This article was first published in 2001.
A turbo upgrade is one of the riskiest ways of improving performance.
Huh? But isn't a bigger turbo s'pposed to do wondrous things for grunt? Well, the answer to that is - sometimes. Other times having a bigger turbo fitted results in a hefty bill, monstrous turbo lag - and yes, good performances gains up the very top. But in city-driven car, that high rpm performance is often nearly unusable, while the performance deficit elsewhere in the rev range can be horrendous.
But what of turbo swaps on the Impreza WRX? Here there are a couple of frequently carried out changes which are said to represent a major, almost-no-downside performance improvement. One of the most popular is installing the IHI VF22 - fuller name, IHI RHF5 VF22. Costing about $2500 fitted, this turbo is available in literally a bolt-on configuration - all oil and water fittings the right size and shape and in the correct spots, turbine flange patterns identical, hose diameters the same, wastegate actuator in the same orientation. And in one fell swoop that overcomes at least some of the common problems with turbo swaps - paying for the labour involved in adapting a foreign body into place.
So we decided to take a look at the installation of a VF22 into an October 1999 Impreza WRX. Dominic Rigoli was the operator, and Tony Rigoli Performance on the Gold Coast the location. (For those who don't know, Dominic's own WRX is the fastest in Australia - if not the world - with a quarter mile time at the time of writing of ten seconds.)
The Starting Point
The silver Rex first headed for the four-wheel drive dyno to let us ascertain a starting point. Importantly - and seldom mentioned - is the way in which WRXs respond to mods can vary substantially from model to model. There have been some cam and other specification changes through the model progression, in addition to the better-known changes in factory turbo sizes. So, while it's a safe bet to say that performing the normal mods - boost, exhaust, intake - will result in a power gain, specifying how much power will be the result is much harder.
This particular WRX already had an Apexi exhaust with a custom twin-off-the-turbo dump, the latter a mod which incidentally Domenic suggests does almost nothing for performance over a more conventional (and cheaper!) single large pipe matched to the flange.
Also under the bonnet was an HKS Super Power Flow filter...
... and a front-mount Apexi intercooler. Yep, there was a reason for that Apexi sticker on the back window! The electronic boost controller wasn't, however, by Apexi - instead it was a G-Reddy Profec B. (As another aside, Dominic suggests that upgrading the factory intercooler should occur after
a turbo swap, rather than before. In other words, he reckons the sequence should be along the lines of: boost, intake and exhaust - and then a turbo swap and an intercooler upgrade to front-mount. Forget the aftermarket underbonnet intercoolers, he suggests, and don't go for the front-mount intercooler until you've got a turbo big enough to really use it.)
After that host of high-priced mods, you might have expected the over-boost defeater to be another brand-name goody. But it wasn't. Instead it consisted of a T-piece in the hose to the pressure sensor, bleeding out the excessive boost through a one-way valve.
So, onto the dyno and the Rex, boosted to a peak of 17 psi, developed 142hp (106kW) at the wheels. This was lower than Dominic has seen on other cars equipped with these mods - as we said, it does vary from car to car. And while boost scaled to 17 pounds, by the redline it was back at only 13 psi - the factory small turbo really starting to restrict flow.
Out came the spanners and off came the turbo.
It really is a quite straightforward removal process, with the only even slightly tricky bit involving the oil return hose from the turbo. This is held onto the turbo pipe with a clip which is almost impossible to get to, but with everything else removed (intake hose, compressor outlet hose, exhaust pipe, turbine flange bolts) it's possible to pull the turbo out without releasing this clip - the clip's spring design giving this degree of flexibility.
Then, as we keep saying, you just reverse the procedure with the new IHI VF22. Again, the only trick involves that oil hose - and don't make the mistake that we heard about where someone thought that they'd pushed the hose home, only to find on start-up that a flood of oil was coming out under the car.... In addition to making sure that you get that oil drain hose back on, be careful when refilling the coolant system, as it's very easy to get air pockets. Fill the system slowly and carefully, squeezing out air bubbles by squashing with your fingers the top radiator hose, and then when you're reckon everything's apples, make sure when the car warms up and the rad fans start, that they're blowing hot air. Not cold air....
Resting in place, the new turbo looks so close to factory that you have to be an expert (or read the numbers, or both) to pick the change. Certainly, this is one modification that a typical uneducated registration inspector or policeperson would never pick...
The one and only change that had to be made to any other component was that the factory heat shield (a new one supplied by the owner) needed to be modified to let the big twin dump pipe escape from the turbo.
So what are the differences between the factory 99-spec TD04L blower and the VF22? In these pics we've actually lined up an internally smaller VF23 - externally the VF22 and VF23 are identical. Compared with the 40mm standard unit, the compressor on the VF22 has an eye diameter of 45mm. (Note that the OD of the compressor intake is the same on both turbos at 50mm; it's the degree of taper of the ID that is different - in addition to the wheel sizes.)
On the turbine side, the std jobbie is 40mm, versus the new 'un's 45mm. So, in plain English, the new roller bearing turbo is just a bit bigger at both ends.
Back on the Dyno
With these mods there are no problems with the car running lean on the factory management; in fact Dominic says that the cars will detonate (ie be running too much timing advance for the mods) long before they run out of fuel. If anything, they can be too rich!
And on the dyno the difference in power with the new turbo fitted was immediately noticeable.
Peak power went from 142 to 176hp, a gain of 24 per cent. Aaah, but was that at higher boost? With the GReddy boost control settings unchanged, peak boost had indeed risen by 1 psi to 18 psi - so a little bit of the power gain would have come from that. But more important was the way the boost level was retained better at high rpm. The boost level still dropped, but at peak rpm it was now 15 psi, versus the previous turbo's 13 psi. A better comparo can be seen at 129 km/h (all dyno runs were done in third gear) where boost rose 3 psi from 12.7 to 15.7 psi - the VF22 was able to hold boost better at higher engine revs. At this point the power was up by 30 per cent!
On the road the car - from the passenger seat at least - is still quite responsive... none of that big-turbo-no-go-down-low feeling. Like the factory STi, you could catch it off boost in some driving situations, but even if you were a gear higher than optimal, it wouldn't be a disaster.
Unlike many so-called bolt-on upgrades, this one is really just that. And while it's hard to work out how many at-the-flywheel horses were gained from the change, a peak power lift of 24 per cent and a top-end gain as high as 30 percent are pretty major improvements for the $2500 cost that this step set back the owner...
Tony Rigoli Performance (Gold Coast)
07 5528 4666
In Issue #120 we covered the engine upgrades to Mick Jones' MY99 WRX, which had similar mods - including the fitting of a VF22 turbo. So how did that car and this one compare for power? Running about 17.5 psi boost (a fraction less than the car covered above) but with very similar mods, Mick's car developed a peak power (on a different dyno!) of 229hp. That's a lot more than 175, isn't it?! However, as we've said time and time again, comparisons between dynos - especially when the runs are done in different gears, as here - don't mean much at all. More relevant is the gain that Mick's car experienced when the new turbo was bolted up after the exhaust, intake and intercooler had been done. Power on his car was up by 23 per cent at the very top end (30 per cent on the car covered here) while peak power improved by 20 per cent (24 per cent on this car). So as can be seen, the proportional gains are similar.