums/attachment.php?attachmentid=85407&stc=1&d=1274938860

That's on a normal inline 4 (like an Evo engine) but you get the idea. So look at the 2010 header:

http://efilogics.com/customer/4%20-%20Subaru%20Customers/AlexM_2010LGT/AlexM_2010LGT_10.JPG

Recall that the firing order is:

Drivers side--------------passengers side

-------------front of car------------------

-------#2-----------------------#1-------

-------#4-----------------------#3-------

Firing order is 1-3-2-4

The 1 and the 2 are connected to one scroll while the 3 and the 4 are connected to the other. The pulses never interfere. Each scroll is pulsed in an alternating fashion.

Mazda actually pioneered the twin scroll turbo 20 years ago on peripheral exhaust port rotary engines (pre Rx-8 Renesis), which only have two exhaust ports and two exhaust runners. The 1989 model Rx-7 turbo has this. Mazda discussed it in an SAE paper. They compared two identical turbos, one with single scroll and one with twin scroll. Both with electronic boost control. Check it out:

http://legacygt.com/forums/attachment.php?attachmentid=85410&d=1274939137

http://legacygt.com/forums/attachment.php?attachmentid=85411&d=1274939137

http://legacygt.com/forums/attachment.php?attachmentid=85409&d=1274939137

Wider torque across the rpm range due to the alternating pulses. Less backpressure up top due to the lack of interference between cylinders (no residual exhaust gas to work against). This is not new technology. It is a proven method of increasing performance.

It is the twin scroll that is the key to the performance! You must separate exhaust pulses according to firing order! The runner length is a secondary aspect. It doesn't matter as much as the pulsing effect and resulting backpressure reduction.

Look at the exhaust manifold design of the twin scroll 4G63 Evos:

http://legacygt.com/forums/attachment.php?attachmentid=85413&stc=1&d=1274940552

http://legacygt.com/forums/attachment.php?attachmentid=85412&stc=1&d=1274940552

because of the firing order, the scrolls are pulsed in alternating fashion. It's probably 1-3-4-2 on the evo. So first the outer passageway receives exhaust gas, then the inner passageway, then the outer, then the inner. The exhaust never interferes. On the peripheral exhaust port rotary engine the firing order is front rotor, rear rotor, front rotor, rear rotor. The manifold is divided and the two exhaust pulses (Rx-8's are different and have 4 exhaust ports with 3 runners) never interfere. This principle that Mazda implemented 20 years ago remains today.

wow... holy perspective batman. Thank you.

On a different note, what I was thinking for the scavenge pump / turbo oil draining is:

Option B: have an upside down 'T' instead. The oil would drain and would T off two directions. One way is to the scavenge pump. Other way is to a breather (not sure if this is needed on a journal bearing BW turbo that uses a lot more oil than a Garrett). Some people have noted that the pumps will create an excessive vacuum and will "suck the oil out of the turbo" or something like that. Makes sense since these pumps all move way more fluid than the turbo will drain. Air must make up the difference, yes? So the turbo oil outlet tube should have a breather to allow drain tube pressure to be more towards normal. However, I would dare say the majority of turbo oil scavenge setups don't have a breather and are just fine. So I don't know.

Not sure whether a oil sump is really needed under the turbo either. Not sure what benefit it would yield. The BMW guys seem to use them a lot. STS doesn't, and STS stuff lasts a long while AFAIK.

Option A: Dedicated small turbo oil sump.

http://img166.imageshack.us/img166/1287/v2ct9.jpg

Not sure if this has one or if it's something else, but you get the idea.

http://legacygt.com/forums/attachment.php?attachmentid=85415&stc=1&d=1274961369

The "sliding structure" may be referring to a two-layer system in the exhaust manifold, a sort of inner and outer shell to retain heat. This is another trend among the OEM's to retain heat for the cat (and preserve energy for the turbo). Mazda began using this in the Rx-8 to improve cat lightoff:

http://legacygt.com/forums/attachment.php?attachmentid=85417&stc=1&d=1274961369

The Taurus SHO uses double layer turbo manifolds for its twin turbos:

http://legacygt.com/forums/attachment.php?attachmentid=85416&stc=1&d=1274961369

The Subaru diagram isn't the best but it seems likely to me that they are in fact utilizing the dual layer design. I think you'd be hard-pressed to find an aftermarket manifold that can meet or exceed the engineering and quality of the Subaru OEM twin scroll header.

Not 100% relevant to the topic but related to the BP/BL JDM twinscroll... Here is a pic of the JDM EL headers I run with my twinscroll TD04HLA-19T (not my pic, but identical minus the heatshields).

http://legacygt.com/forums/attachment.php?attachmentid=78420&d=1262190618

There is a ton of info about the turbos capability in my (Underdog's) "JDM Twinscroll Project" thread, although with my current mods I'm certainly not using it to it's full capability. Anyways, my point is that with this setup I have boost on tap at pretty much any RPM in any gear, including 1st. Boost response is as close to instantaneous as I can imagine and my boost gauge needle seems to follow my left foot directly.

So, even though the exhaust pulses are being routed up around the passenger-side head to the stock location, I feel that there is still a considerable benefit to the EL twinscroll header/up combo (in conjunction with the turbo of course).

Food for thought.

I knew Boostin would be on here.

I have a few theories I want to throw in as well.

ie: FR has a very nice TS manifold. But I have yet to see it perform notably well on smaller set-ups. On a 40R turbo recently on NASIOC it spools up fantastic for a 40R. (spool at around 4000 rpm) I believe this is because the extra VOLUME required in our set-ups. That nice pulsing affect shown here:

http://legacygt.com/forums/attachment.php?attachmentid=85406&stc=1&d=1274938860

Is reduced the farther away from the cylinder head we position the turbo. Now of course the pulsing affect is amplified by the higher rpms a 40R turbo spools at. But it doesn't seem to work exceptionally great (I didn't say bad) at 3000 rpms where we would like the smaller TS turbos to spool.

Now, have a normal TS manifold, where the turbo is close to the head, and you have a recipe for a great combo! I think a 35R turbo on a 2.5 liter would spool at 3500 rpm or faster (vs 4000+). Better heat retention, and better pulsing FTW!

Is there an easier way?

Drop in a 2010 motor and swap the turbo?

Drop in a 2010 motor and swap the turbo?

 

The engine mounts are in the same place or are they different? It looked to me like nothing would line up . What is the correct answer?