I’ve written more about rolling roads and dynos than just about anything else on this web site. It shouldn’t be hard to guess that one of the things I really dislike is the nonsense flywheel power figures that get bandied around in the car magazines and by tuning firms based on rolling road tests where only the wheel bhp figure was really measured with any accuracy. Plenty of people have expressed a view to me that it doesn’t really matter whether the figures are accurate or not as long as they are repeatable and you can tell whether you achieved a decent power gain from your tuning modifications. My answer to that is that the only data that is of any use to someone who tries to do things scientifically is accurate data. In previous articles I make it clear that the only figure you should place any reliance on from a rolling road test is the wheel bhp figure. The dangers of relying on the supposed “flywheel” bhp figure is illustrated below from an actual tuning session I attended some years ago. Copyright David Baker and Puma Race Engines
In the mid 1990s I used to build a lot of CVH engines for the Fiesta XR2 Challenge and also the Stock Hatch series. The class rules mean that the engines have to stay standard internally and any power gains can only be achieved by very careful blueprinting and attention to detail. Building identical engines over and over again isn’t really very interesting and after winning the XR2 Challenge for three years in succession I moved on to other types of engine.
I still build the occasional race CVH these days and a couple of years ago I was approached by a competitor (we’ll call him Steve) in the Stock Hatch series who felt his engine was underpowered and wanted to discuss me building him a new one over the winter break. He’d built his current engine himself but had never had it set up on a rolling road on the basis that being basically standard it should be fine on the standard ignition timing and carb jetting. I suggested it would be a good idea to pay for a tuning session to find out how underpowered it was and that would help decide what needed to be done during the rebuild. I recommended a rolling road local to myself that I could pop down to during the tuning session and have a chat about things.
He duly booked a day and I said I’d turn up an hour or so after his start time to see how things had gone. A bog standard XR2 engine is rated at 96 PS by Ford (that’s about 94 bhp) and they show about 75 bhp at the wheels on an accurate set of rollers. With the exhaust system and carb mods that this race series allows I expect to see about 80 bhp on a home built engine and up to 10 bhp more for one I’ve blueprinted myself. So the target power figures for this test day were fairly clear – to me at least. When I arrived at the rolling road premises the job seemed to be just about finished. The fuelling and ignition timing had been checked and adjusted and a couple of power runs made.
I asked the tuning guy (let’s call him John), how things had gone. He said it was actually a pretty good engine and was showing about 100 bhp which was a decent bit up on standard. He was happy enough with this and Steve the driver looked happy but I wondered how this tied in with an engine that was supposedly underpowered. The first thing that should strike anyone used to reading my website is I don’t take a blind bit of notice of flywheel bhp figures so I asked John to let me see the wheel bhp data and that’s when things started to go downhill.
This alleged 100 bhp engine was only making 69 bhp at the wheels. That’s 6 bhp down on a standard car and at least 11 down on what I expected a home built Stock Hatch engine to make. The ridiculous 31 bhp the system was adding for transmission losses was blinding this supposedly experienced rolling road operator to the fact that the engine was actually crap. Now John knew my views about wheel and flywheel bhp well enough because we’d discussed them many times but he still wouldn’t look any further than the flywheel figure his machinery was producing.
So I asked them both to explain to me step by step what had been checked and how. First thing they’d done was the ignition timing. The Ford Figure is 12 degrees BTDC with the vacuum advance pipe disconnected. In my experience the race engines like a bit more advance than this. It turned out that the driver had never realised the vacuum pipe should be disconnected and when John had checked the timing he’d forgotten too. They both thought it was set at 12 BTDC but when I put the strobe light on the car myself it showed the timing was actually at 3 degrees ATDC with the pipe disconnected.
In other words the vacuum advance had been adding 15 degrees to the centrifugal advance and the true timing was retarded by this 15 degrees. Now that’s about the most basic tuning mistake you can ever make and no professional worth his salt should have been caught out by something that stupid. With the timing properly set to 14 BTDC, which is where I normally run them, the power leapt to 80 bhp at the wheels.
I asked John somewhat tongue in cheek what he thought about the previous 100 bhp at the flywheel now. I think the answer was mainly in the form of a red face and some muttering. Next thing I checked was the fuel mixture. 5% CO is the figure to aim for on a race engine but a lot of rolling road operators think it’s safer and less chance of the owner melting a piston and coming back to complain if you set them up a bit rich. Sure enough this one was at about 7% and I had to almost force a by now somewhat unhappy John to decrease the main jet by one size. With the CO showing 5% the power went up another 2 bhp to 82 bhp at the wheels.
The outcome of the story was rather ironic. John didn’t knock a penny off the tuning bill despite the fact that he’d cocked most of it up to start with. I spent an hour sorting things out which I didn’t ask for payment for because I’d only intended to go there for a break and a chat. The engine was now so powerful that it was fully competitive and I never heard back from the owner and effectively did myself out of an engine rebuild.
I suppose I should have kept my mouth shut and let Steve trail round at the back of the pack for the rest of the season. I’d have got an engine rebuild out of it and a reputation as a miracle worker because the very least he would have gained with one of my engines was the missing 13 bhp we found that day with a proper tuning session.
The moral is clear. By relying on a nonsense flywheel figure it looked like the engine was producing its target power output. So no one had been motivated to check any further to discover their mistakes in the timing and jetting. Basically just stupidity compounded on top of even more stupidity. There’s another angle to this as well. Very few standard engines actually make as much power as the manufacturers claim.
Partly because the average engine has already done 50,000 miles and is over the hill and partly because the quoted power figures will be towards the top end of what a randomly selected batch of engines would produce even if they were in tip top condition. But because nearly all rolling roads produce inflated power figures it makes a below average engine look better than it really this.
The brilliant bit is this – the rolling road operators then say “well my rollers must be accurate because every standard engine I test shows the standard bhp figure”. The total lack of logic in this is so mind blowing it renders me almost speechless (although thank god I can still type). Using engines that on average will be 5% or more below the claimed power output to justify as accurate rolling roads that on average read at least 5% high.