Castrol Top Shop Challenge Results


 EngineTeamTurbo or NAPeak BHP & BHP/L corr.Magazine
#1Mitsubishi 4G63 2.0LAMS (Chicago)Garrett GT42R @ 41.5psi1035 whp 258.8hp/LModified
#2Nissan SR20DET 2.1LG-Dimension/Phase 2Garrett GT42R @ 34.3psi827 bhp 196.9hp/LImport Tuner
#3Audi 034 2.3L (5 cyl.)034 Motorsports
Garrett GT42RS @ 38.2psi848 bhp 184.3hp/LEuro Tuner
#4Nissan RB26 2.8LSP EngineeringHKS T51RSPLBB @ 34.8psi966 bhp 172.5hp/LTurbo
#5Honda F22A 2.3LBisimoto
Normally Aspirated360 bhp 156.5hp/LSuper Street
#6Nissan VQ35DE 3.8LCosworthNormally Aspirated438 bhp 115.3hp/LSport Cmpct Car
#7Chevy Big Block 427 7.0LAce MachinePrecision T70 x 2 @ 10.2psi1092 bhp 78hp/LLow Rider

The above table uses the corrected horsepower/Liter using the displacement multiplier of 2 for turbocharged engines. All engines were supposed to be run at the same facility and on VP 100 octane unleaded race fuel. The dyno facility was Speed-O-Motive in West Covina, CA where they use a reputable DTS engine dyno. The contest consisted of several portions: hp/L, area under the curve, a 30 minute durability run, and a subjective judging portion. Turbo engines used a displacement multiplier of 2 (e.g. turbo 2.0Lx2=4.0L to calculate the hp/L). For some reason in the pamphlet that Source/Castrol published, they did not use the horspower/L calculation in the rules.

This was the first time that Sourceinterlink (mag publishers) did this kind of dyno contest and while there were things that were learned for next time, there probably will not be a next time. This contest was a huge undertaking to begin with. From the beginning I was asking Elliot and Jay, “Dude are you sure you’re going to engine dyno 7 different engines?!?! That’s 7 different harnesses, ecus, bellhousings, splines, etc. you have to design, build, and configure each time.” I’m sure it was a pain in the ass both physically and logistically, but Elliot/Source did a great job organizing the whole thing and Castrol was really good with supplying all of us contestants with plenty of Syntec lubricants. Anyhow, running out of time and man power, the Cosworth VQ was not run on the same Speedomotive engine dyno. It was run at our own engine dyno facility. That’s too bad since our dynos are super greedy and calibrated to OEM standards. It didn’t matter too much since we probably weren’t going to win anyway.

I didn’t think of it from the beginning, but boosted engines should have had a displacement multiplier based on their boost. The seemingly most heavily weighted part of the contest was hp/L. Since we live and breathe in 1bar or 14.5psi of atmospheric pressure, every additional 14.5psi of boost pressure is another atmosphere. So if a team decided to run 14.5psig, their displacement multiplier should have been x 2. If they ran 29psig, their multiplier should have been x 3. If they ran 43.5psig, their multiplier should have been x 4. If they ran 58psig, their multiplier should have been x 5. Here’s how the results probably would have looked like if the results were based on horsepower per liter alone:

 EngineTeamTurbo or NAPeak BHP & BHP/L corr.New Multiplier
#1Honda F22A 2.4LBisimoto
Normally Aspirated360 bhp 150.00hp/L1.00x
#2Mitsubishi 4G63 2.0LAMS (Chicago)Garrett GT42R @ 41.5psi1035 whp 134.06hp/L3.86x or 7.72L
#3Nissan SR20DET 2.1LG-Dimension/Phase 2Garrett GT42R @ 34.3psi827 bhp 117.14hp/L3.36x or 7.06L
#4Nissan VQ35DE 3.8LCosworthNormally Aspirated438 bhp 115.3hp/L1.00x
#5Audi 034 2.3L (5 cyl.)034 Motorsports
Garrett GT42RS @ 38.2psi848 bhp 101.56hp/L3.63x or 8.35L
#6Nissan RB26 2.8LSP EngineeringHKS T51RSPLBB @ 34.8psi966 bhp 101.47hp/L3.4x or 9.52L
#7Chevy Big Block 427 7.0LAce MachinePrecision T70 x 2 @ 10.2psi1092 bhp 91.76hp/L1.70x or 11.9L

Now the numbers look alot more reasonable when you compare them against each other, eh? Yes, you read that right: Bisi made the most hp/L if the rules took into account boost pressure (using a displacement multiplier based on boost pressure/number of atmospheres). It’s a shitload more difficult to be making 360bhp out of a single cam 1991 Honda Accord engine than it is to make big power with any turbocharged engine. It’s taken Bisi YEARS of F22A development for him to build an engine that wicked. I dig the shower injectors too. I remember when he used to come to XS to dyno his race car back in 2003 and he was racing the F22A for years before then. Anyhow, I’m not taking anything away from any of the other teams since all of the results were pretty strong (except for Cosworth’s in comparison – see below), but if the contest were based solely on hp/L with a displacement multiplier based on boost pressure, then these would have been the results. Of course the contest was not based only on hp/L because then every team would have used a GT60 turbo for an outrageously big hp number. I made this chart purely for the sake of comparison. I’m not saying things should have been any different (this time). It was a learning experience for everybody.

As for a VQ35DE engine that you could build yourself using off the shelf Cosworth parts, don’t you think 438 normally aspirated hp is pretty strong? I sure do. And that’s exactly what we did: build an engine we were using for customer long block development. Sure it’s our fault we brought a knife to a gunfight, but we knew it going in and more or less did it to show what a customer’s VQ35DE could do (and the media coverage of course). That’s probably a pretty damn realistic 360hp at the wheels on a chassis dyno. Oh well, maybe next time.

There was a vote on the Castrol Top Shop website, but it’s no longer there for whatever reason. Sport Compact Car Magazine and the Cosworth VQ build won the vote. Thanks guys for voting (especially my high school buddy Jim in Taipei).

There’s Martin from AMS and his guy. Great job guys. 1035bhp on 100 octane is buff regardless of any multipliers.

Here’s Bisi. This guy builds the world’s buffest F22A!

Here’s Tyler on the left, my Swedish buddy Magnus in the center, and me on the right. Maybe since Tyler and I are both from the Eastside we never smile in pics?

Leave a Reply

Your email address will not be published. Required fields are marked *