znod

Thanks g-man:

I calculated one more set of weighted averages. I move the averages forward five torque entries at a time until I get to the last two--where I drop one entry and add another. These averages may show the phenomenon even more clearly. In this regard, no average is caculated across passes such that all were made at the same density altitude. Here they are without labeling.

238.8
239.1
238.9
239.0
239.3
239.5
239.8
240.8
240.7
240.9

Looking at these averages, I am not sure how we would come up with an appropriate weather correction. In this regard, note that my operative density altitudes cover a great deal of ground. As shown above, their range is 787.9 to 2151.3. The only way I see that we could come up with an adjustment is (a) to do something arbitrary (e.g., add or subtract 10% of actual altitude less density altitude to actual altitude) or (b) work with the density altitude tables. With respect to (b), let's say that we buy the argument that plus or minus 10 HP (of some type) equates to minus or plus .1 in the 1/4 on average for average-performance cars such as ours.

Next, let's say, for example, that a given density altitude (1800) implies a .1 decrease in my 1/4 time and it is approximately the same as the average of the density altitudes that produced the 239.0 torque reading above (i.e., interpreting the weighted-average torque values as weighted average HP values for simplicity). Then, I might adjust the density altitude (DA) to {1600 (my altitude) + [(239.0 - 238.8) / 10](1800 -1600) = 1602. For simplicity, conservatism, and because the change in DA would be so small, I might adjust as follows: {1600 + [(240.9 - 238.8) / 10](1800 -1600) = 1642. I am pretty sure that I am going the long way here since everything implied by the SMOKEmUp tables appears to be linear. I would have to check for sure, but I think the ET adjustments, ultimately implied by the two formulas, would reduce to -.1[(239.0 - 238.8) / 10] (= -.002) and -.1[(240.9 - 238.8) / 10] (= -.02), respectively. Neither of these adjustments would make a hill of beans (not to imply that beans aren't good ).

Thoughts, por favor? I am pretty convinced that we are being ultra conservative in the way we are reporting. And, they say that a foolish consistency is the hobgoblin of little minds. I think it's the same for a foolish conservatism.

grogan545

Well said Zman.I don't mean to oversimplify these calculations but something about the NHRA chart has bothered me.The differences between altitudes for both the et & speed corrections are the same for ea 100' interval(.0013/et,.0014/speed) regardless of the altitude number(1000' or 4000").I may be wrong but I would think that this diff should be nonlinear and be based on a sliding scale.If you take the extremes and multiply the diff(.0013,.0014) by the number of 100' increments(4000'=40 X .0013=.052+.9510(corr factor at 4000')you get a factor of 1.0030 at sea level for the et.The factor at sea level should be 1.000.You get similar results with the speed correction factor.As you get further from sea level the factors seem to overcorrect.The differences are not great(.3% & .4%) but they are significant.Even if these factors are corrected mathmaticaly we don't know if they represent the actual performance changes due to weather.

Having said all this it still doesn't tell us a better way to correct for weather.It also doesn't take into account that maybe the factors should be less for cars that have computer controlled management systems.I guess we could continue to test under various weather conditions and develop our own correction charts,but I don't think we want to do the amount of runs necessary to get good data.What do you think Zman.

znod

Good work g-man. I wasn't aware of the small problem with the NHRA altitude adjustment table, but, as you say, the error looks like it would not be at all significant. I'll look more closely at the table. And, I agree about your nonlinearity conclusion. Everything that I have been able to infer about the all the tables implies that they are based on a linearity assumption.

On the weather correction issue, I adjusted each of my 38 passes so that each allows for only 25% of the weather correction and calculated the resulting averages. The averages are given below. Ignoring a rounding error, the procedure I used yields exactly the same results that are produced when adjusting my averages for the 38 passes using a 25% weather correction applied to my average density altitude of 1360?. For example, a 25% weather correction based on the 1360? value yields an adjusted density altitude of 1540 = -.25(1600 ? 1360) + 1600. Using 1540, adjusting 13.727* (see * below) to sea level gives 13.493. Then, adjusting this value to 500? yields 13.564 = approximately 13.562** (see ** below).

...........................1...................... .......2.......................................... ....3.........................................4
..................Unadjusted.....Weather/Altitude to 500'.....25% Weather/Altitude to 500'....Altitude to 500'
1/4 Time.........13.727*..................13.592...... ...............................13.562**........... .................13.556
1/4 Speed.....102.878..................104.046........ ............................104.351............... ..............104.134

Taking the altitude effect of .171 (13.727 - 13.556) out of 2 and 3 gives:

2 = 13.592 + .171 = 13.763 adjusted to 1600' given standard conditions or
3 = 13.562 + .171 = 13.733 adjusted to 1600' given standard conditions

Now, the question is whether the weather conditions, on average, were better than standard so that:

a. they imply a .036 increase ? time at 1600' assuming standard conditions (i.e., 13.763 - 13.727) or
b. they imply a .006 increase in ? time at 1600' assuming standard conditions (i.e., 13.733 - 13.727)?

The more reasonable answer depends on how my car?s HP is affected by weather over the 1360? (my average density altitude) to 1600? (my actual altitude) range of altitudes. If the answer is .036, then, crudely, my car?s performance should be handicapped, in effect, by the absence .036(HP) = .036(325 HP) = 11.70 HP [(= RWHP of 9.36 = .8(11.70 HP)]. And, if the more reasonable answer is .006, then, crudely, my car?s performance should be handicapped, in effect, by the absence of .006(HP) = .006(325 HP) = 1.95 HP [(= RWHP of 1.56 = .8(1.95 HP)].

The values below show my car?s Pro RR-measured average RWHP for each of the given density altitudes (DA). The range of these values approximates the range mentioned .above.

..............Average
...DA........RWHP
1695.0.....243.3
1554.4.....239.3
1516.0.....244.8
1509.3.....240.0
1415.4.....248.0
1239.0.....244.1

The average of the RWHP values associated with the density altitudes averaging close to 1360? [(1415.4 + 1239.0) / 2 = 1327.2?] is 246.05, while the average of the RWHP values associated with the density altitudes averaging close to 1600? [(1695 + 1554.4 + 1516.0 + 1509.3) / 4 = 1568.7] is 241.85. The difference between the two averages is 4.2?which is closer to 1.56 than to 9.36 (i.e., 4.2 ? 1.56 = 2.64, while 9.36 ? 4.2 = 5.16).

The above observations suggest that the more reasonable answer to the question raised above is .006. In turn, this conclusion suggests that adjusting for 25% of a density altitude adjustment is more reasonable than adjusting for 100% of the adjustment.

From an overall perspective, what seems to be implied, for my car, is finding the percentage of the 1360? density-altitude adjustment that yields a RWHP decrease of 4.2 RWHP. Trial and error reveals that 42% yields a decrease of 4.28 RWHP?which actually is as close as one can get using the tables. The density altitude yielding this decrease is 1499.2 = -.42(1600 -1360) + 1600.

I believe that the above procedure is appropriate for use on my car. But, I am uncertain about whether it applies to the cars of others running especially at greatly differening altitudes. On the other hand, I would bet that using my procedure on other cars, etc., at least would improve accuracy. Also, I would bet that the use of my procedure could not hurt accuracy appreciably since the average results produced should differ very little from what 100% adjustment yields--i.e., my 40%-adjustment values compared to my 100%-adjustment values for the 1/4 are as follows.

........................100%........42%
1/4 Time..........13.592.....13.572 (difference .02)
1/4 Speed......104.046....104.077 (difference .31)

Thus, the advantage of my procedure does not subsist in it effects of averages, but rather in its effects on the identification of superior passes. Using the 25% adjustment caused some rearrangement, and minor changes in times/speeds, for my superior passes. I am not yet sure if the 40% adjustment will have the same effects.

Ultra conservatism is implied by using 100% adjustment. However, as shown above, the effect of doing so is not likely to be significant.

znod

I finally finished the above analysis. Any thoughts? All appreciated.

grogan545

Zman,you have my head spinning with all of the data & calculations but the numbers you came up with make more sense than just using the NHRA chart.I know that many readers are wondering why we go to such extremes to fine tune our data,but it is the only way to compare data obtained under varying conditions.It is also the only way to compare cars located in different locations of the world.

Until now I have been ignoring the ave HP numbers from my GT2.I will start recording them along with my other data for comparison.I noticed in your RWHP data that the density altitude/rwhp numbers did not necessarily follow the norm.Sometimes the rwhp numbers were higher even though the density altitude was higher.There was not a big difference,but it shows that other factors influence hp other the altitude.

znod

yes, all of what you say is true. I also noticed your observation about the "norm." That phenomenon made me wonder if adjusting for weather makes sense at all for our cars. But, being somewhat, but not ultra, conservative is a good idea. What do you think about, at least me, using the 42% weather adjustment? Doing so would at least pay homage to the idea that our cars are not "average" when it comes to the effect of weather on performance. The charts have to be based, in principle, on the averge car assuming no compensation for weather.

grogan545

The 42% adjustment sounds fine to me Zman.Like I said previously some correction is necessary,but I don't think the NHRA chart is accurate for our cars.

znod

Right, g-man. My overall results will barely be affected. I'll let you know if there are any surprises related to individual passes.

big_ipaq

Does anyone has any experience with Beltronics FX2? It seems I've found a place where to buy this one locally...

znod

Sorry bigi. I have no info at all on the FX2. I know that Escort and Tesla will ship to Europe though.