The AUSubaru.com.au Forums

Hi all if anyone has links to the above car let me know please , I tried searching but didn't find much .

Cheers A .

I found the member in

"members list"

if that helps. QikRX is what I found under Q

Hey guys, long time no chat! What info you after?

Hi and thank you for helping us out .

I am doing a spider Vortex EA82T engine also for an early RX Turbo and have made a few internal mods to get a bit more urge out of the old dinosaur .
I am lucky to have the Rally people from Road and Rally and SWR to do things like port the heads and any fab work . Stu Wilkins rallied an RX in the late 80's so he has a good idea of their strengths and weaknesses .

Basically someone had rebuilt and used this engine in their AWD Vortex , could tell that it was not factory original and had the engine cases decked to get the piston crowns closer to the case decks faces . I had the chambers cc'd and when the static CR didn't come out at the factory 7.7:1 we looked closer . I wasn't then sure how far I wanted to develop this engine and at the time settled on 8.6:1 CR , wished I'd have asked for 9:1 now but thats life . Also the std pistons were in really good condition and the bore clearance perfect so it was honed and fitted with genuine rings for its model . They were ceramic coated on the crowns to keep their temps down as well .
Its still on the engine stand ATM minus manifolds and I can't decide if I should put better cams in it or not . The factory ones are 14-56 56-14 while the NA ones are 16-60 60-16 .
Where you could help me is to let us know how grumpy these engines get with extended cam timing , you mentioned originally that your cams were modified but no specifics .
My guess is that being in QLD you may have used Tighe profiles as they list a couple of grinds for EA82's , I think a 20-60 60-20 like late hi comp (9.5:1) NA Spider Vortexes used in the US , and a rally grind for the go fast on the dirt brigade .
I'm torn over using the L EFI 16-60 profiles as they should not be too dificult to find .

Another thing I noticed in your posted specs , was it 2.25 header pipes into a 2.5" pipe up to the turbo ? This sounds on the surface to be real big but I guess you must have had your reasons .

To turbochargers , I should have an IHI VF8 and a VF10 here next week and I noticed that you had used both of them .
VF8's look pretty big in the wheels/trims/housings for an EA82 and I'm curious to know if it was pretty lazy until the engine got some revs aboard .
I think I'll lean towards the VF10 with its smaller trim turbine , smaller A/R turbine housing and smaller than VF8 trim compresor - in the same sized compressor housing .

Gearbox wise my RX has a dual range converted Vortex AWD transmission but
with the closer ratio 1st gen DR Lib main ratios . Its a little shorter on the highway than the std Turbo L ratios but being closer will stay on the boil more easily when leaning on it .
My clutch is an uprated one but still with organic facings and a sprung center , was you button clutch plate solid or sprung ?

Overall you car certainly sounded like it had get up and gone but was it a gruff grumpy thing when just cruising around .

I applaud you for having the testucular fortitude to persevere with an engine that many consider to be a waste of space , it would have cost you a fortune even without the breakages .

I'm sure you embarrassed many a capable car and you can't put a price on that . A wolf in sheeps cloathing no doubt .

All help much appreciated , cheers Adrian .

Hey Adrian,

Has been a while since I sold the beast but am happy to answer your questions to the best of my ability. The main issue with answering your question is that I don't know the exact spec of the cams. The standard cams were modified when it was rallied back in it's day as aftermarket item weren't available at the time. I know they had about 4mm of material added to the lobe at it highest point but as to the exact lift/duration, I'm not sure. I wouldn't even know who did the job originally & they came with a spare "rally" engine when I bought the car about 8 years ago. The said engine had the notorious cams, forged pistons (which I later found were made by Specialised Piston Services in Melbourne), 440cc injectors as when I sent down an original undamaged forgie to get 4 new replacements, were made by them back in 1986, what are the odds! I didn't take any photos of the cams even to get an idea of the lobe spec, as the engine was built for me with the CR of 8.1:1. The new pistons weighed a fair bit more the the originals (which was requested by the builder) which contributed to a healthy 420nm's or torque. Maybe ORX-18 can help you out with the cams as he owns the car now & has rebuilt the engine again (bastard took my plates as his nick) Sorry Allan! The setup was a bit of a hand grenade with that sort of fuel/boost but the cams were definately one of the main contributors to the power the car made. Probably about 30-40bhp when it was at the pointy end of the build with all the good bits.

In regards to the turbo, there is a big difference between the VF-8 & the VF-10, albeight the relatively small difference in efficiency. The VF-8 RHB-52 Legacy turbo was noticably laggier, coming on full boost (19psi) at around 4200 in 2nd, 4000rpm in 3rd gear & 3750 in 4th in comparison to the VF-10 @ 3500, 3300 & 3000rpm respectively. Peak power with all the work & the VF-8 was just shy of 280hp at the wheels @ 6300rpm & about 250bhp with the VF-10.

I started off with stockie then upgraded to the VF-8 RHB-52 Legacy RS turbo. The VF-8 then spat the oil seal after about 3 years of hard driving & was blowing smoke like a 1975 diesel landcruiser so bought a VF-24 & found the impellor wheel was chipped on install & was advised from the seller (Option 1 Garage in Brissie) that I could get it rebuilt no charge. Thought that I was sweet & was getting a refurbished turbo until I found out that noone in the country could get parts for it. So then I swapped it for the VF-10 with some money back & the RX was back on the road. Little to say, I was disappointed with the VF-10. Although it was much more responsive & was came on boost under 3000rpm, it was lacking the crazy torquesteering/wheelspinning characteristics of the VF-8.

I had the car with the VF-8 on a standard engine while I was getting the tough one built (with all the bolt ons, ECU, boost, exhaust, intercooler etc) & was told not to wind the boost up to 19psi on the boost controller as it was a stock engine & wouldn't last as long as my built engines. Well having more money than sense at the time I went out highway racing on the M1 that night on full boost. That stock engine lasted about 10hours then cracked the RHS (turbo side) head down the seat of the exhaust valve.

So all in all, the VF-8 was the pick of the bunch (minus the VF-24 which I never got fitted). The exhaust manifold was as big as we could fit (relatively no clearance between the head & the crossmember) which contributed to the top end power and the laginess under 4000rpm. This thing took about 12 hours of fabrication to make as we had to take the engine in/out about 8 time to get it cut/tack & welded. Off boost, the engine would make about 50bhp but then made power from 4000rpm up until 8200rpm until the turbo ran out of puff.

The bottleneck of the setup was the intake manifold which you already have sorted by running the spider manifold. That was next on the list for me (was going to make a customer plenum with tune length runners & 100mm XF throttle body) but by that time I'd already thrown too much money at the thing & was already munching gearboxes for breakfast. Making it more reliable was next on the agenda, not making more power. That's when Allan called me about buying it, at which my bank balance was happy with. He bought it for less than the cost of the engine, but dems the brakes with modding cars. Especially old ones.

Sorry for my rant. I miss the grunt bucket.

LOL no problems with longish posts , I'm famous for them .

Header wise I was thinking more along the lines of something like 2" (or smaller) tube because this is primarily basic transport so has to be a reasonable drive at more pedestrian engine revs .
I like the way WRX crossover pipes have the stainless bellows so one of those with something to join it to the exhaust ports and up pipe may be a goer .
Did your header attempt to equalise the lengths of the engine pipes before the collector or was it mainly aimed at high flow and ease of packaging ?
There is a mob in the US called TWE that aims the turbo side heads exhaust pipe forward and then through a tight 180 and up throught the hole in the "crossmember" to merge with its opposite number just before the turbos mount flange .

When my engine was rebuilt and AJ ported the heads he comented that it wouldn't be impossible to fit a plate up through the header flange to meet the port divider and end up with a dual port type system , probably not woth the effort on a roadie though .
I supplied him cooked heads to cut up mainly because I wanted him to try and take as much of the hydraulic pivot hump out of the two inlet ports but he didn't seem interested - I don't know why .
I really think that its the smallish inlet ports in these EFI heads that ultimately limits how much air you can get to the cylinders , once they're as large as is possible (still not enough) the only other way to increase airflow is to increase boost pressure unfortunately . I reckon Subaru intended these engines to make torque at low revs and small ports (inhibited by the remnants of pushrod passages) was up to their ask .

If I had to guess I'd say that the VF8 worked better for you because of its larger trim turbine and the larger 20R vs 15R turbine housing .
I don't think its impossible to fit the VF10's smaller trim compressor wheel and matching comp housing onto the front of a VF8 but its the cost if it didn't work so well .
I've been on some of the Mazda boards , and the turbomaster one hunting up specs of some of these early RHB52 turbos , Mazda versions tend to use a bit more turbine (or a bit less compressor depending on which way you view it) and that fits in more with my way of thinking .
I like the idea of an engine that can make acceptable torque off boost (hard I know in the EA82's case) and use a turbocharger thats a touch healthy on the turbine/housing side and just about adequate compressor capacty to meet the power (airflow) goals .
So I've upped the static CR/ported the heads/fitted the spider/looking at a little more valve timing (cams) , a will do something with the header and turbo .
What I ultimately seek is about a 50 odd % increase in Hp (over the std 105) but I'd like it a bit more linear and less agro than I think yours was .

Finding middle ground should prove interesting .

Thanks and any pointers most welcome , cheers Adrian .

280WHP in an EA82T, thats over 300 at the crank and more than an EJ20 STI??? or do you mean 280 at the crank?

Have either of you (both QikRX and Discospud) ever followed the work a guy on the USMB who went by the name of WJM?

He had a RX that made 230HP with a TD04 and a WRX top mount. He did extensive porting work, copper headgaskets and a TWE header up pipe.

Have you got a dyno graph of your car. Im curious of the torque curve and how much it differs from that of a normal EA82T (which peaks at 2800rpm)

280HP at the wheels ?? Sheesh !! I am happy with the gains of a standard EA82T over standard EA81's and 2's.

Eating gearboxes - hows the new owner coping then ?

Gannon getting reasonable grunt from an EA82T is not difficult , expensive yes , but doing any total rebuilt engine is .
It's not really any better from a cost point of view with an EJ because you're doing the same sorts of things .

Both were a bit over engineered in that they don't fall to bits with normal use .
I think you gain on the swings and roundabouts becaue while the EJ is along more modern lines the extra 1-200 Kg of the Impreza or Liberty is not .

I think the real trick is to strike a balance of power output and linear power delivery .
Also to make EJ type power with an EA means your going deeper into its physical limitations and life expectancy because of things like its three main bearing crank and head casting issues .

The same basic principles work with all engines , doesn't matter if its a lawn mower or an F1 example . Like all the masters say , breathe more air to burn more fuel to make more horsepower .

EA82Ts have lots of breathing restrictions and thats why they don't crank out higher power numbers in bog stock form . To increase volumetric efficiency (cylinder filling efficiency) need porting better manifolds and cams . 7.7 CR is dark ages stuff with todays pump fuels .
To do anything serious you have to start out with rings and valves that seal and bearings and pumps that are known to be good . Its the age of EA82's that makes most original examples a bit breathless at best .
You really have to go inside to get at the things that contribute to a good overall result , there is no other way .

The "bottom" end of EA82's is not rocket science to pull apart and they are not that expensive to buy , just return everything to factory spec and if you want the CR to be this centuries thinking get the machinist to take a bit off the tops of the cases (head faces) . Cut four grooves for wire rings , not hard . Coat the pistons if you can afford to .
The real mods after the base is good is the heads and manifolds , need good or repaired heads and port them . NA cams would have to make a difference and are JUST a tad warmer than the best OEM turbo cams .
They need a slightly larger diameter header and something better than where the turbo side engine pipe (stub ?) joins it .
At the very least I think you need the 15R turbine housing off a VF10 or VF11 , this should go bolt on your standard EA82's turbo but you have to change to EJ type up and down pipe flanges .
It surprised me that the compressor and turbine wheels in the EA82Ts turbo are very nearly the same as in the US spec VF11 , the compressors inducer is a tiny smidge smaller and the turbine housing is 15R compared to 12R on the exact same turbine .
The fact that Subaru used an EA82 flanged 15R turbine housing on some far flung models means they knew the 12R was border line smallish , face it EA15R + NA like cams got these engines another 30 horses possibly with arachnid manifold and minor exhaust system differences . Thats nearly 25% extra with basically bolt ons . With a real header better exhaust and say WAIC it may get to 40-50% .
Not sure where they run out of fuel at but I think slightly larger electrically compatible injectors and a bit of a play with the "clock spring gear" inside the early gate AFM may just get things workable mixture wise .

QikRX , how much of a hairy goat was your car to drive around normally ?

Thanks all , cheers A .

Agreed with Adrian in regards to the flowing limitations of the EA82T, but once those limitations are addressed as much as possible, the results can be had.

Reliability is the main issue. As you may have read, Allan who bought the car twisted the block on the dyno not too long after he bought it. I went through 3 built engines. Piston failure (which were forged items), smashed rod/crank & then head failure. I'm the first to say that making power is not reliable.

Drivability was poor in the scheme of things. Idle needed to be set at 1250rpm with the said cams & sounded like a rotary on idle. First gear was all lag & literally had no power off boost. General cruising would require downshifting under 2700rpm to maintain speed going up a hill. On boost was a different story. Fuel economy was about 22L/100kms. The build was intended on it being a weekender only & a challenge for the builder (who is a mate of mine that owns an automotive engineering company).

I am really kicking myself now that the internal specifications & photos of the engine build weren't documented. Kinda hard when someone does the work for you...

I sort of wondered if she was a bit of a Tassie Devil at those power levels , to idle like that I'd say you had rally spec cams and they don't care what low rev performance and manners are like because in competition the revs don't get that low .

I will look up Tighes profiles and from memory they are ~ 270-280 degrees duration where Turbo L's were ~ 250 with 28 degrees of overlap .
Our spec NA MPFI cams are a bit longer at 256 with 32 degrees overlap but the 9.5:1 CR makes all the difference .
To get to 270+ duration you're probably looking at I guess 40 to 50 degrees overlap and the valve timings trapping efficiency is nowhere at low revs , and you don't have the NA engines higher CR to pull the bottom end power up with .
Plus siamesed exhaust ports and a single throttle inlet system mean reversion plays havoc at idle/low speed . As I said its of no consequence in a competition engine with a close ratio dog box behind it .

Yes I think I'll stay mild on the cams and try to get the most from a smaller bore header and slightly larger turbocharger than std like a VF10 .

Thank you very much for your input , its great when others can benefit from your experience and try to minimise the breakages .

Cheers A .

I like your honesty and frankness QikRX

QikRX can you remember precisely which injectors you ran , I know you had an aftermarket engine management so resistance values not so important but I'd like to know if what you used was electrically compatible with std 3 pug computer .

I won't need much bigger than std but can't afford to have it go lean at peak torque .

Cheers A .

imo if you wanted to be sure you're not going to go lean at good power, i'd get something like a microtech microfueler for around $550. simple to set up, once you reach a certain manifold pressure, it'll add fuel via 1 or 2 injectors. that way you can keep your standard injectors, and be fine for normal driving (especially economy wise)

IMO microfuelers are iffy because the injectors are rarely placed to get even fuel distribution .
Really the right way is to use larger injectors in the std position .
You can make some reasonably large across the board changes with those 3 plug computers and the vane type AFM .
When fuel injection doesn't involve ignition timing interceptors can be used to shorten the pulse width of larger injectors to maintain normal light load mixtures because it's usually just a 0-5v signal your interfering with .

Also large injectors are not a problem if you have a smart computer to run them with .
Skyline users often fit 740cc Nismo (Denso) injectors to an engine that had 370's std , run real clean if tuned properly .

Cheers A .

Tighe EA82 cam grinds . http://www.tighecams.com.au/cars.htm

Note the "Turbo Improved" profile , 23-68 65-25 meaning 48 degres of overlap timing . Bit more than the std Turbo cams 28 degrees or NA ones 32 .

then dont stuff around with half-a'd fixes, go for a full standalone unit and never have to worry about it again. get rid of the vaf, put in whatever injectors you like, run whatever boost you like, get good timing out of it. Think of how much you're going to spend on a half-a fix, think of how much your engine is worth, $1100 for a microtech lt10s is cheap insurance.

by the way, that's a bucket load of overlap for a turbocharged engine. absolutely massive. i guess 2 valves per cylinder doesnt help so it requires more overlap. dean tighe certainly knows what he's doing. we've had a lot of dealings with dean over the years. great guy but from what i can gather now he's semi-retired.

The injectors weren't standard but were never flow tested. They came with the rally engine & am not sure if the restrictions in the competitions allowed larger injectors. Used to run 18psi throught the stock ECU at one time with the VF-8 but without the cams. Not sure how the stock ECU handled it without pinging.
The tuner estimated the injectors to be 550's according to the pulse durations, A/F ratio & duty cycle.

The homologated rally parts kit included alternate computers ie vane or mass air sensors .
I don't remember about injectors though I wouldn't think you'd get 190 Hp with std ones .
I'll look up the homologated parts list later , domestic stuff ATM .

A .

The stock ECU is still mounted underneath near the steering column. Could get Allan to check the part number to see if it's the rally version.