Putting in the Boost

We borrowed a G-Tech meter for this exercise (thanks to Dave Smith) and found some flat back roads to do some testing. The G-Tech measures acceleration (or G-force) and is rated to 1/10^th of a second accuracy. It can measure 0-100 km/h and standing 400m times as well as the 400m terminal speed. The cars comprised a bog standard MY99 WRX with 14,000km on the clock, and a 20,000km MY99 with homemade cold air induction, upgraded air filter and a pneumatic variable boost control very similar to the AutoSpeed EXA system ("Project EXA"). The over-boost cut was defeated using a simple check valve and T-piece, allowing vacuum but not full boost to reach the sensor.The total cost for the above mods was just over $220.

Gentlemen - Start Your Engines!

We had often wondered if Subaru technicians tweak the turbo when loaning their cars to the magazines in order to get better acceleration figures. Our times show that this is probably not the case as they matched the Wheels and Motor magazines test times almost exactly. Our best 0-100 km/h time for the standard car was 5.71 secs. The 400m for this run came up in 14.14 with a trap speed of 156 km/h. The best for the modified car was 5.44 sec with a 13.89 sec at 159.1 km/h.

For the record, Wheels have recorded a 5.8 sec 0-100 in the MY98 and 14.2 sec at 147.6 km/h for the 400m. Motor rate the MY99 at 5.91 and 14.12 at 157.7 km/h. Note the standard MY98 runs a turbo pressure of 12.76 psi (0.88 Bar) whilst the MY99 standard boost is set to 13.6 psi or 0.94 Bar. Also of interest is that Wheels normally test with a full tank and single passenger aboard, while Motor tests with half a tank and only the driver aboard. This extra weight does make a difference! We ran the 0-100 and 400m with a near full tank and with only the driver on board. In order to reduce error, our figures in the table below show the averaged times for two runs in opposite directions.

	MY99 WRX Motor	MY98 WRX Wheels	Our Cars Standard	Our Cars Modified	STi WRX Wheels	STi WRX Motor
0-100 km/h	5.91	5.8	5.78	5.49	5.7	5.38
0-400m	14.12	14.20	14.23	13.96	13.80	13.59
400m speed	157.7	147.6	156.0	157.0	159.9	163.6
40-70km/h	-	4.2	3.8	3.9	2.8	-
60-90 km/h	-	3.2	3.2	2.8	2.1	-
80-110 km/h	-	3.3	3.1	2.6	2.3	-

For the in-gear acceleration figures, we had to rely on good old fashioned stopwatch technology. This obviously affected the accuracy of our times but it does show the relative difference between the two cars (which was the whole point of the exercise anyway). We recorded at least 6 runs for each speed increment, dropping the best and worst and averaging the rest to reduce error and obtain more consistency. The test car was also carrying a rear seat passenger for the exercise (who was doing the timing).

Despite the possible inaccuracies of the stopwatch method, the figures proved surprisingly close to that of Wheels. We only tested the 3^rd gear response as 4^th and 5^th gear are not something you would use in the real world (ie what does 5^th gear acceleration from 40 km/h really show?). It was bad enough running in 3^rd gear at 40km/h as only about 2500 rpm shows on the tacho. This proves an eternity when you're waiting for the boost to kick in! In reality, you'd change back to 2^nd and mash the go pedal.

We suspect the significantly faster 40-70 km/h time we obtained was due to the differences between the Wheels tested MY98 and our standard MY99. As the MY99 has a slightly smaller turbo than the MY98, spool up is quicker and this would make for a slightly quicker time when accelerating from off boost. Note we were unable to source any MY99 in-gear acceleration figures from any of the motoring magazines.

As you can see, the differences between the standard car and the boosted car were quite significant when under boost. The marginally slower 40-70 km/h time recorded for the boosted car may be stopwatch error or a function of the pneumatic boost control. We have seen data for a modified MY98 (pneumatic boost control at 17psi and air filter) that shows power with the bleed in place is actually lower than standard ECU below 3000rpm. This would confirm our marginally slower 40 - 70 km/h time.

Horses

Another parameter the G-Tech claims to measure is the delivered horsepower (imperial, not metric) at the wheels. We recorded the HP numbers for several runs in both cars including the HP at variable boost (from 7psi through to 18 psi) in the modified car. The figures we got were:

Standard Car	174hp	130kW
Modified Car at 18psi	201hp	150kW

We found that the G-Tech HP figures were consistently too high, based on dyno-proven figures of approximately 100kW at the wheels. This inconsistency may be due to the 4WD drivetrain of the WRX, or the high instantaneous G's that 4WD cars provide on launch, which can possibly create error in the G-Tech measurements. Although the recorded figures are too high, the relative difference between the cars is 20 kW at the wheels - although of course only a true dyno test would confirm this.

To estimate the power at the flywheel for the modified car, we simply scale the result for the modified car by the known figures for a standard MY99. This assumes that the standard car has exactly 160kW at the flywheel as per Subaru's specification, and that the drivetrain losses between the two cars are identical.

150 * (160.0 / 130) = 185kW at the flywheel.

Based on the acceleration figures we recorded, 185kW for the modified car would seem about right, given the STi is quoted at around 196kW on Australian fuel and our boosted car was only marginally slower.

The Final Word

There was no doubt that an extra 5 psi of boost in conjunction with the induction mods gives the car far more grunt. The times show that significant performance gains can be made at very little cost by simply tweaking the turbo pressure. In fact the performance gains that can be made almost match the Ferrari-fast STi! When comparing this to the $3000+ cost of an enlarged exhaust, upgraded ECU and other go-fast bits, this would appear to be an attractive option.

However, having said this, there is some concern about the long-term effects on the MY99 running at higher boost (especially 18 psi) without a properly dyno-tuned aftermarket ECU. Anecdotal evidence suggests the WRX can run up to 16 or 17 psi without problems. We do not wish to argue against this but keep in mind that, although the STi runs around 17 psi, it has different engine internals to cope with the extra stress of running higher boost pressures.

(However, as with all turbo car boost increases, if the mixtures stay appropriately rich and there is no detonation, the chances of engine damage are remote. With the top-mount standard intercooler in place, it would be wise to turn the boost down to standard in hot weather, especially if travelling at sustained high speeds or in slow stop-start city traffic - Editor)

One very noticeable drawback of running 18 psi boost was the increased fuel consumption. Both tanks were filled prior to testing and after a morning of roughly equal road usage and testing, the boosted car had used almost double the amount of fuel used by the standard car. Immediately after testing, the standard car was sitting on 3/4 full, whilst the boosted car was just above half. This result was most unexpected. One theory for the increased fuel consumption is that because the MAP sensor is being bled off (to bypass the fuel cut at 15.5 psi) the factory ECU is supplying more fuel as a safety measure to prevent detonation because other sensors in the car tell the ECU that boost is higher than the MAP sensor measurement eg rpm, airflow, throttle position, load, air/fuel ratio etc.

A final point of interest is the marked difference in feel between the two cars. Although feeling faster, the boosted car felt very unrefined and lacked the smoothness of the standard car in its power delivery. The modified car was very jerky when backing off the throttle during acceleration. This was noted by both drivers although ironically, each preferred the feel of the other's car........