I had a busy, but good Saturday. I calibrated my gtech pro com, weighed my prelude and did about 20 test runs, and a few runs at the drag strip. All that plus helped my wife in the garden, made lunch for the in-laws and ran a few errands, now i want to
I did this test to evaluate the gtech meter. I am very happy with the data it yields and think it will serve well for my own testing (and of course i'm still game to dyno the production parts). I could yap about how happy i am that this unit will allow me to do enough, good testing and how that will help make a better product in the end etc etc etc, but instead of words for you guys, i have some data you can play with. There is alot to look over here, so grad a cup of coffee, i hope you enjoy!
consistant enough? find me a DYNO that will do +/- 1.45 hp in a given set of runs. This method requires more data collection, but hey im not paying by the hour for this, so 20+ runs is no big deal. The tricky part was finding a good place (flat 1/8mile+ long and NO traffic). The dragway is good to, but the lines are a little long.
Here are my notes for the day:
******************
04/24/04 testing data
- acceleration tests performed at **** Plant parking lot
- prelim runs at Milan Dragway (not calculated) best 1/4 ET of 15.3, but i nailed a 0.443 reaction this time
- v3.5 4-1 header tested on 98 Prelude SH (with I, E, CAT, P, VAFC, mounts)
- tire pressure 32psi
- 93octane pump gas
weather:
local measured
temp (7:30pm): 55f
temp (11:00pm) 49f
wunderground.com data for 04/24/04 at 7:53pm
temp: 51.1f
dew point: 37f
humidity: 59%
pressure: 30.25"
wind: from ENE at 9.2mph
aero drag corrected, estimated, averaged, at-the-wheel, peak HP = 169.3
notes: light drizzle on pavement for last few runs, motor temp increased also.
***************
I'll share the data i generated today (*.gtp) you can get it HERE
you can view that data with the free GTech PASS software, found HERE
If anyone is an MS Excel wiz, want to help me by making a template for copying all HP/ TQ values from this*.csv file and averaging them? i could figure it out, but id guess someone else could do in in 5min, i only know basic Excel stuff.
can anyone point me to a link that has functions for doing SAE HP weather correction (pressure, humidity, temp)?
oh yeah, H22A4 redlines at 7400RPM right? does anyone have a more specific number?
header v1 dynojet measured, peak hp = 166.0 check it here
not to bad eh... thats right where i expected the number should be.
One last thing: please remember (or read up on) how the gtechPC works, in some ways it is more accuate than any dyno (engine loading, traction, aero, no straps), and in some ways it introduces other errors. But when you can do 50+pulls per day to test parts, i think it will yield equal if not better design test data. The important numbers will come when i do the same-day, same place comparative testing, but todays work makes me confident that that effort will be worth while.
I hope this was a fun read for you! time for
-Brett
Where or how did you measure your car's weight? That should be the most critical aspect of gauging performance. I thought about driving onto one of those truck scales, but how accurate can they be for a little import car?
__________________
97 EGP Base 5spd Prelude. Engine: DC short ram, DC SS headers, AEM pulleys, Random High flow cat, Mugen sport exhaust, Apexi VAFC. Suspension: Tein SS, Tanabe front lower tie, Neuspeed front upper strut, Neuspeed rear sway, Tanabe rear upper strut, DC rear lower tie, 17X7.5 5Zigen Typhoons with 215/40/17 Goodyear F1 GS-D3s. Other: Bride Ergo II, Spoon oil cap, Magnecor wires, relocated Odyssey 680 battery, Denso iridiums, Mugen pedals, Mugen lugnuts, shortened shifter w/ Momo airleather, Rotora slotted rotors and Endless NA-Y pads, Goodridge SS lines.
It actually doesn't matter what weight that you enter into the GTECH just so long as you enter the same weight in for the next test. The G-tech isn't any good for measuring real whp down to any specific number, it's all about repeatable tests that can be measured again and again. You could enter 3 tons as the weight, and you just want to see the benifit of whatever it is that you tune/do.
ive weighted the vehicle twice, once at a truck scale and just yesterday at a local rock yard, both times it was at 3200 - 3250 with gas and driver. empty curb wieght seems to be about 2950lbs. DR's race proped 5thgen is several hundred pounds lighter. but time is right, doesnt matter what the weight is...
I weighed once at 2950 at Texas Motorplex, though that was when I weighed around 170 and the car had NOTHING in the trunk and only about 1/8 (at best) a tank of gas. Car was a bone stock 97 base.
Thanks, Brett, for this... I think this will yield very consistent results. I would like to see how temps/correction affects results. I think the ONLY way to do this will be do do a baseline the same day (same weather conditions) or when conditions (temp/humidity) are within a few percent of each other (compared to any baseline). This is MUCH more convenient than a dyno, though.
That software looks nifty. What can you tell us all about the gtech pro as a product. what does it do that you think is very important... and what doesnt it do that you wish it could?
seeing these types of results (seeing an entire power curve), i may be very interested in one.
__________________
1999 Prelude 2.2 VTi 4WS
I speak fluent typo
Thanks Todd, yeah i'll need to be realy vigilant with the weather watching when im testing. I think I could test 2 manifolds, and do the test runs (at the same spot) all within 3-4hrs (ie drive to spot, test, drive back, add a little gas, swap parts, drive out, test). Hmm lets see what the weather change was yesterday in that time.... www.wunderground.com for 48105 04/24/04 says:
7:53 pm = 51.1f, 59% and 30.25"
11:53 pm = 48f, 61% and 30.18"
Thats not bad. It looks like the 12:00am- 3:00am time range has more stable weather, maybe i should make a "test after the sun goes down rule." Also, i could apply SAE correction to the Ecel files...
Nuro, the gtechPC is a pretty good unit, but the things i would change are: higher contrast screen, USB output (serrial is slow, but i know its easy to develop for).
The PC software is better than i expected (agorythmicaly finds powercurve)but could use a little more work though. For what im doing i would like it to be able to average all HP/TQ results from any number fo files (maybe all the files in a given dirrectory...), and also have a mannual adjusting factors (ie enter weather data, estimated aero/traction drag). In general though its a pretty "trick" device (they've obviously done alot of R&D and testing themselfs) and it will work for what im doing...
side note: GO BarF1 HONDA !!! Honda back in F1 again, next year they'll be up with Schumi i bet... Sato blew a motor though race replay is at 9est on speedtv.
Even the regular GTECH (old one) is pretty accurate and repeatable. Or as accurate and repeatable as the driver in the car. It is a good tool, and with the pc software is is a good way to do an a-to-b comparison.
Leptolude, the stock redline is 7500RPM (shows about 8000 on the stock tach) and the fuel cut hits @ 7600RPM (8200 on stock tach) This has been independently verified by both the helms, OBD2 dataloggers and rpm readings on a chassis dyno by myself. Most other people who have checked show similar results.
notorious, to get the HP data from a gtech you have drive in a "non-racing" way. If you want power data from the whole RPM range, you have to rev through the whole RPM range... just like on a dyno. so you get the car moving in 1st, then shift to 2nd at a low rpm, such that the engine speed is as low as possible, then floor it to do the whole sweep....
Looks like there have been vast improvements over the old model. Good to see. Glad it went well. I'm ready to move, just need notice as I do it after hours.
That makes sense. So basically from a 2nd gear roll our cars still post a better 0-60 time than a lot of the cars on the road (well except for those DTEC Hyundais lol)
I did another quick test tonight, to try for repeat accuracy. Considering wind and temp differences (see log texts) im happy with the results. When i get the time (this weekend hopefully) I'll test both manifolds late at night when the weather is more steady (after sun is down) and the test sight is totaly empty. I'll control as many of hte variables as i can!!!
todays work:
04/28/04 testing data
- acceleration tests performed at ****** Plant lot
- v3.5 4-1 header tested on 98 Prelude SH (with I, E, P, VAFC)
- tire pressure 32psi
- 93octane pump gas
aero loss estimate of 13.5hp (Cd = ~0.39, front area 1.84m^2 peak speed 64mph)
aero drag corrected, estimated average peak HP = 158.2
notes: All runs were done in same dirrection, (N to S) against wind, conditions were clear but gusty winds were present. At 7:00pm on weekday there was occasional traffic at test sight so I only did 3 runs. Weekend late night will be best testing time.Saw a cute fox running around, i'll wait until hes' asleap when i test next. This test was performed to estimate repeat accuracy of GTech unit. Considering gusty wind and temp differences from last session i'd say its doing pretty well.
*********************
Anyone have an old Auto textbook sitting around? I'd like to find the formula for weather correction, and apply it mannualy (just to the peak # for now, hopefully Tesla will add this to PASS one day).
This works pretty well, and if my memory serves me, it's the same as SAEJ1349 which seems to be the industry standard. It should work well for what you are doing.
Does Tdry need to be solved for with observed temp - vapor pressure also? I'll convert all of my measurements and make an Excel sheet for all of my averaging and correction.
Back to work for me. Today its finnishing investment casting patterns from the RP machine, 3D work for our VR project, finding a sand casting foundry to work with in MI and delivering some more RP models
I have never done anything to resolve Tdry, other than stick a thermocouple in the air. I've heard of some people getting really particular about modifying temperatures, but I've never seen it done, and I've never done it myself. Even HVAC engineers I've been around don't seem to consider it when working with temperature and humidity control.
Good Luck. I'm glad to see you working so diligently with that GTech.
I did a 23 run test last night with both manifolds, the test went well. I didnt get to bed until 2am and work is busy today so i havent had time to look at the data, or do the averaging excel work. i'll post ASAP The GTech was a good choice !!
time for another tripple latte
w00t!! I just got accepted to the ME grad program !!!!
realy busy with work, havent had time to make a nice excel sheet that will average multiple runs and graph them. Its a little tricky as RPM, HP and TQ are sampled by time not RPM so i've got to do some processing to do if i want to match up the cureves based on RPM. But, if anyone is currious, here are 2 graphs (red = 4-1, black = OE), i pick the two closest to the average of thier sets of runs (did 23 runs total). This is isnt as impressive as i'd like, but alot closer to what i assumed the manifold would do:
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
find me a DYNO that will do +/- 1.45 hp in a given set of runs. This method requires more data collection, but hey im not paying by the hour for this, so 20+ runs is no big deal. The tricky part was finding a good place (flat 1/8mile+ long and NO traffic). The dragway is good to, but the lines are a little long.
i could figure it out, but id guess someone else could do in in 5min, i only know basic Excel stuff.
race replay is at 9est on speedtv.
The GTech was a good choice !! 
Linear Mode
