The Temple of VTEC - Honda and Acura Enthusiasts Online - Articles - 00-01 Plots: Could It Be a Ringer? We Find Out

After the initial results of TOV's dyno test were published, a firestorm of discussion began on internet forums about whether or not the results were accurate, representative, or had bearing on real world performance. To be honest, we wondered the same thing since the 04 S2000 didn't "feel" 26 hp faster, but I've always said that the butt dyno lies, and I include my own posterior based acceleration measurement device in that category.

The next set of questions TOV had to ask were, "how do we verify that these results are (1) representative of production vehicles and (2) that the results are not just dyno numbers, but translate into real world performance?" The second question was somewhat easier to address, so we began there. We knew we had a car (representative or not) that dyno'd at 243 whp. We also had my own personal MY00 S2000 that had dyno'd around 218 whp with a cold air intake. We also had dyno curves for both cars and a Vericom VC2000 measurement system. All we needed now was an open stretch of road and a little computer time for a simulation.

The first portion of the test involved measuring acceleration times for the cars over a multi-gear interval. At first we thought we'd stick to one gear to avoid shifting issues, but our real world experience told us that if we started below VTEC, the 04 was going to kick the tar out of the 00 due to the higher lower end grunt and tighter gearing (as we saw in side by side roll ons). So we decided to make a run starting at 30 mph in 1st gear and end it at 70 mph in 3rd gear. Yes, an arbitrary range, but it seemed as good as any and allowed us to avoid major speed limit violations.

Up first was the 04. With 1/3 of a tank of gas and a 210 lbs driver we made 4 passes down our favorite stretch of road, 2 in each direction. The car was very consistent in its performance, and averaged a time of 5.415 seconds with a standard deviation of only 0.042 seconds.

Next was the 00. With a half tank of gas and the same driver on the same stretch of road, within 20 minutes of the first test. It averaged 5.668 seconds with a standard deviation of 0.107.

The difference between the averages was 0.253 seconds, but what does that mean? To find out, we fed our dyno data into a nice little simulation we've developed in Excel to compute acceleration times. The absolute numbers predicted by the simulator were off by a couple of tenths, but the predicted difference was 0.237 seconds - very close to our observed difference. Furthermore, when the extra weight of gasoline in the MY00 was added to the simulation, the predicted difference reached 0.245 seconds. While we don't claim to have written the world's greatest simulator, the fact that the dyno data, simulator predictions and real world testing all support each other suggest that what we measured on the dyno performs as it should in the real world.

Next, we decided to see if Honda gave us a ringer. We have tested many Honda press cars in the past and HMC has never given us a car that didn't perform/test similarly to the consumer models, so we have no reason to doubt them. But, good journalism demands you question everything, so we tried. The most likely source of a "ringer" would be a tweaked ECU - and in fact this 04 S2000 did have a development ECU in it (as noted by a different part number). So, we obtained a production ECU from a local Honda dealership. This dealership had no idea we were testing a press car, did not know we were from TOV, etc. This was an off-the-shelf unit. We had the ECU rekeyed and put the car on the dyno to see what happened.

And what did happen? The answer is that nothing happened. Oh, the car started, ran and performed just fine, but the peak power number was within 0.1 hp of the previous peak power number. At a couple points in the curve the new ECU was up or down by 1-2 hp, but that's well within normal variation. Even the mixture curve was the same as exhibited with the original development ECU. I think its very safe to say that Honda did not put in a ringer ECU.

Furthermore, we datalogged ignition timing on the 04 to see if Honda had made any changes. On the older models, Honda would roll off the timing above 8000 rpms, which is why power began dropping after the 8300 rpm power peak. Not so on the 04. The timing stays flat right up to the limiter. Given the flat timing and good mixture, there isn't much to be gained in peak power with retuning (although there may be some low end gains - that's another article in and of itself).

So, does the 04 make more power? We're pretty damn sure it does. Oh, I suppose Honda could have slipped a hot motor in there with bigger cams or something, but you would think they'd reprogram the ECU too. Simply put, we believe the 04 makes more power and puts that power to effective use on the road. And all the data points we've collected so far support that belief very strongly.