I think your test procedure is excellent, but still subject to minute variances. I am a little skeptical about an improvement of over a half second just with a filter change. But I did think I felt an improvement in throttle response with the Comptech filter over the stock unit so maybe another test will verify it.

 

The stock filter was really dirty, that may have a lot to do with it. As for the procedure, I'm not sure if I described it before but, believe me, it is very very accurate.

This kind of precision is only possible with the S2000 and not with other cars. I'll explain why:

The S2K rev counter is extremelly accurate. Furthermore, the lines that light up as the engine revs up have a different width at the krpm divisions. So they are extremelly easy to spot. With a passenger it's even simpler as you can focus on the driving while the passenger does the measurement. You can measure improvements in whatever rpm range you choose.

You do have to make sure that you do exactly the same stretch of road. The way to achieve this is to pass a given point ingear at a set speed but under the rev range you want to measure, then just wait for the fatter line to light up and start your chrono.

It's a poor man dyno. But I have repeated the test several times with reproduceable results (looking at my records: once on another test I got: 13.11 13.06 13.28 13.10s. This is the kind of variations you get under the same testing conditions).

Short of a dyno run, I would like people to provide this kind of information when they do performance mods. The "the car feels much more responsive now" or any other butt-o-meter type of assessment is fraught with uncertainty. It's a psyche thing, people spend money and their perception will be altered...

 

set up a specific confidence interval (90%, 95%, 99%, etc.) all you need are the samples, calculate the mean, standard deviation, and use the t table (less than 30 samples) so you can find the mean time (normal distribution).

i'm studying stats, and it's filling my head with weird stuff

 

[QUOTE]Originally posted by mingster:
[B]set up a specific confidence interval (90%, 95%, 99%, etc.) all you need are the samples, calculate the mean, standard deviation, and use the t table (less than 30 samples) so you can find the mean time (normal distribution).

 

JR Filter website does not seem to show the filter as available for the S2000. What is the part number? Where did you make your purchase? TIA dgl1 [QUOTE]Originally posted by Luis:
[B]Just did an initial in gear acceleration test.
From 6Krpm to 9Krpm with the stock filter I got 11.45 seconds and 10.84 seconds with the JR filter. I just did one run, so take this number with a grain of salt, and consider as well that the stock filter has 25,000 kms on it, so it's not spotlessly clean.

Still, it looks promising...

The site is www.jrfilters.com.

 

I'm game for a GB for one of these...anyone else?

--------------
from their website:
JR Airfilters are manufactured with 4 layers of chirurgical cotton, which are hold together by
2 layers of steel wire mesh. To prevent rust, they are galvanized and covered with plastic
which give a double security.

Already at the factory we charge the filter with a special oil of blue colour. This oil is an very
important element of the filter. Due to the construction, the material and the oil, the dust
which is in the air, is retained outside of the filter and serve as supplementary filter element.
That

 

i contacted them for a GB, but either they don't understand English or they're just ignoring a group of good customers with real cash.

 

[QUOTE]Originally posted by dgl1:
[B]JR Filter website does not seem to show the filter as available for the S2000.

 

Mingster, They will probably get back to you. I had a question on a different matter and they took a while to reply. Try emailing directly to jans@happynet.lu.

 

i did bro, and i'll wait.