Engine and 'box upgrades on a WRX
 

So a bit of a question regarding what I want to do with my ’03 WRX. No, selling and buying an STI is not a choice I’m considering ;) If I was going to go down that route I would have already done it. Where I’m at with the car is it’s now a toy for driving on the road aboyut 1500 miles a year. It’s got AP brakes, Koni dampers and RB320 springs. It’s also running just under 300bhp courtesy of Andy Forrest through the stock 5spd box and is at just under 100k (mapped about 50k ago).

With it no longer needing to be practical or in any way economical I’m considering my options for where to go next with it. I’d like to go DCCD box and get some more power. This would probably entail doing the box first as the actual driveability of the engine is absolutely fine, I’d just like a bit more go).

So, drivetrain;

I’ve got multiple sets of STI 5x100 wheels and new hardrace pillowball rear arms etc so I don’t really want to go to 5x114, much though I like the look of the stock Enkei wheels. What’s going to be the best setup here? I think the only 5x100 DCCD 6spd was on the WR1 and some imports, correct? I have seen reference to using the stock R160 rear diff with the 6spd, as long as both use the stock 3.9 ratio. As I’m not really aiming for more than 400bhp (more likely 350ish), is this likely to encounter any issues? The big downside of this to me would be that the WRX diff is viscous (only works when both wheels have some traction) whereas the STI is plated and so is better on roads where it’s rougher. That said, the Konis have a huge amount of droop so this isn’t as much of a concern vs when I was running BC’s. Is this a feasible option to avoid having to replace rear arms/hubs etc to fit the 6spd setup? And if this is the case, is the prop the same fitting to both the R160 and R180 diffs?

With the DCCD I’m guessing it’s easiest to integrate a custom DCCD ECU rather than the stock one, or am I overcomplicating things?

Engine;

I guess there are a few options here. I won’t be selling the 5spd setup as I don’t ever intend to sell the car and so being able to put it back to standard in the future is a big appeal (regardless of what anyone tells me about the sensibility of this plan!). Similarly, rather than just throw a bigger turbo onto the WRX engine, I’d like to consider the route of a second engine to simply drop in. It’s already 2.5” fully decatted which I believe is absolutely fine up to the power levels I’m looking at (feel free to jump in if I’m wrong). The only change would be for putting a sport cat in to make MOT’s a bit less of a palaver.

So with 350ish bhp a target, and a nice torquey and well road driving option, what would the ideal spec be? I guess one option is to simply buy a used STI motor and mod that but as this is a toy, and a bit of a plaything, that seems like a bit of a cop out. Do I buy a blown WRX motor, closed deck the block from the likes of AS and then throw in something like a 2.5 crank with Mahle stroker pistons, some forged rods and then what else? Does the 2.1 even give me what I want? I mean I like a revving engine but if it only builds power up high then it’s not quite what I’m after. Put it this way the best driving Impreza I’ve driven was a classic running an AF TD04H at about 320bhp simply because its spool was so good low down (and the classic being about 200kg lighter than the blob of course). Turbo wise I have no idea what the best options are. TD05H? Ultimately I’m not after this being a budget no question power monster, more the best bang for buck to get me what I want while giving the best ‘area under the curve’ so to speak. It would be nice to have something that spools solidly at low RPM and goes consistently to redline rather than have something that’s all about low end that runs out of puff at 5k, or all high end that doesn’t get going until 4k. As an example, if I could get to say 350bhp with a nice torquey motor and it was going to cost me say £3k with some careful parts choices (doing most of the work myself), I would go for that all day long over £5k to get 370bhp. Then there’s the question of cams. Back in my days with other brands people would always talk about changing the cams yet in the Impreza there’s very little talk of this. Are the stock cams the way to go, or is it just that there’s not much choice out there for the EJ? Similarly, are the WRX and STI options the same?

Any pointers (or even threads that anyone has suggestions for me to go and read) then please feel free to give them, or alternatively tell me I’m daft. This is, at present, all hypothetical planning, it’ll be the end of the year before I start doing anything although if I come across stonking deals for some of the parts I’d need then I certainly wouldn’t be averse to start collecting them on the way.

I've got Quaife atb front and rear diffs on my newage wagon - my rear diff is r160 with 3.54 rear diff - as my six speed has the drop gear 1.1:1 to slow the prop speed.
I changed the pinion flange on the diff to match the flange on the STI prop.

I didn't bother with a dccd version box, I had the centre viscous diff uprated from 4kg to 12kgf.m/100rpm - you can certainly feel it in reverse manoeuvring.
There's no cable speedo drive connection on my box, so I used this:
https://mapdccd.com/vss.html

You could fit a TD05 or similar for 350bhp. But every time you increase turbo size you are also going to increase spool. Maybe not by a lot but it will noticeably drive different to how it does now. If you want the early torque of your current turbo but more power throughout the rev range, then you really need a 2.5 motor or convert to a twin scroll set up. A 2.1 conversion may give you the numbers but it's still probably too small to drive a large turbo at low revs.

Maybe forget about 350bhp and go for a billet vf35 instead?

[QUOTE=2pot;12098361]I've got Quaife atb front and rear diffs on my newage wagon - my rear diff is r160 with 3.54 rear diff - as my six speed has the drop gear 1:1:1 to slow the prop speed.
I changed the pinion flange on the diff to match the flange on the STI prop.

I didn't bother with a dccd version box, I had the centre viscous diff uprated from 4kg to 12kgf.m/100rpm - you can certainly feel it in reverse manoeuvring.
There's no cable speedo drive connection on my box, so I used this:
https://mapdccd.com/vss.

Prodriverules..work your magic.. if you have a spare 10 mins have a read

Random note about the drive-train: I believe the STI hubs (even the 5x100) have bigger bearings. This is a modification that I'd be interested in myself (though I haven't had any bearing issues in my 46k, it has to be said).

Regarding engines:

If you want early torque and not run out of puff at the top, you need as much volume as you can get, IMO. I'm using a td04HL hybrid on my 2.1, and there is lots of torque, but it does die a bit past 6000RPM. I don't want to, at least at this moment, fit a bigger turbo however, as I'm happy with the way the torque is delivered at the moment.
I've had a td05h-16g (the standard subaru one) on a 2.0 blob wrx engine, and I don't think I'd go for one of them again. It was cheap at the time (80 quid I think), and it made close to 320bhp on bad fuel (on purpose, because petrol stations..), and mapper said he thought it would make about 15-20hp more on 98RON. However, the spool point was worse than on an sc36, which I also had, and it made less power up top. I've also ran stupid stuff like td06-20g billets, but they become tedious to drive due to the late spool point... early spool is where it's at for me nowadays.

One thing to note - Linksfahrer started a thread on here recently about his experience with an SC36 on a built 2.0. I think he is still trying to eek out some more bottom end, but it does look from the graphs as if the 2.1 is much torquier than you'd expect by just scaling up the torque by 2.1/2.0. Something to keep in mind if you want to stay 2.0/2.1. This all becomes a moot point on a 2.5, where there is torque all the time, but at the cost of thinner cylinder walls :)

I would personally not spend money on cams or headwork under 350-370hp, unless you wanted more revs, in which case the STI heads have a higher rev limit IIRC. However, potential wiring challenges with this approach (or maybe fit some sti v5 or v6 heads, which should fit straight on IIRC).

linksfahrer's thread: https://www.scoobynet.com/general-te...-wrx-05-a.html

For a good fast road car then 2.5 is the way to go.

And although you say you are not considering swapping for an STI, you really should be, you'll save yourself a fortune and alot of hassel.

Yes, I think I'd be inclined to agree... The WRXs seem to be making good money too nowadays (long gone are the days of the 2500 pound blobs/bugs).

I'm personally so far down the road with my WRX, that I don't think I'd swap at the moment, but for a car with standard engine and box, it surely must make sense to swap?

Example: sell WRX for 7000, buy nice STI for 13-14k. Cost to swap is 6-7k, but it already has the sti box, the sti engine and sti heads, and it will always command a premium over a WRX if/when sold.

i've moded a hawk wrx and a hawk sti as a direct comparison and i can tell from experiance i'd sell the wrx and buy the sti.

Only way i'd start with a wrx is if im going the absolute full smash, bare shell strip, full roll cage, sequential box, race fuel system etc etc, short verison when all you want is the shell.

been there, done that, deleting the cage :D

edit: not that it ever got finished... best laid plans, and all that.

im yet to finish any of my project cars lmao

Interesting. What characteristics does the viscous diff upgrade give compared to the stock setup? The pinion flange setup swap seems easy enough from a quick google. The only thing is that a quick look at prices a bare DCCD box with no diff etc seems to be about the same cost as non-DCCD plus the viscous diff upgrade. What was your reasoning behind going down this route?

This is the thing that I'm trying to balance in my head. I don't do launches and for overtakes etc it'll be sitting above 3k anyway, but at the same time torque low down is definitely just nicer to drive. I get that 2.5 may be the way to go, in which case picking up an already blown 2.5 may be the best option and just forging it, but then the 2.5 is just flakey. In my head I'd rather build something bombproof if I go down that route, although I guess at 1500 miles a year that's perhaps not necessary?!

I've actually read that on and off since you started posting it. I'll need to go back through and read it again with a different frame of mind!

I do like the drive of the 2.5, even in stock WRX form, so I agree with your first point. I also agree with your second point, but it's not going to happen. I've had this car 8 years and I'm attached to it, which may sound nonsensical but I'm really not doing it for any sensible reasons. After all, if I was being sensible for a 1500 mile a year car I either wouldn't spend a penny on it other than oil changes, or I would sell it and not replace it! That and I've gone through the car to sort out all the underbody corrosion, all the bushes etc are new, the interior is mint, it's got the suspension setup on it that I really like etc. And sure, I could get into an STI for not much of a jump but a) that's a one hit cost (this is far from my only hobby), b) if I went 2.5 and it wasn't forged I'd be looking at an engine rebuild into the budget anyway, c) I'd have all the potential for bodywork issues still to be sorted and d) all the suspension etc would either be disgusting BC coilovers or knackered OEM stuff. I'd rather stick with my own car and tweak it, even if it's a gradual process. Like I said, I do take your point, but there's more to it than pure financial sense. The only time this car has seen anyone to be worked on was Andy Forrest for the original map and the MOT every year. All other stuff has been done by me and there's a lot of satisfaction in that. If I was being really ambitious I'd even consider building the short block too once the necessary machining had been done to it...

Right, replying to this one separately as it has lots of bits to question and link from.

Interesting. Are the R180 end splines different at the hub end too then? I hadn't clocked this before (I was going to pick up a used set of hubs to powder coat and replace my WRX ones with) although it seems obvious given the faffing with the handbrake shoes if you're putting Brembo rear callipers onto a WRX

What's the spec of your 2.1, I think there are a few ways to go about getting that aren't there? Or maybe I'm just mis-remembering reading old Jap mags twenty years ago and I'm thinking of 2.2/2.3? Do you have a power graph for your TD04H? Is it an AF one, or from someone else? I am still kicking myself a little for missing out on one that popped up on eBay for £300 or so a few months back but then I'm also still kicking myself ten years on for not buying my mate's MY2000 Classic with AF TD04 when I had the chance...

This was the dangerous starting point; I was considering picking up a cheap stock TD05 and either rebuilding it or hybridising it but using it on my current engine. But then I thought that it would be better to start with something more homogenous and spec'd more closely to what I'd want with the block etc...and the slippery slope started. I mean I could pick up a TD05 for a few hundred, another few hundred to get the map tweaked by AF and some pink injectors and it would be a power boost for probably £5-600. But I'd quite like to build another engine, and do things right...and well, things then get expensive. Last engine I rebuilt was the Ecotec V6 in my old ST200.

So in your experience of the TD04-13G, TD04H-19(?), TD05-16G, SC36, how would you describe each of them? I know you said you're currently on a TD04H but did you have that on a 2.0 at any point or has it always been on a 2.1? I guess if I wasn't actually giving up low down torque going to something bigger then it would still be reasonable as it wouldn't be worse than what I've got, and then just get better up top, but I'd still rather have more low down if I can get it! I do think what I'm looking for isn't a million miles away from your ethos.

I'll give that thread a read too. I think my concern with the 2.5 build is that they do seem life limited, even when forged. I could possibly be convinced that they are the way to go but at present I'm not sold on the bigger picture.

Do you mean the STI head are just able to take higher revs, rather than they have gas flow properties to help with higher revs? I think the complexity of fitting AVCS heads is probably out of the question, so V5/6 would probably be the answer if I went down this route but it's probably not worth it if sticking to c350bhp. Again, a high revving engine is nice occasionally but I just don't find myself driving like that any more so it doesn't seem a sensible direction to take.

Thanks for all your input, it's much appreciated.

I think some more research is needed (someone who has done the 6mt swap probably knows - i'm on an uprated 5 speed box myself (PPG)), but this thread https://forums.nasioc.com/forums/sho....php?t=1412230 seems to suggest you can re-use your WRX front shafts when doing the STI swap, which I think should mean that the hub spines are the same?

Mine is a 2.1 made up with: 92.5mm wiseco pistons, +1mm longer rods (k1) + STI crank. IIRC the compression ratio is about 8.5. Swept volume = 2.12 litres
Note of caution regarding the 2.1 engines. Some companies call their 2.1 (i.e. the recipe above) a "2.2", because if you bore it out to 93mm you get to 2146 cc, which apparently rounds to 2200 :confused: . This then confuses matters, because the EJ22 from Subaru is an actual 2.2 engine. The EJ22T (the turbo version, with closed deck block) can be made into a 2.3ish litre engine with a 79mm crank, and this is probably the beast you've been reading about in old mags. Many big power imprezas use a 2.3, but they are massively costly nowadays (even a block seems to be about 3k on its own, before any machining etc etc).

I built my td04 hybrid myself - see the project thread I linked to above for details, but basically it's a TD04HL (bigger exhaust wheel than the TD04H), with a 19t billet wheel.

If you have built an ecotec, I think you can probably build a subaru engine too. There are lots of videos on youtube etc to help you on your way, but careful research is needed piston/bore clearances etc. Nasioc seems better for that kind of information - lots of people building their own motors over there over the years.

Unfortunately I never ran the hybrid turbo on the 2.0, so it may well be that the HL turbine on a 2.0 would be quite sad in terms of spool - not sure. Should still make the same peak power, but that doesn't mean that it'd drive nicely. I've written up my own thoughts on the turbos I've ran here https://www.scoobynet.com/1060270-20...l#post12098181 . Word of warning - I think people get attached to their own turbos over time, because they learn how they drive etc, so we all end up biased in favour of our own setup :-) What I like may be different to what someone else likes.


This is my fear with them too, and it's why i'm staying with a 2.0/2.1 engine. It may be an unrational fear, but I don't care :)

The STI heads should both flow better and rev higher. The cams are also more aggressive on the STI5-6 heads. HOWEVER, they're now 22-23 years old, so maybe they need a refresh before running, which could make the cost prohibitive.
Several years back I had a set of STI6 heads + valvetrain etc, but I sold them before I fitted them. Back then I wanted to build a big power engine, and I would have wanted the AVCS heads, but _even if i had them now_, i wouldn't necessarily take my WRX heads off just to fit them. For me personally, I also don't think that at 350bhp there will be enough of a difference to justify the cost of swapping the heads over (if the HG was gone anyway etc etc, then yes, but if the WRX heads were working properly I wouldn't spend the money).


I've been doing things more or less on a shoe-string (my mrs will disagree), as I don't want to spend every last penny on the car, and also as a hobby-proposal it's more fun for me to learn about the things and get my hands dirty, rather than pay someone else, so my choice of components probably reflect that... I've bought things that looked interesting as and when I came across them. In retrospect I should have bought an STI, but I couldn't afford the insurance at 23, when I got the WRX :)

Why not just get an appropriate VF series turbo and all the cheap bolt-ons/additions that help spool as much as possible that aren't already on the car (up-pipe, sports cat/decat, ported manifold, turbo blanket etc), get it mapped for 350hp and enjoy?

If you gearbox breaks, then replace/upgrade. If you shorten the life of the gearbox to only 15k more then great, you've got another 15 years of use. Likewise, if the engine has an issue after a while, you can rebuild then/replace with an STI one.

I'm not sure I understand the logic of spending a fortune to achieve what people are achieving with a turbo and bolt-ons with a WRX engine, especially when you don't need to rely on the car as a daily. Most of the bolt-ons are useful if you then did a rebuild, and likewise possibly even the turbo. May as well hit that power level cheaply and see how you like it and just hope it stays reliable.

I'm going to read both your thread properly, Linksfahrer's and a couple of others I've come across and I'll come back to you with some more questions ;) Thanks for all your pointers and input :)

The man makes a very good point, and logic I guess doesn't really come into it which is why I'm considering this! I think it's the fun in having a project and the sense of achievement in building an engine that really appeals (the Ecotec was a bare bones rebuild with new rings and a few other things rather than a ground up performance build). At present I have no family to provide for and therefore have some spare cash, so my view is that I would like to get the car to where I'd choose it to be before budgetary constraints really kick in in the future. But there is something to be said of the bolting stuff on approach. That said I really don't like breaking stuff, particularly when the mode of failure is so uncontrolled.

I'm already down the route of having another set of manifolds there which I'm going to get cerakoted as the OEM ones on there just now have completely rotten heat shields that need replaced, and I hate exhaust wrap on a road car as it simply absorbs salt and everything else and causes increased corrosion to the pipes. So they'll be ported, as will the spare STI up pipe I have too. But I'll probably hold off putting them on until I've decided what direction to take with things. There's already more breathing available than when it was first mapped (originally it was; full defat, WRX I/C, stock centre silencer and STI back box [290bhp/305lb/ft], now it's got an STI cooler, straight centre and Prodrive box at the back) so I'd rather not risk any potential lean running issues by pushing my luck.

With regards to the gearbox the main reason isn't so much the power capacity of it but the DCCD. I much prefer the drive of the cars with the power delivery, albeit in my car I have got it set to rotate very effectively in corners. However, the downside of that is that it's like this all the time rather than just under power and you still don't have quite as much line adjustment with the throttle. I think I'd probably choose a DCCD box over more power I I was being brutally honest and had to choose.

Reading your opening post is like a DejaVu , you already have my full understanding while aiming at 350bhp from a 2.0 WRX.
Before spending your hard earned cash you have to make some hard choices. the danger here is going down the blind alleys.
Id like to give you some to do's / not to do's.

Clearly you have recognised already that , the drivability (availability low down torque) is a key fun factor.
Henrik has most of the T shirts on this and I have been following a lot of his long experiences,
its a good idea to read his threads for more detail.

Some assumptions already made.

You have a WRX and you are going to keep it. because it costs considerably less than a 6 speed Sti
You think you will be happy with 350bhp (and for arguments sake we will say 350lb/ft for now)
You are going to spend some money on supporting mods / suspension / brakes etc.

Well at this point don't bother with the 6 speed box. you don't need it for this power level , however anything more you will.
equally "if" you build a nice torquey engine then a 5 speed is comfortable to use. Driving a 6 spd is quicker but you are far more aware of stirring it about.
but they are very useful for keeping a larger turbo on the boil when you press on across country.
Looking at you initial post though I don't think a 6spd is 100% for you.

Ok so if we discount buying a Sti / fitting the 6 speed with 5x100's then we are left with the 5spd box gearing.
That's both a blessing and a curse.

To get the 03 WRX moving faster you need loads of torque.

Retroplug has an excellent common sense point at this stage:

"Why not just get an appropriate VF series turbo and all the cheap bolt-ons/additions that help spool as much as possible that aren't already on the car (up-pipe, sports cat/decat, ported manifold, turbo blanket etc), get it mapped for 350hp and enjoy?"
This approach will likely bring you to 340@ 5600 / 340 lb/ft @4000 and indeed that's a very good road car.

Getting above this requires a bit more thought on either raised capacity , or an even bigger turbo. I will assume for now that you will stay with the ECUtek standard ECU and a 1.5 bar max

I went for a tiny bit more capacity 2038cc ( 93mm bore/ 75mm std stroke ) and the SC36 turbo ( 7cm housing) similar to a TD05 16G And I'm now at 343bhp @6500 / 330lb/ft@4500 hmm not exactly a success is it.
low down torque is so la la, but I do have a wide power band >300 bhp from 4950-7650 rpm and on the 5 spd box that means you can concentrate on where you are steering and braking rather than stirring the box.
So in the speeds 60-100mph its pretty good blast.

However if you are moving through small lanes , the low torque is disappointing for the important 20-45 mph in 2nd gear .
This will also the case for larger Turbo's , you will always sacrifice the low down performance for top end. Ending up with 380-390 bhp on a TD05-20G won't help that and you will run a constant danger of blowing the gearbox.

Gettings round that problem on a 2.0 requires that the turbo is brought up to speed much faster, and that requires a lot more gas speed.

You can port headers / fit tubular RCM's Either or is essential anyway to get the engine breathing / 3 inch downpipe/ headers a Harvey uppipe for 350bhp ( not a big one for +400hp) get this from AS performance

You can raise the CR from 8.0 : 1 And remap safely to 8.5 :1 on Ron98 Maybe 8.7:1 on Ron 99 with some subtle combustion chamber work and yet still run 1.5 bar ( Twin map this to 1.0/1.2 bar for emergencies for 95 RON only )
this will give you more of a Normally aspirated throttle response and will bring you 2.8 % more power if you go up by the 0.7:1 The low down torque ( sub 3000rpm torque levels will be improved by 200-300rpm )

You can raise capacity.

And here the performance of Henriks / Malc B's old car with 1mm longer con rods and 92.5mm bore appears to be a very good solution. ( 350.1 bhp ) please take a look at his Dyno graph it is impressive .

You can go to a 2.5 crank / new rods and this will give you even more gas speed. But I am not a believer in this build in a EJ205 engine without at least going forged pistons and rods,
possibly a Semi closed deck EJ207 if you want to think ahead for a later time and 400 bhp

With a 2.5 crank you will find easily your target bhp /lb/ft and more likely end up at @ 375 / 390 lb/ft with a TD05 18g / SC38 but that's out of 5spd box comfort zone. so you would have to run lower boost and drop CR for expedience
Maybe a Vf35 / SC36 and a lower boost level giving a very nice flat wide curve 360/360 lbft would be best v gearbox top limits. However the longer stroke does mean more bottom end stress as the piston speeds are increased ,
so doing this really needs an 11mm uprated oil pump / the forged rods / pistons / ACL bearings , Its a 3k engine build. That won't like to rev as high as the 2.1 does but will satisfy you yearn for torque.

Camshafts are generally not the way forward to improve low down torque , except if you were to reprofile your existing ones for a tiny bit more lift. but staying at the same duration ,
here you need to watch out for piton valve clearance, everything else cam shaft wise I have seen on the internet will end up making more power higher up the rev range at cost of bottom end performance.

Sti heads on AVCS would give you more low down torque as the cam timing is optimised according to rpm for the best of both worlds , and this could be a solution for you with a Sti loom adaption to hook them up.
I'm informed they are worth 20-30bhp and presumably also same for torque depending on the level of tune. This is a bolt on , and could be an interesting conversion.


Given the choice again . Id go with a new forged build from the start.

Henriks 2.1 92.5mm ( still allows for a 0.5mm rebore ) +1mm stroke
Aim for CR 8.6 : 1 by either gasket / or head / deck skim
Adding the SC36 Turbo ( But do use the SCR36 this is the Blouch ball raced XTR version ) to get even better spool.

The 2.0 alternative is a TD04 Hybrid maybe not quiet reaching 350bhp ( ask Henrik about his 19t or find a Andy Forest S/H one ) this will be a very quick car due to the spool.

Fit an exceedy pink clutch.
Tumbler delete
TMIC STi 8
565cc Hawk06 Blues
New FP 255lph
Cosworth Air filter
NGK 7's
3 inch Downpipe 200 Cell / adapted to 2.5 Prodrive Mid/back
or Decat 3 inch Milltek mid back box which gains some torque/bhp
exactly how much I'm not sure, others will know better, I suspect at this level 10-12bhp / ft/lbs

IF you stay on the 2.0 engine .

I'd do above mods the Sti AVCS upgrade, and the TD04 Hybrid which in the short term is the cheapest way to get very close to target.
Spending any other budget on good coil springs / set of larger front disk brakes 335mm will fit in the 17inch rims
when the engine goes pop you do the forged 2.1 and you can still think about a 6spd on that with a MD321H for 400 bhp


All the best.

Only thing i would add to links post is possibly a 2.5 conversion. the low down grunt they have makes for mega fun road car and will make even the 2.1 look like a lag monster by comparison

On the DCCD point, I‘ve owned two classics with viscous diffs, as well as a blob and a hawk both with DCCD-A. Of these I most enjoyed the handling on the blob with its 66% rear bias. It felt more neutral than the classics or the hawk and felt like it gave you more options. You could get it to rotate under power which was nigh on impossible with the other cars, but was still pretty foolproof as the front would still pull you safely out of the corners, often in a nice four-wheel drift.

By comparison the other cars required more trail braking and weight transfer to rotate and overall felt more inert and “nose-led” compared to the blob which felt more tail-led and more fun as a result.

I’ve recently acquired a classic with manual DCCD which I’m expecting to be relatively tail-happy but I’ve not had the chance to put it through its paces.

I too feel that the WRX 02 -05 is more nose led , I rarely get the back end to flick , which is another reason to look for the low down torque from a Hybrid TD04 , I deal with this by putting the car slightly off line and brake heavy wirh 335mm disks front and good biting pads Carbon Lorraine RC5 / RC5+ rears then as it starts to drift , John Felsted can explain / do this better and is a master on DCCD Setup (search his threads), but if you don't have this kit then you have to do it by changing your driving style and optimise what you can on the car. To get the power down balance steering with throttle , this I manage fine on wet / rough surfaces on winter tyres but the grip level is too high in the dry on Nankang Nsr2 , that's why I need more lb/ft in the 2000-3500 rpm band. So its think about mild boost 2.5 or 2.14 engines time if staying on 5spd or keep to the Hybrid TD04 on the 2.0

Lots of interesting reading on the subject of DCCD's, it's not a subject I've really investigated the tech of too much beyond the superficial knowledge I have, and my experience of driving them on the road (WR1/2004 WT/Hawk Spec D). They all exhibited a more throttle adjustable experience for me, with less understeer. Then again I've managed to dial understeer out of my chassis (although I'm still yet to install those bump stops from you last year!). Referring back to your 12kg centre diff in the WRX 5spd, how does the drivability on the road compare to the DCCD-A and WRX stock box with VC (my two references!). I generally dislike an understeering car in favour of something that is more rear led, albeit twitchy Mk2 MR2 heroics are a bit too far in that direction for my tastes!

I didn't notice much difference between the DCCD setups in the models noted above, but that probably has more to do with the time gaps between each drive than there not being a difference. Similarly, the difference between these models and my WRX were the noticeable traits.

My current suspension setup is Koni inserts in original struts (much better than the BC's I had before for the roads I'm on) which are matched with S05 front springs, RB320 rear springs with 10mm spacers to achieve positive rake, an ALK (which @2pot has suggested to swap for a Gp.N item, and I will in time, alongside installing his suggested bump stops), and a 22mm rear ARB. I've also got 330mm AP CP5200's on the car. It's definitely not a stock handling WRX any more which relied on more aggressive steering alongside left foot braking to really get it turned in. My choice would be to have a slightly more neutral handling chassis and a more throttle driven response from the back end, akin to what I've found in DCCD cars. @2pot, would the upgraded centre diff in the WRX box give this? My concern with going down this route is that I would commit to spending reasonable money for the box to not be any stronger while also losing out on the extra ratio etc. I'm not averse to the idea, far from it, but I definitely like to go into things with my eyes open.

Even in stock form I've found the 2.5 WRX to be a lot more provokable than my mapped 2.0 and that is purely down to the improved torque low down so all the suggestions of 2.5's and/or building for low down torque are pretty much along the lines I'm already thinking. I'm not one for leaving black lines on tarmac no matter what I'm driving but real throttle adjustability is nice to have.

I'll reply to your engine comments shortly (massive thanks for such a comprehensive response!)...

Adding power/torque with more cubic capacity will of course allow you provoke oversteer if you have a diff balanced to rear ( 2wd rally style) Having more power at lower revs in tight turns in the WRX will see you (on loose/ wet surfaces ) break loose the rear , and on all four in esses, but still with a tendence to understeer . Without the DCCD that will stay the same, so getting the tail out still requires you to get the tail into rotation by steering/ braking, Personaly I have yet to master left foot braking in the WRX , alrhough I have used it in 2wd Audis and to good effect in a Volvo V70 D5 , to steady the ship as you power thru the Apex , realeasing pressure as you open the angle of steer, I have never felt that it would help me go faster in the Subaru to pull out of the Apex. On the rare occasions that I use the left foot its a very short jab, when Im too fast in a longer curve , where I need to steady the car. Anyhow you need to have empty roads for all that. For just poking the rear end out the WRX has the wrong spec

Mine started life as a standard WRX PPP, watch out :D

On the DCCD front I can recommend a mapdccd controller, really good bit of kit :)

First set the chassis to the balance you prefer - rake, springs, dampers, bars, alignment, bump stops.

dccd-a still understeers.

The dccd - 4kg open to 20kg locked - 50:50 when locked

dccd open = rear biased torque split, sends drive to the end with least grip.

My box is a non-dccd six speed conversion, 50:50 torque split - 12kg centre is a 4kg with a change of fluid viscosity. Predictable handling in the dry and wet.

I see your logic on the suspension, although I'd still say my experience is that these cars have felt significantly more throttle adjustable into an oversteer or at least tail led attitude. Is this simply the way I've been driving them? Am I mis-describing or mis-understanding what's driving this sensation? I'm not being facetious, I'm seriously asking to be told I'm wrong so I can work out what the correct course of action is! On those cars with a 60/40ish torque split, what is it that causes these to still understeer?

With regard your diff on the 6spd, the more I think about it the more it does seem to make sense. However, this does limit you to one split rather than adjustable than DCCD, and also fixes you at the 50/50 torque bias as it seems the non-DCCD boxes never deviated from this, unlike DCCD. So this leads to the next question(s), did you do this yourself or get someone else to do it? And if you did do it yourself, where did you get the higher rated fluid from, or if you got someone else to do it, who did you get to do this work? From some research it seems that it's also possible to do it to the 5spd which might be an interesting and cheaper test to see if I like what it does before committing to doing the same on a 6spd box, or indeed a DCCD unit.

@Henrik and @Linksfahrer I'm really not being ignorant and ignoring your posts, I'm combining them with some other reading (four stroke performance tuning and forced induction performance tuning) and some other threads, including those you linked to and putting some thoughts down on paper. I will be back to you on the really informative posts, but ultimately it's a complex set of different parameters, options and thoughts and it's taking me some time to get my head around the options :)

Bara Motorsports for viscous mods.
01527 880011

Uprate the R160 viscous rear diff as well?

The viscous centre is reactive, responding most slowly at the 4kgf-m/100rpm open setting - increasing from 4kg increases its speed of response. Too stiff (20kg) and you'll get understeer/binding during tight manoeuvrers, in dry/grippy conditions, at road speeds.

Maybe this?:
If the front loses traction, in the open position, torque will be transferred to the front = more understeer.
If the rear loses traction, in the open position, torque will be transferred to the rear = more oversteer.
35/65, then later, 41/59, maybe they changed it for a reason? Maybe it was catching out/unsettling the average driver, as the stock dccd-a couldn't compensate quickly enough? Particularly as braking opens the dccd?

We need @johnfelstead

Well there's definitely a lot of info in this thread, and what's been said has done a great job of sowing some additional seeds of thought I hadn't considered before. I guess there are two options; go safe and with bolt ons to the existing setup (and thus keep within the box limits), or go big and accept the increased budget. Given where I am in asking advice on all this I think there's an element where I need to see what it is I actually want in the real world.

So where's my head at? Well I'm sitting here thinking that I'd quite like to see what I can do with the easy stuff, which then lets me try stuff out before going for the big plan.

I like the idea of the uprated viscous diff option @2Pot, thanks for the link to Bara too, which would potentially work in the stock 5spd box and R160 rear diff that I already have and allow me to see how it works on the road, I mean a full spare diff and box setup is hardly bank breaking. It's interesting that this really doesn’t get talked about on here very much but on other forums it does. I guess this is down to Nasioc etc seeing a lot more autocrossers on there compared to here. Anyway, more investigation is going to be done on this.

With regard the power upgrades, I'm increasingly thinking that long term the 2.5 is probably the way to go for me despite their potential fragility. That then gives the option of some bigger turbos while still keeping low down tractability. A friend of mine had a 2.5 with the SC36 turbo (I think this was before the SCR36 was made available) and that thing was an absolute weapon. Going for the 400/400 mark would probably be the aim for a build like this I think but again, I'd like to do a lot of the work myself if I can.

However, this leads back to the interim and there I think the bolt on option might be fun, and cheap while I spend more time thinking about the long term aspirations. I've got a spare manifold there which I got to cerakote and wrap as the one on the car has rotten heat shields and a knackered lambda thread. This would be an ideal option to port, and while it's off it would seem sensible to go for the Harvey up pipe (or would this be negligible for c350bhp? vs a ported stock STI pipe?). Then I guess it's a question of going for either a TD04 hybrid or a VF of some description. While the former will obviously run out of puff earlier, it may give more in the range where I really drive it at 3500-5500rpm. Again, more driving to be done and thinking. So are there any specific VF turbos to look out for that are better than the others in terms of gains? I've been looking at this piece and it seems that they're much of a muchness in terms of what they can offer. I guess my thinking is that the ones which bolt straight in are going to be the most cost effective without needing any oil supply mods etc, albeit they're not that much effort to do in the grand scheme of it. And similarly I'm guessing not all TD04H's are built equal so there's more education to happen there.

With regard to the air intake side of things, is doing the TGV delete going to net me much? I mean it's all well and good reading things saying 'this'll give you +10bhp' and then adding all those bits up but clearly these things don't compound, but with ported exhaust manifolds and a different turbo, is this likely to give any performance gains? Power I'm not too bothered about, but if it gives better spool and torque in the mid range then I'm game for it. But then I've also seen it's an utter PITA of a job so perhaps best left to a new build from the ground up? Additionally, is there any other intake porting work that is worthwhile for these power levels?

Finally, fuelling. A new fuel pump is a given as the one on there is 7 years old now and 290lph so given the price of a new one I'm going to call that a consumable. Which leaves injectors; what's the best option here for a 340-350bhp aim (well aware that I could come in well under all this despite the best efforts)?

I guess I may come across as a bit of a walter mitty on all this but it's all a question of timing. I do want to build an engine, and I do want to get a 6spd into the car but it would also seem that I'm not quite decided on what build the engine would need to be to suit what I want. I think the answer is probably to get some seat time in a few different cars to see where they are for me; do I want a low down torquey build or actually, given my daily is a 335xD wagon, would something that's a bit higher revving actually make the car feel more special? No idea on that one, the only way to find out is to try. Maybe going for the slightly slower spooling VF turbo may allow me to test that theory? I mean if I didn't like it then it would simply be a case of taking that off and putting a TD04H on for the interim and then realising that the 2.5 would be the build to go for long term. Then again, at 1000 miles a year I suspect I could probably live with it.

When mine was around 340bhp it was a great road car, and the WRX can be very reliable (even the box) if they have been well maintained which yours sounds as it has.

I had been in a few cars with an SC36 which is essentially a TD05-18G, a quick spooling turbo for 360ish and they all went like a stabbed rat! If I were you I’d get one of them, a set of STI pink injectors would flow enough for that power level, if you can do the tgv delete then do, other supporting mods like an Fmic would be worthwhile too.

Then you can go have fun, and if you have any engine or gearbox issues then you can go down the building route at that point, you never know you may decide that the power is fine for what you want !