Just how much potential has the 2.0 impreza got ?

I can afford to buy and insure a turbo model but this project is to give inspiration to the non turbo owners that good power can be had from a standard 2.0.

Lets do something different Subaru did do a concept supercharged impreza car but it never went into production which is a shame.

I did see some supercharged impreza's done across the pond in America with varying success so I decided to take the project on.

choice of supercharger ?

eaton was a good choice and reliable I could have gone with the m62 however noise was an issue as it would have to be run a such a speed that the supercharged would have been audible although it would have been able to create enough displacement I decided to go for the m90 it would be able to create more than enough displacement even running at less than half speed making it extremely quiet so no supercharger whine and the capability to increase boost further into the project.

ok guys lets get started with some specs and numbers

eaton 3rd gen m90 supercharger (90 cubic inch displacement per revolution)

impreza gx 2.0 na
comp ratio 10:1
specific torque 92.28 nm/litre or 1.12 ft lb cu
max torque 184 nm/litre or 136 ft lb at 3600 rpm
specific output 61.7 bhp/litre (101 bhp/cu2)
max power output 124.7ps (123 bhp) (91.7 kw) at5600 rpm
(cid) cubic inch displacement (121)
crank pulley 5.25"

atmospheric pressure at sea level 14.7 psi

so we now have some vital stats to do some maths, you cant just fit this and hope for the best we get this wrong and the boxer engine will throw them pistons right out of the block like a heavy weight champ.

lets be sensible and only aim to achieve 5psi of boost
so intercooled or not ?
no intercooler no pressure drop
intercooled pressure drop of 1-2 psi depending on efficiency of intercooler

so Im going intercooled so we will do the calcs on 7psi with a pressure drop of 1-2 psi giving a total of 5ish psi

pressure ratio
14.7 +7/14.7 = 1.47 to achieve 7psi

volume efficiency (ve)

ve = hp x 75213/ap x cr x bhp x rpm
ap =14.7
cr = 10.1
bhp = 123 at 5600

123 x 75213 / 14.7 x 10.1 x 123 x 5600

9251199/102266136 = 90.4%

so we now know gx is 90% ve

air flow rate

can be calculated in two ways both should be same answer
cid x rpm/3456
121 x 5600/3456 = 218.8 cfm
or
cid x rpm x 0.5/1728
121 x 6250 x 0.5/1728 = 218.8 cfm

so
218cfm x .90 = 196.1 cfm = 100% ve

now calculate boost (pr)x air flow
air flow x pr
196 x 1.47 =288.12 cfm
218 x 1.47 = 320.46 cfm max ve 100%

so now we know how much air we need and the pressure ratio required
all we have to do now is finding a way of getting that out of the m90

7000 rpm on the m90 performance map generates around 320 cfm not taking in account any delta calcs for heat or wear in the rotors

pulley ratio

so at last we can transfer calcs to the gx crank pulley to achieve 7000rpm from supercharger to red line of 6250 rpm engine to establish a pulley
so
7000/6250= 1.12 pr
5.25/1.14 = 4.68"

so pulley size is 4.68"

so supercharger is going to run at 1/4 of max speed according to eaton map so extremely quite.

 

So weapon of choice is the m90 out of a 4.0 jaguar and she's a beast and big but chosen for displacement and quietness not it's size I don't want people to know what's under the bonnet.
What goes in must come out so before I started in the engine bay lets deal with the exhaust so I fitted a stainless borla unequal header now this will bring out the flat four sound, but there is a slight power drop on a standard engine by replacing the equal headers but since I am supercharging i'm not bothered, I fitted a decat then a modified straight through 2 1/2" wrx stainless middle section with a vortex afterburner back box so the exhaust is larger in diameter over standard and it sounds great.
This system will deal with any excessive back pressure issues, I then welded in a boss and fitted a wide band lambda into the decat so that I could monitor air fuel ratio readings once the project was complete and running.

 

Interesting project, I look forward to more updates. How will the supercharger mount, or is it a case of cutting up a manifold and welding it on top? I take it the drive will come from the front of the pulley instead of the aircon? Would a supercharger require any electrical control via the ECU or is it simply just a mechanical fitment?

It's always interested me what a Supercharger could do, and I'm sure proven results might inspire some twin charges, especially if it aids lag.

 

Hi, the supercharger has its own independent cradle, drive taken off the front pulleys but the air con system will remain.
Control of the supercharger will be through a remote tb and a bypass valve.

This is a completed project and a running car I just thought I would share the journey I had with you guys and give some inspiration, also what works and what doesn't. Ive got loads of pictures of the build which took months to complete but hopefully I can give to you in a couple of weeks.

 

Really interested to see more on this

 

Sounds good. I would love to see some pictures and some ideas of the results you've had .

 

Next thing after the exhaust I considered upgrading the brakes so I fitted dimpled and grooves discs all round with mintex pads and what a big difference it made so that was the stopping power
taken care of.
M.O.T time threw up the dreaded impreza rear subframe failure and leaking rear shocks so I took the opportunity to upgrade with the project in mind so on went wrx shockers with ibach lowering springs and a rear subframe from a wrx which has the rear anti roll bar with isn't fitted to the standard gx and again it transformed the handling the car is much more sure footed.
The only thing that I was concerned about was with the standard 15" alloys every time I went over a speed bump I left a piece on the Scooby floor behind so when I was offered a set of sti 17" alloy for £180 I jumped at the chance and it solved the clearance issues and just looked right now.
I bought a high level sti rear spoiler and aluminium turbo bonnet as I would need the scoop for the intercooler.
Now I felt the car looked the part and handling and stopping issues had been addressed it was time to lift the bonnet and get started !

Im going try and find out how to get pictures in the posts now

 

www.photobucket.com

Copy and past the [IMG] links once you have uploaded them, and get on with it, I have to see this

 

Thanks for the link

Space is an issue there's a lot to cram in the engine bay
First the abs pump was going to be a problem so I opened up the cut out in the inner wing and re bent the brake pipes to allow it to sit further back to open up the space where the supercharger is going to sit.







Next was going to be the power steering bottle and connecting hoses.
The bottle was removed from the inner wing and a new bracket fabricated and new high pressure hoses to the p/s pump, the hoses from the steering rack were cut and lengthened and re braised.





That's the space opened up now

Time to think about the intercooler the gx doesn't have enough room to fit the wrx/sti intercooler due to the distance between the tb and the bulkhead so trying to keep this build as much Subaru as possible after much hunting I opted for the intercooler from a forester turbo its not as deep so there's 25mm clearance from bulkhead.




so after a bit of colour coding its ready to be fitted when needed.
You will notice that I made a plate made where the b.o.v would be that's because supercharged systems simply don't work properly with a b.o.v which I will cover later on.

 

I spoken about supercharger choice the m90, flaking on the rotors can be an issue and wear in the coupler inside the gear casing for the drive.so I took mine apart to inspect it and indeed my rotors were peeling so I applied paint stripper on the rotors and removed all the paint which took days that stuff is hard as nails then polished them, last thing I needed was loose paint flake get sucked into to cylinders. I ordered a service kit which consists of front and rear bearings and a coupler. then on re-assemble replace the oil in the supercharger so its basically a brand new supercharger, I gave the outside casing a good clean and repainted it with vht silver paint.
I didn't take photos of every part of the build but I trial fitted the s/c into the new opened up space to fabricate the bracket using only the existing holes in the manifold and block making sure everything was aligned both horizontally and vertically for the drive off the front engine pulleys this took around 6 hours to do check and recheck, once I was happy I gave the brackets a paint in vht paint.







You will notice that the injector rail guard has also been remove for space and the upper s/c bracket uses their tapped holes to mount to and the lower into the block.

 

Somebody started to supercharge a Bugeye on here not so long ago if I remember correctly.

 

Yes, it was blownimpreza I did try to give him some helpful input with his build but he more or less said I didn't know what I was talking about which I thought was a bit rude, so I left him to it and it appears his build has died of death. so I thought why not show my build experience

 

Before I go onto the supercharger i'l talk about how this system works.
All oem supercharger systems are run like this THROTTLE BODY-BY PASS VALVE-SUPERCHARGER-INTAKE MANIFOLD this is the way its designed and this is the way i've done my build, the t/b controls the flow of air to the supercharger, some people have put the t/b after the s/c but this causes lack of control, driveability issues and buffering against a closed t/b under braking. cruising and de acceleration as the supercharger is still able to boost against a closed t/b. this is not a good set up and is inefficient.
The t/b controls the flow of air to the s/c
The bypass valve is open in cruise and high vacuum situations this allows excess air the engine does need to re circulate through the s/c whilst getting air from the t/b.
At w.o.t position (zero vacuum) the b.p.v closes and remains closed even through boost situations, only when the actuator sees any sort of vacuum does the valve start to re open to allow excess air to again recirculate improving spool up time so to speak when the valve re closes to create boost.

Supercharger with remote impreza t/b and integrated bypass valve photos


s/c with bracket fitted and b.p.v actuator and vacuum hose



Custom made s/c outlet



custom made air intake with Subaru t/b and iacv and tps still fitted
so it can still communicate with ecu and control air to the air assist valve



By pass valve still requires to be connected to s/c outlet pipe
I will try and get a photo of this connected





That's the beast ready to go into the car and as you can see its quite a size.

 

Very interesting reading.

Look forward to seeing the results!

 

Took a while to find the photo



Bypass valve connected up to s/c outlet

The choice of material for the inlet and outlet on the s/c was aluminium to make it light weight and strong.
Custom made gaskets none of that instant gasket rubbish
For the eagle eyed viewer you will notice tack welds appear to be holding it together however all sections have been fully welded and polished on the inside for cosmetic reasons and the full system was pressure tested to ensure integrity for boost and vacuum leaks prior to installation.
The Subaru t/b has been modified to make it as compact as possible as space is a big issue.

cruise control cam removed
Heater inlet and out cut off
Internal orifice added to s/c inlet pipe so that the iaav can still supply metered air to the s/c when idling or closed t/b conditions.

so into the car it goes






As you can see space is an issue and to be honest the fabrication and mock up stage to get to this point was the most time consuming part of the whole build ensuring enough clearance was given so that under driving conditions nothing rubbed or hit anything such as brake pipes, abs pump, inner wing, ect. But more importantly the bonnet could still close and not hit anything.

 

new bracket for the throttle cable to remote t/b

 

This is going to be interesting. Looking forward to the results.

The kid in me would want to opt for the smaller charger - just to get that supercharger whine!

 

hi,
There's no reason why if someone wanted to replicate this they could'nt keep the equal headers on to suppress the flat four sound and make it quieter but exhaust flow and excess back pressure is going to be an issue when you use forced induction on a standard n/a exhaust system, this why I decided to up the entire diameter of the exhaust purely for flow not noise I suppose choice of back boxes is up to each individual taste. I have run the car on equal headers, de cat, modified middle section and a slightly larger than standard back box and it was very quiet.

 

Lets talk about the drive system.

Drive will come from the outer belt coming from the crank over s/c then custom idler, power steering pulley then belt tensioned by the alternator.

The idler bracket locates behind the power steering pulley its thickness is 5mm. Due to the float in the power steering pulley this is the maximum thickness the bracket can be so that the power steering pulley does not contact the bracket.


idler bracket


power steering pulley removed and bracket secured using original tapped hole in power steering pump


The purpose of the idler is so the belt can be deflected under top radiator coolant hose as seen in picture.

Next is to refit power steering pulley fit s/c pulley and fit belt and tension



A coupler from a mercedes water pump was used on the s/c this was machined to suit the s/c shaft and pressed on to accommodate the Mercedes pulley.

 

Lets get the supercharger connected up to the intercooler.

prior to fitting this,the plate was removed from the original t/b butterfly this now makes it technically now part of the inlet manifold and control will now be from the remote t/b on s/c


From s/c outlet


To y pipe on intercooler
(worm drive clamp has since been replace with correct hose clamp)


Front view


intercooler and connecting pipe fitted


Looks a bit different to the standard engine bay now

 

Nice work Mike, great idea and knowledge !!

 

You will notice that there is no silicone hoses on the build except where hoses join this is due to the fact that silicone hoses are good for boost but useless when dealing with massive vacuum from the t/b in a closed system they just simply collapse under high vacuum conditions such as idle this is one of the reason I decided to use aluminium pipe work.

Now when it comes to existing vacuum hose we are going to have re pipe or can off certain part due to the manifold now being under boost and vacuum conditions the last thing we need is to boost things such as purge valve, air assist injector valve or through the crankcase/camshaft ventilation hoses.

The only part of this system that has vacuum relative to throttle position is the inlet side of the s/c


On the inlet side of the s/c you will see (3) 6mm vacuum connections these are relative to the remote t/b position and will never see boost conditions this is where we connect up the hose for the purge valve to dispose of fuel vapours, this also where we connect the hose for the actuator for the bypass valve, it is important that the bpv hose is not tee'd into as this will interfere with the signal preventing the valve from opening/closing properly and will prevent boost situations.

The crankcase/camshaft ventilation system is connected to the pipes on top of the intercooler then vented to atmosphere via a small filter



I do have an oil catch can but finding a home for it was difficult and to be honest I never see oil in the pipework so I wasn't to concerned about it but I may install this later and have the vacuum from the s/c inlet to draw the vapours through the catch can back into to combustion chambers.

 

From memory I think is should only leave the air assist injectors to deal with. It is funny how they only fitted them to the n/a bugeye then decided to drop them on the blobeye maybe because the system just isn't the greatest but none the less has to be addressed.

These systems require a steady flow of fresh filter air to the air assist valve/solenoid which is controlled by the ecu under certain condition so we need to cap off the air port on the iaav on the original t/b as this now technically now part of the inlet manifold as the plate has been remove and the iacv is unable to control metered air to the valve so we connect this to the remote t/b iacv on the s/c to supply this.
Now I have heard that some people cap off the air assist injectors and I did do an experiment with this and as I suspected I did notice performance was down slightly under certain throttle states which didn't surprise me due to them be designed to have a fresh air feed to aid in atomisation so I went back to my original set up as above because it simply works better.

communication

Next both multi plugs need to be disconnected on the original t/b for the tps and iacv and a wiring harness extension made so we can communicate signals from tps and iacv from the remote t/b on s/c to the ecu to keep it happy.




tps harness extension




iacv harness extension

So now the ecu can now control the iacv and read signals from tps and is happy.

All multi plugs are Subaru compatible so its just plug in, none of that cutting and slicing of wires because I think it would just look nasty

 

Thanks smiffywhu,

 

I've been out to the car today to get some more up to date photos that I missed during the build.


iacv blanked off on original t/b



front view of s/c


s/c belt alignment


Clearance between pulley and power steering bottle



Throttle cable and custom bracket to remote t/b on s/c



wiring harness extension fitted to remote t/b





Breather hose piping and filter using intercooler pipework


Sti alloys with dimpled and grooved discs.



Still loads of the build to go yet

 

Keep it coming buddy.

 

Lets go into the car for a while and install the gauge pod.

Now I made an error here and bought a gauge pod for the bugeye and when it arrived there was no way of fitting it into the hole after the clock display facia was removed the instructions recommended using screws right into the dashboard which I thought would of looked unsightly to say the least, or glueing it down which is no good if u need to get it back out, so I thought i'd have to think outside the box because I want to use the standard clip holes that the clock display facia used.

so I took a standard clock facia


I cut the top off leaving only the base plate and clips



base plate with standard clips

I then removed the clock and cut out an access hole so I would be able to install the gauges and run wires into the pod and then glued it to the gauge pod so it now is removable and fits the standard clips in the dash with no screws.


bottom view


top view with hole drilled for Subaru alarm led


now to make sure it fits and looks like it is belongs there







Good fit and sits nicely in the recess and removable

There was again no hacking into wires and no birds nest of wires behind the dash, I made a custom loom with multi plugs to aid in install and removal.


gauges fitted with multi plug connection




custom made loom


so at least I will be able to know what is going on now when it comes to turning the key.

 

We have now covered the air and communication lets have a look at the fuel system.

As the build cost starts to creep up I thought lets see where the limitations of the standard ecu in regards of fuelling before we look at piggy back/ stand alone ecu systems.

The injectors fitted to the 2.0gx are the same fitted to the 2.5 rs, so as far as injector fuel flow rate is concerned that gives me some scope to work with. I could of gone with the 2.0 n/a blobeye inlet manifold without the air assist injectors and that could give more options in regards to fitting bigger injectors but I am sure for the current set up i'm fine but at least its an option should it be needed if I decided to increase the boost.

First of all out came the standard fuel pump and I fitted a walbro 225 pump from a wrx. I wasn't quite sure what the standard pump delivery rate was but for £20 it was worth the upgrade.

Next I did the parallel fuel set up for the injectors to make sure all cylinders got equal fuelling in all condition

Then off came the standard fpr fitted to the end of the injector which is rated at 38psi max, now for some sort of adapter. I couldn't find one for the n/a bugeye but I found one on ebay for a classic impreza after a bit of modifying I got it to fit but the problem with it is the fuel outlet is not the correct orientation to sweep back at 90 degree to marry up with the standard piping and I wasn't happy with the way it sat on the flange and because it is part of the fuel system I decide to throw it in the bin. Now I then found that the bugeye turbo adapter with a 90 degree fuel outlet does fit, it has the same internal diameter fuel rail the only thing was the centers of the two securing bolts are slightly different and just require to be slotted and it is a perfect fit and matches the flange. When it comes to high pressure fuel you just cant take the risk especially when its right over an exhaust manifold.



Next I fitted an adjustable fuel pressure regulator and decided to set it at 38psi rising to 44psi so I had more fuel pressure at the injectors.

 

Now I could turn the key and see what all this work had accomplished.

Most importantly the engine not leaning out and engine knock were my primary concerns and just what would the ecu make of the changes and would it throw a cel light.

As I had increased the fuel pressure from the standard 38psi there was no surprises it ran richer than normal during idle, cruise conditions but what was going to happen in boost conditions ? I don't have a very complex data logging programme but it gives me all the information I need to make sure I don't wreck the engine.
The car starts fine every time and has a steady idle as usual, cruise conditions are as normal, mild acceleration is much more crisper and faster and when in boost it pulls like a train with no fuelling issues such as leaning out.
The torque increase is massively noticeable on full boost which is 4.5 psi so I now know that the intercooler and system has a parasitic loss of .5 psi due to the calcs I used for 5psi.
I've had no problems with the ecu throwing codes.

Performance is considerably up

standard 0-60 = 10 second
s/c 0-60 = 7.2 seconds running 4.5psi

so other than the fuel pressure increase the standard ecu appears to cope with 4.5 psi but I did do a test and I do know where the ecu draws the line and throws codes up but il cover that later.
I have done around 7000 miles with this set up and I have not had one problem with the car

 

I like this thread, very much in the spirit of how I work. Well researched and a good understanding of what you are doing and how to achieve it by the looks.
I have turbocharged many NA cars, some on standard compression, some lowered.
I think the next step you need to make is engine management, either piggy back or megasquirt.
I would opt for an AEM fic6 system, easy to use and fit and very reliable. This is what I am currently using on my car.
Injectors defiantly need swapped for bigger ones and an increase in boost to 10/12 psi will see the limit of your compression ratio.
The clutch may be fine at this boost due to the linear boost rise that a supercharger has, I bet your only a couple of psi at peak torque, 3-4k? Full boost at the redline?.
Interested what your AFR is at WOT just now?
We did a similar thing with a gti6 engined 205 turbo, 5 psi max boost, no Intercooler. 236bhp up from 165.
Keep up the good work