Go Back   986 Forum - for Porsche Boxster & Cayman Owners > Porsche Boxster & Cayman Forums > Performance and Technical chat

Reply
 
LinkBack Thread Tools Display Modes
Old 03-09-2015, 09:53 AM   #1
Registered User
 
steved0x's Avatar
 
Join Date: Dec 2012
Location: FL
Posts: 3,878
Durametric pulls data before and after UDP

I finished the install of the UDP in January and have been sitting on this data, and decided to post it. I was going to make some more "after" runs but it doesn't look like I am going to get to that anytime soon so I thought I would post this here.

The baseline is the first set of pulls, 2000 S ~115,000 miles, all stock except for 2nd cat deletes. Then I installed the UDP and did a second set of pulls.

Both days, it was about 70F outside, and I had a full tank of gas, and around 500 pounds of people.

For each pull I got close to 3000 RPM and then WOT through 6,000 and I was able to do this for 1st, 2nd, 3rd, and a little bit of 4th. Otherwise you get going too fast

I was able to get around 5-6 data points a second using the Durametric (since then I have heard that you can disable the display of graphing on the laptop and get more reads per second, I may investigate that in the future). One problem with this, is that I don't always get a data point exactly at 3000, 4000, etc....

To work around this I had to do some extrapolation in order to get start and end times that fell on evenly divisible RPM ranges. With the durametric and my laptop I would get around 5 readings per second. So if one reading was like this (I am doing a very simple example and just using the seconds part):
1.25 3950
?.?? 4000 (I want to approximate the time at 4000)
1.75 4050

I would figure where 4000 was in between 3950 and 4050. I would see that 3950 to 4000 was 50% of the value, and 4000 to 4050 was 50% of the value. Then I would subtract 1.25 from 1.75 and split up the difference using the percentages from the rpm split. it is not exact (and I tested my formula and compared it to actual values) but it was very close to within a hundredth or more when I compared with actual values. Since it is not truly linear it was the best I could do. I guess I could see at what rate the acceleration changes and use that rate but for this I just used the formulas described above)

For example here is one of my 1st gear pulls. See how I had to extrapolate the time at 3,000 and 6,000 RPM.

The formula for the time in seconds for 3000 RPM is this: =((M7-M6)/(M8-M6))*(L8-L6)+L6
The 3000 is in cell M7 and this formula was pasted in L7. The label "Time in seconds" is cell L4. I also played with a formula by which for each reading I calculated the interval for a single revolution and used that, also where I did the same formula as above averaged with the result of subtracting from the subsequent time instead of adding to the previous time. They all came out very close and several decimal points out so I just went with this.
===========
Time in Seconds____RPM_____KPH_____MPH
45.8996134_______2854_____23______14.3
46.1086253_______2972_____26______16.2
46.12119012______3000
46.3186373_______3440_____29______18.0
46.5296494_______3814_____33______20.5
46.7486619_______4239_____36______22.4
46.9596740_______4683_____40______24.9
47.1806866_______5106_____44______27.3
47.3956989_______5608_____48______29.8
47.61621154______6000
47.6207118_______6008_____52______32.3
47.8327239_______6465_____55______34.2
===========

Before UDP - Dec 31 2014 (or maybe Dec 30)
After UDP - Feb 8 2015

1st gear pull 3000-6000 RPM
Before UDP
Run 1 - 1.570449764 (time in seconds)
Run 2 - 1.5407802

After UDP
Run 1 - 3000-6000 RPM 1.495021421
Run 2 - 3000-6000 RPM 1.519046372


2nd gear pull 3000-6000 RPM
Before UDP
Run 1 - 4.264609568
Run 2 - 4.158200102

After UDP
Run 1 - 4.120548992
Run 2 - 4.017960617

3rd Gear pull 3000-6000 RPM
Run 1 - 8.726394609
Run 2 - 8.979013562

After UDP
Run 1 - 8.710858422
Run 2 - 8.813156252

4th gear pull
Before UDP
3000-4000 4.941685313
4000-5000 4.973688321
5000-6000 5.187585148
Total
3000-6000 15.10295878

I did have another 4th gear pull before UDP for reference but it only went to 5000:
3000-4000 RPM 4.937129145
4000-5000 RPM 4.821128471
3000-5000 RPM 9.758257615

I only got 1 run through 6000 in 4th and it was before UDP. Below are comparisons for 3000-5912 which I did get to in my after runs.

4th Gear Pull 3000-5912 RPM (Ran out of road/room...)
Before UDP
Run 1 - 14.59862228

After UDP
Run 1 - 15.32816243

The 3000-4000 and 4000-5000 also were behind on my post UDP run... I only was able to make one run, maybe I failed to "launch" somehow? I did not log throttle position, but even if I was not on 100% throttle by 3,000 RPM, certainly by the 4000-5000 RPM band I would have been caught back up, but that pull was slower too

Range/Before UDP/After UDP
3000-4000 4.941685313 5.102526305
4000-5000 4.973688321 5.143059695
5000-5912 4.683248645 5.082576431
3000-5912 14.59862228 15.32816243

In 1-3 all of my pulls were faster post UDP, and in 4th they were all slower. I wish I could have done additional runs and I just might. (I am going to check elevation data where I did my post UDP run to see if maybe I was going up a hill?) I am grasping at straws to find a reason

I can definitely say that I can feel it in the "butt dyno" after installing the UDP, but I can offer no explanation as to why the 4th gear pull was slower after the UDP than before, except to say that I was on different roads, etc... and those same conditions would also apply to all the other runs so I can't say how valid they are either....

Last edited by steved0x; 03-09-2015 at 10:32 AM.
steved0x is offline   Reply With Quote
Old 03-09-2015, 10:50 AM   #2
Registered User
 
woodsman's Avatar
 
Join Date: Feb 2013
Location: Land of naught
Posts: 1,302
Wow neat idea. Can you run in reverse direction on the same stretch of road? Could also have been a headwind.
__________________
Death is certain, life is not.
woodsman is offline   Reply With Quote
Old 03-09-2015, 11:52 AM   #3
Registered User
 
Join Date: Feb 2005
Location: It's a kind of magic.....
Posts: 5,102
Quote:
Originally Posted by steved0x View Post
I finished the install of the UDP in January and have been sitting on this data, and decided to post it. I was going to make some more "after" runs but it doesn't look like I am going to get to that anytime soon so I thought I would post this here.

The baseline is the first set of pulls, 2000 S ~115,000 miles, all stock except for 2nd cat deletes. Then I installed the UDP and did a second set of pulls.

Both days, it was about 70F outside, and I had a full tank of gas, and around 500 pounds of people.

For each pull I got close to 3000 RPM and then WOT through 6,000 and I was able to do this for 1st, 2nd, 3rd, and a little bit of 4th. Otherwise you get going too fast

I was able to get around 5-6 data points a second using the Durametric (since then I have heard that you can disable the display of graphing on the laptop and get more reads per second, I may investigate that in the future). One problem with this, is that I don't always get a data point exactly at 3000, 4000, etc....

To work around this I had to do some extrapolation in order to get start and end times that fell on evenly divisible RPM ranges. With the durametric and my laptop I would get around 5 readings per second. So if one reading was like this (I am doing a very simple example and just using the seconds part):
1.25 3950
?.?? 4000 (I want to approximate the time at 4000)
1.75 4050

I would figure where 4000 was in between 3950 and 4050. I would see that 3950 to 4000 was 50% of the value, and 4000 to 4050 was 50% of the value. Then I would subtract 1.25 from 1.75 and split up the difference using the percentages from the rpm split. it is not exact (and I tested my formula and compared it to actual values) but it was very close to within a hundredth or more when I compared with actual values. Since it is not truly linear it was the best I could do. I guess I could see at what rate the acceleration changes and use that rate but for this I just used the formulas described above)

For example here is one of my 1st gear pulls. See how I had to extrapolate the time at 3,000 and 6,000 RPM.

The formula for the time in seconds for 3000 RPM is this: =((M7-M6)/(M8-M6))*(L8-L6)+L6
The 3000 is in cell M7 and this formula was pasted in L7. The label "Time in seconds" is cell L4. I also played with a formula by which for each reading I calculated the interval for a single revolution and used that, also where I did the same formula as above averaged with the result of subtracting from the subsequent time instead of adding to the previous time. They all came out very close and several decimal points out so I just went with this.
===========
Time in Seconds____RPM_____KPH_____MPH
45.8996134_______2854_____23______14.3
46.1086253_______2972_____26______16.2
46.12119012______3000
46.3186373_______3440_____29______18.0
46.5296494_______3814_____33______20.5
46.7486619_______4239_____36______22.4
46.9596740_______4683_____40______24.9
47.1806866_______5106_____44______27.3
47.3956989_______5608_____48______29.8
47.61621154______6000
47.6207118_______6008_____52______32.3
47.8327239_______6465_____55______34.2
===========

Before UDP - Dec 31 2014 (or maybe Dec 30)
After UDP - Feb 8 2015

1st gear pull 3000-6000 RPM
Before UDP
Run 1 - 1.570449764 (time in seconds)
Run 2 - 1.5407802

After UDP
Run 1 - 3000-6000 RPM 1.495021421
Run 2 - 3000-6000 RPM 1.519046372


2nd gear pull 3000-6000 RPM
Before UDP
Run 1 - 4.264609568
Run 2 - 4.158200102

After UDP
Run 1 - 4.120548992
Run 2 - 4.017960617

3rd Gear pull 3000-6000 RPM
Run 1 - 8.726394609
Run 2 - 8.979013562

After UDP
Run 1 - 8.710858422
Run 2 - 8.813156252

4th gear pull
Before UDP
3000-4000 4.941685313
4000-5000 4.973688321
5000-6000 5.187585148
Total
3000-6000 15.10295878

I did have another 4th gear pull before UDP for reference but it only went to 5000:
3000-4000 RPM 4.937129145
4000-5000 RPM 4.821128471
3000-5000 RPM 9.758257615

I only got 1 run through 6000 in 4th and it was before UDP. Below are comparisons for 3000-5912 which I did get to in my after runs.

4th Gear Pull 3000-5912 RPM (Ran out of road/room...)
Before UDP
Run 1 - 14.59862228

After UDP
Run 1 - 15.32816243

The 3000-4000 and 4000-5000 also were behind on my post UDP run... I only was able to make one run, maybe I failed to "launch" somehow? I did not log throttle position, but even if I was not on 100% throttle by 3,000 RPM, certainly by the 4000-5000 RPM band I would have been caught back up, but that pull was slower too

Range/Before UDP/After UDP
3000-4000 4.941685313 5.102526305
4000-5000 4.973688321 5.143059695
5000-5912 4.683248645 5.082576431
3000-5912 14.59862228 15.32816243

In 1-3 all of my pulls were faster post UDP, and in 4th they were all slower. I wish I could have done additional runs and I just might. (I am going to check elevation data where I did my post UDP run to see if maybe I was going up a hill?) I am grasping at straws to find a reason

I can definitely say that I can feel it in the "butt dyno" after installing the UDP, but I can offer no explanation as to why the 4th gear pull was slower after the UDP than before, except to say that I was on different roads, etc... and those same conditions would also apply to all the other runs so I can't say how valid they are either....
Sorry, but this why dyno runs are done under controlled circumstances. From Dec to Feb, a lot of things (air temperature, air density, humidity, barometric pressure, wind direction, road surface, etc.) may have changed dramatically enough that you could have seen similar differences without any changes. The validity of experimental data depends upon controlling all of the possible variables other than the one under study.
__________________
Anything really new is invented only in one’s youth. Later, one becomes more experienced, more famous – and more stupid.” - Albert Einstein

Last edited by JFP in PA; 03-09-2015 at 11:58 AM.
JFP in PA is offline   Reply With Quote
Old 03-09-2015, 12:09 PM   #4
Registered User
 
Join Date: Apr 2012
Location: Saint Paul, MN
Posts: 105
+1 on the headwind idea. While your "S" certainly has more grunt than my '99, I've noted a significant difference at the end of the main straight at BIR with a headwind v. tailwind. I can routinely hit 130 mph going into Turn 1 there with a tailwind, but *struggle* to hit 120 with any sort of headwind. Further, it feels like a struggle down the entire straight with a headwind, too.

The headwind idea would also explain why it would show up at the faster speeds (higher gear measurements), too, and not the slower speeds. Drag has a much larger effect at the higher speeds.

Last edited by pharaohjb; 03-09-2015 at 12:11 PM.
pharaohjb is offline   Reply With Quote
Old 03-09-2015, 12:31 PM   #5
Registered User
 
steved0x's Avatar
 
Join Date: Dec 2012
Location: FL
Posts: 3,878
Good points, I did get lucky to find a day with approximately the same outside air temp but who knows what the rest of the variables were like. I never even thought of those. I was just happy that the temp changed and it was 70 that second day.

I never thought about wind resistance at higher speeds before either but that makes sense too. I can definitely feel the wind resistance when I am driving my Toyota FJ Cruiser and I go to pass somebody on the interstate...

I know I can definitely feel it in the "butt dyno", but even on the runs where the "After UDP" was faster it is by fractions of a second... I can't imagine trying to "poor man dyno" a de-snorkel operation. (For the record I still have my snorkel)

Thanks

Steve
steved0x is offline   Reply With Quote
Old 03-09-2015, 12:33 PM   #6
Registered User
 
steved0x's Avatar
 
Join Date: Dec 2012
Location: FL
Posts: 3,878
Quote:
Originally Posted by JFP in PA View Post
Sorry, but this why dyno runs are done under controlled circumstances. From Dec to Feb, a lot of things (air temperature, air density, humidity, barometric pressure, wind direction, road surface, etc.) may have changed dramatically enough that you could have seen similar differences without any changes. The validity of experimental data depends upon controlling all of the possible variables other than the one under study.
I would have to say this (these) are probably the most major factors. I wasn't even on the same roads as I was trying some different roads to try and avoid traffic.

It was a fun exercise, until I got the part of the results where my 4th gear pull was slower, then I was really bummed with visions of potential engine failures robbing me of my horsepower....
steved0x is offline   Reply With Quote
Old 03-09-2015, 12:40 PM   #7
Registered User
 
Join Date: Feb 2005
Location: It's a kind of magic.....
Posts: 5,102
Quote:
Originally Posted by steved0x View Post
I would have to say this (these) are probably the most major factors. I wasn't even on the same roads as I was trying some different roads to try and avoid traffic.

It was a fun exercise, until I got the part of the results where my 4th gear pull was slower, then I was really bummed with visions of potential engine failures robbing me of my horsepower....
I completely agree it is a very interesting and creative experiment; I would like to see something like this repeated, but in a more controlled manner.
__________________
Anything really new is invented only in one’s youth. Later, one becomes more experienced, more famous – and more stupid.” - Albert Einstein
JFP in PA is offline   Reply With Quote
Old 03-09-2015, 12:45 PM   #8
Certified Boxster Addict
 
Join Date: Nov 2010
Location: Los Angeles
Posts: 7,429
I also worry about the use of linear interpolation on a data set (acceleration) that is decidedly non-linear - especially in 1st and 2nd gears.
__________________
1999 996 C2 - sold - bought back - sold for more
1997 Spec Boxster BSR #254
1979 911 SC
POC Licensed DE/TT Instructor
thstone is offline   Reply With Quote
Old 03-09-2015, 08:29 PM   #9
Registered User
 
Join Date: May 2011
Location: Miami florida
Posts: 1,591
Quote:
Originally Posted by thstone View Post
I also worry about the use of linear interpolation on a data set (acceleration) that is decidedly non-linear - especially in 1st and 2nd gears.
Exactly. Look at the published HP and torque curves and they are definitely non linear.

But i applaud the OP's thought experiment and attempt.
__________________
Current car

2000 Boxster 2.7l red/black

Previous cars

1973 Opel Manta
1969(?) Fiat 850 Convertible
1979 Lancia Beta Coupe
1981 Alfa Romeo GTV 6
1985 Alfa Romeo Graduate
1985 Porsche 944
1989 Porsche 944
1981 Triumph TR7
1989 (?) Alfa Romeo Milano
1993 Saab 9000

Last edited by san rensho; 03-09-2015 at 08:33 PM.
san rensho is offline   Reply With Quote
Old 03-10-2015, 05:42 AM   #10
Registered User
 
steved0x's Avatar
 
Join Date: Dec 2012
Location: FL
Posts: 3,878
Quote:
Originally Posted by thstone View Post
I also worry about the use of linear interpolation on a data set (acceleration) that is decidedly non-linear - especially in 1st and 2nd gears.
Quote:
Originally Posted by san rensho View Post
Exactly. Look at the published HP and torque curves and they are definitely non linear.

But i applaud the OP's thought experiment and attempt.
That was a hard thing to try and figure. On a couple of my pulls I recorded an RPM reading that was 3000, 4000, 5000, or 6000, and as an experiment, I would delete the elapsed time for that record and use one of my formulas to see what the calculated time would be for that RPM using the 2 surrounding records. I was able to get close, but in terms of the differences in the runs, it was not that close. Especially in 1st and 2nd where the RPMs increase by hundreds in between readings, it was way off.... I think I could tweak my formulas to extrapolate a better time but it exceeded my current math knowledge

If I could get more readings per second (maybe by disabling the graphed display of the data on the laptop using the durametric) there is a residential airpark near my house that has a paved runway with a crossing street at both ends. When I drive by it on the way to one of my favorite curvy roads I fantasize about turning in and making a "wrong" turn and making a high speed run down that runway... It would be good for making runs in both directions...
steved0x is offline   Reply With Quote
Old 03-10-2015, 05:48 AM   #11
Registered User
 
steved0x's Avatar
 
Join Date: Dec 2012
Location: FL
Posts: 3,878
Here is a thread with more tips on logging durametric that I wish I had read again before I did mine.

Durametric Logging 101: How to Log, What to Log, and What to do with it Afterwards - 6speedonline.com Forums

Disabling the virus scanner and disabling the graph are two of the ways they say that help log more data points per second. I would add, set your laptop on a high-performance power plan before starting.
steved0x is offline   Reply With Quote
Reply


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On
Trackbacks are On
Pingbacks are On
Refbacks are On




All times are GMT -8. The time now is 09:46 PM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Content Relevant URLs by vBSEO 3.6.0
Copyright 2018 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website -    DMCA Registered Agent Contact Page