The AUSubaru.com.au Forums

Here is my step by step plan for fitting a TD04- standard wrx turbo- to an ea82t engine and L series shell.

this mod is not quite finished yet but i will update as i go.

Materials needed:
-basic tools like socket set, pliers and screwdrivers.
- a big awesome pair of circlip/snap ring pliers- the type to squeeze together
-grinder
-vice
-welder if you plan on doing the zorst pipes yourself
-TD04 - very similar turbo to original VF4 just a little bigger. i got mine off a forum for $100
-std wrx/rs lib turbo downpipe and up pipe for their flanges to suit the td04's flanges.
- about 2-3 inches of fuel hose to extend the skinny oil line that goes to the top of the turbo
- longer coolant hose- i think about 40 odd cm will be needed.this one replaces the small line coming from the bottom of the engine running with the up pipe.
- random other vacuum hoses need to hook up. its also a good time to replace any other rubber hoses near the turbo whilst its off.

u will need most of this stuff plus a little more.




Step 1) Remove old turbo and dump pipe. take care not to snap any other hoses or oil lines.


Step 2) remove any intercoolers that may be attached. note-you will need one after with the new turbo to dissipate the increased compression heat.
engine bay should look like this



Step 3) Clocking/Re-Clocking the TD04.

in the above photos notice the difference in the exhaust outlets and compressed air outpipe/'big silver pipe', between the standard turbo on left and td04 on right. i should of taken the photo with the exhast ports the same way but with mental rotation you can see that the 'big silver pipe' will be rotated 90 degrees further than the stocko turbo one. so to make it point (almost) straight up to into existing intercooler, instead of pointing into the engine you need to reclock it.
this involves undoing the comporessor side of turbo and spinning it and doing it back up.this process is done as follows.
note: the exhaust side of turbo can also be removed apparently by removing centre holding clamp and lots of love taps from a hammer.


Step 3a) unbolt gold wastegate bracket- you should be able to line up the bottom hole of bracket with the top hole on turbo. this is where it will bolt up once clocked. one bolt is enough for wastegate to function properly apparently.

3b) put turbo in vice compressor/silver side down. you will see a giant metal circlip lurking in there which needs to come out. its a bitch so just persist with it. a small flathead helps here.
note: some turbos have this circlip and stuff to clock it to any angle and others have a few bolts evenly spaced for set rotations.

there is also a locating pin inside which couldnt see before which i snapped it off in the process of getting the clip out. REMOVE this locating pin by tapping each side of it with a hammer and screwdriver. else the compressor housing wont go back on properly.

3c) the whole compressor housing is now removeable, like pic below, (it shouldnt need any love taps from the hammer) but you just need to spin it 90 degrees so the wastegate hole lines up as said above. bolt it up loosely and tighten later.


3c) get the biatch clip back on. this helped me check that the turbine shaft still spins freely. else you have a problem.


3d)bolt turbo bits back up.

turbo will now look like this with all exhaust ports and 'big silver pipe' lined up to match the old turbo.

Step 4) time to cut flanges off wrx up and down pipes. now as the td04 is slightly larger than the stocko it hits a little on the engine. so what ya have to do is space it up about about 2 inches so that nothing fouls the engine. so this means leave a bit of pipe length on your flange cuts so that these are extended when you weld it on.
like this pic of the up pipe below *not mine*


note: for the up pipe flange, the stock ea82t one will fit with a minor adjustment made to the furthest stud. it needs to be moved about 1 cm outwards as in pic below *again not mine*- this dude has also added a stud but im pretty sure its not needed.


Step 5) this is where im up to atm. i plan to use lots of the wrx up pipe so that mine is fatter then the stocko skinny pipe and like the above picture.

next week i am towing car to the workshop and getting them to weld the pipes up. hopefully shouldnt cost too much. or take too long.
i am going to change my intercooler to another bigger unit and whack a new bov on because not having one before contributed to my turbo dying. and also i get cool psssht sounds. but no more fluttersluut

i will also try to squeeze in a tune before shops close for chrissy because i am running an aftermarket ecu im guessing it will run like a pig.

more to come as it happens

looks like you got it sussed matto. great idea

well my turbo conversion is all finished. the zorst shop got it done in 4hours, and a few hundred later i was hooking up the rest of the pipes to make it fit. was all good till my mate n i started it up and then it pissed out some oil from the bottom banjo bolt on the motor. after running round for new one and new oil, we drained oil and put it in but to no avail. the shop locked the car in as it was closing time. this morning i used some thread locker and she was all good. dont know how long it'll last tho.

this is the end result



sounds different now spooling up, and its noticeably smoother to boost up too- not like the instant surge of the stocko on 10psi still pulls hard tho. not a huge difference to the stocko but noticeable.
ill have to see how it runs now with my top mount put on straight- had to drive it home askew with like 1/4 exposed to scoop. and also im getting a bigger replacement top mount with a blow off valve. and also a tune to make it even better. cant wait.

after a little bit of driving on it - its great btw- ive upped the boost a little bit to 9si and put a bigger intercooler on, and there is a good difference in this. i just need to get it tuned now to stop the surging.

Im having problems with surging too (i did a TF035 swap about 3 weeks ago). But im using stock ECU.

Must have different air flows at different RPMs

well that didnt last long. my td04 has crapped itself already! bah! bloody shaft play.

btw EA82t's suck having no boost.

ditto my last comment x2.

something has gone wrong in this swap. the last 2 turbos have lasted less than a month each. and i cant find out why. the only thing i can think of is the oil pump not providing enough pressure for the slightly bigger turbo then the bearings overheating and crapping themselves.

I suggest you strip them down and see if the bearings and turbine shaft journals are worn and or the materials overheated .

I have not seen if you people are getting the water cooling system right either . The top hose must NOT form a high point or air trap because if it does the turbos water cooling system will NOT thermosyphon when the engines shut off .
Also are you using the Std VF2/4 turbos water plumbing fittings and banjo bolts ?

Also something to remember is the EA82T's headers internal restrictions .

By changing to a less restrictive turbo and dump/exhaust the engine may be running leaner and having higher exhaust gas temperature . So many things can damage turbos nowdays .

Also BTW , if anyone wants to sell me a dead TD04 I'm looking .

Cheers A .

discopotato03 wrote:I suggest you strip them down and see if the bearings and turbine shaft journals are worn and or the materials overheated .

I have not seen if you people are getting the water cooling system right either . The top hose must NOT form a high point or air trap because if it does the turbos water cooling system will NOT thermosyphon when the engines shut off .
Also are you using the Std VF2/4 turbos water plumbing fittings and banjo bolts ?

Also something to remember is the EA82T's headers internal restrictions .

By changing to a less restrictive turbo and dump/exhaust the engine may be running leaner and having higher exhaust gas temperature . So many things can damage turbos nowdays .

Also BTW , if anyone wants to sell me a dead TD04 I'm looking .

Cheers A .

how many do you want? ive got a few

man i wish i had known that before because the coolant line had to be lengthened and does come in from the top. oops

mm yeh this is why the Lib cooling system has the radiator header tank as the highest point in the system, so bubbles form there rather than on the turbo - could be the culplret.

discopotato03 wrote:I have not seen if you people are getting the water cooling system right either . The top hose must NOT form a high point or air trap because if it does the turbos water cooling system will NOT thermosyphon when the engines shut off .
Also are you using the Std VF2/4 turbos water plumbing fittings and banjo bolts ?

So do you reckon this would be able to kill a previously good TD04?

i had a quick look at Subafury's today, and the Top hose was going fairly high, probably about as high as the rad cap, and with coolant levels usually not being as high as the rad cap, it wuld surely be creating a problem, no doubt there would be an air pocket there which will limit or stop flow of coolant through the Turbo.


i could definetely beleive that lack of coolant would seriously affect the cooling of the TD04, being that the oil doesnt have as effective cooling like the water system.
Also with the shocking header design increasing exhaust temps even more before it gets to the turbo, means greater turbo heat, meaning it will reduce the characteristics of the Oil (less lubrication, less detergent like properties - ie wont 'clean' the surfaces, increased heat, Shortened life of bearings/materials/metals/Actual Turbo's etc.)


The only real problem with the old turbos seems to be that they develop a serious amount of 'shaft play' (one probably moving around 1cm each side - which is a damn lot
the shaft play will obviously affect the turbines ability to spin altogether but most cases not fast enough i expect to produce any more boost (as they keep hitting the walls or stop dead).


Of course to actually find otu if all the problem is, is infact that Return line that would involve buying yet another TD04, which would be a real paint o have it die again.

(also having a radiator which doesnt keep its level full would be a problem to)

your input would be appreciated Adrian

ditto what matatak said. i was gonna post that.

I have always fount the easiest way to keep a radiator full is to use an overflow bottle and a rad cap that allows water for the overflow reservoir to return to the system as it cools down .

Mostly the water cooling jacket in the turbo is there to keep the conducted heat in the turbine/turbine housing/exhaust manifold (many are cast iron) low enough to prevent the oil from baking itself onto the turbine shafts bearing journals . When the engine is running you have oil being pumped into the bearings but once the engine stops so does the oil pump .

Water has a high thermal mass meaning it can absorb a lot of heat before it boils and becomes a gas - water vapour .
Ideally how the turbos water cooling system works is this .

Water supply from engines cooling system hopefully somewhere down low where it's temperature should be low - for an engine .
Golden Rule : Water MUST be able to run uphill to the turbos water inlet and then preferably uphill from the outlet and return to the cooling system at a high point .

Process : Engine is shut down hot so water pump stops . Water in the turbos cooling jacket absorbs local heat and begins to boil . The steam bubbles and hot water rise out of the turbo and travel up to the high point in the cooling system . The heated water is lighter (less dense) than the cooler water below so it will rise and be replaced by cooler water from below (lower down) .
It becomes a constant cycle for about the critical 2-3 minutes it takes to bring the turbos core down to a little below 100c which as we know is waters boiling point .
A properly thermosyphoning system is completly automatic and requires no pump or power source , how good is that !
My FJ20 powered Bluebird was set up like that and you could hear the muffled gurgling of the water boiling in the turbo and the bubbles etc gently rattling up the outlet pipe (rubber) and returning to the engine on the engine side of the thermostat (EA82T's are like this too) .
Provided the thermostat is lower than the radiators top tank the post turbo bubbles flow through the open thermostats valve and up the top hose to the radiator and via the pressure relief radiator cap to the overflow bottle .
I'm not familiar with Liberty RS's but they would have that water reservoir there to become the engines cooling system high point and to prevent gas bubbles getting trapped in the engines or turbos cooling lines and getting localised hot spots where the gass displaces cooling water . Sierra Cosworths and RS500's have a similar type of thing .

Where turbo damage can occur is that when overheated the lube oil bakes itself to the journal bearings they are not smooth any more so when the engine is started from cold the less than perfect surface gives to not ferrous bearing bushes a real caining . The hole through the bushes opens up so the shaft instead of spinning concentrically starts to get an "orbit" effect which wears the bearing bushes faster and eventually the turbine or compressor blades have a big rub on their housings and things start to go west fairly quickly . The whole idea of the turbos water cooling system was to prevent it getting hot enough to cook the oil onto its shaft/bearings when shut down hot .

Also second hand turbochargers are an unknown quantity , they can feel ok by wiggling the compressor wheel but thats no way to tell if it's been abused by overspeeding/overheating/using cheap oil and not changing it regularly .
To be 100% sure means buying a brand newin or getting one cheap 2nd hand and fitting a new cartridge between the existing housings .
I know I know people get them rebuilt but bearing housing bores wear and so does the turbine piston ring seals bore .

Remember how I recently said that just screwing up the boost can be dancing with the devil ? Turbochargers are a high speed device but when run as designed can lead long lives . When the boost is screwed up the turbine/compressor speed goes up and almost always so does the engines heat output (EGT) .

Run out of time again , cheers A .

info is greatly appreciated Adrian- i understand now how it all works in there. And also from pulling one apart the other day.

although i dont think it will solve my problem.

just spoke to andrewt and he said when he did his first ej20g conversion on the now monsterwgn he didnt even connect up the turbo's water lines- just bypassed it quick and then never remembered to fix it and put on his extra turbo tank with cap. that turbo has copped some abuse and was running fine when he took it out.

so that makes me think my previously high coolant line isnt so bad- it still had some coolant going through it.
maybe it can be blamed on oil that is almost ready to be changed, i dunno.
One thing is for sure i cant afford to keep spending money on this old engine. I need something easy and reliable - which is why im going ej22.
this motor will live and breathe again properly. i have a plan for it down the track.

The trouble with hybrid turbo conversions is that you go beyond the manufacturers parameters so the safety zones (AFR/EGT/Turbine Inlet Pressure) and the results are not always pretty .
Unless you have the ability to measure these things and make sure the results are in the safe region things can get out of hand . Because the engine can't talk it can't say I hurt .

It will be same deal with an EJ in a L/MY because the exact conditions are not the same as a Lib or Impreza . The pain is generally less with an NA engine because they are usually less thermally challenged .

Gotta run - steel train awaits , cheers A .