Test data for a few mods

[saand]

Hi All,
I did a test on a bunch of mods I have done. I did the tests in one go expecting that it would be more efficient. Unfortunately looks like too many of the trips were influenced by the very few passing cars as the baseline tests are very variable.
The results do give a bit of an idea that the modifications appear to do something for efficiency however the levels of improvement aren't too reliable due to the high variation of the baseline tests

Either way the data is below if anyone is interested

So the test method was as follows
Drive on a freeway at 90 kph (56 mph) with cruise control on, going one direction then the reverse direction. At a defined location I turn on my fuel usage meter data logging and I turn it off at another defined location. I then make a change and repeat the run both east and west and end up at the start location. Keeping the engine running while im doing the change so that the engine stays warm.

The sequence of runs were
- Baseline standard configuration 1
- Turned the headlights to parkers (to see how much the electrical load affects efficiency)
- Tyres up to 45 PSI from 30 PSI on all other tests
- Baseline standard configuration 2
- Reattach my normally removed passenger side windscreen wiper
- turn the radio on which puts the antenna up in the air
- reattach both side mirrors
- Put flat wheel hub covers on
- Baseline standard configuration 3


Results are as follows
Numbers are in litters for the run
East West
Baseline 1 0.646073067 0.708983172
Head lights off 0.637612481 0.726116467
tyres 45 psi 0.646266763 0.710757363
baseline 2 0.656333128 0.722960581
wiper 0.659809485 0.726815202
antenna 0.660399757 0.732200836
mirrors 0.656161305 0.732990613
wheel cover 0.648702133 0.696455623
baseline 3 0.657330163 0.708763732


Processing this data I averaged the 3 baselines then got the percentage change, results are as follows
Baseline 1 -0.87%
Head lights off -0.32%
tyres 45 psi -0.73%
baseline 2 +0.89%
wiper +1.43%
antenna +1.85%
mirrors +1.58%
wheel cover -1.55%
baseline 3 -0.02%

So what can you get from this data.
- All of the mods work as expected, they do all improve efficiency.
- If you ever do tests, make sure you can’t see another car on the road if any cars get in front of you then stop the test and restart

So if I have enough time one night I may retry the tests, it did take about 5 hours so im not too keen to do it again too soon but maybe I’ll get bored one night.

[Piwoslaw]

Thanks for spending time on testing. It really is a bummer that it takes so long, but once it's done it's well worth it

Originally Posted by saand

So if I have enough time one night I may retry the tests, it did take about 5 hours so im not too keen to do it again too soon but maybe I’ll get bored one night.

So one of the things you tested at night was "Head lights off -0.32%"? I'm glad noone else was testing that at the same time on that road... Just kidding

[saand]

My pleasure Piwoslaw, happy to do testing as long as i get some good data for the group so it might help newer people to decide on what mod to try.

I was waiting for someone to catch that i was driving fairly fast with no headlights on.
I still had parkers on so wasn't too bad, It was a little concerning when the stretch of freeway didn't have street lighting, was a bit hard to pick out the lanes.
Im definitely not suggesting people should drive without headlights in the dark though. I was trying to characterize the effect of electrical drain on the drive train. Hoping to eventually get a value that will say you will loose X % efficiency if you take out X amount of watts from your alternator.

I did start out the tests thinking id like to get some good data for the wiki but in doing the tests I have found out some interesting things that will now get me doing some different modifications so I figure it is worthwhile even if the data was not too good. At the end of the day it only cost me about $15 in petrol and a bit of my time so not too much of a waste for good information.

Well I just have to find a location that has very few cars around at night and is a nice long stretch of road and no traffic lights. So im really looking for a freeway that isn't used much. Maybe ill find one eventually and ill get some reliable test data.

[RobertSmalls]

It would be nice to see multiple runs for a few of the mods, to establish the resolution of your testing methodology. We need to know how much noise is included in the data.

Baseline 1 -0.87%
baseline 2 +0.89%
baseline 3 -0.02%

If all of these baselines represent the same configuration, then unfortunately most of your other trends fit within the "noise", and no conclusion can be drawn about whether they work.

Statistical significance - Wikipedia, the free encyclopedia

I know that additional test runs cut in to your budget of testing time, which means you can't test as many mods. But it's better to be sure of a few mods than to have a hunch (or worse: bad data!) about a large number of mods.

[saand]

RobertSmalls i agree completely, that is why i initially wrote that it is not reliable data. Appears that there was too much noise in the measurements as you have indicated the baselines were the same configuration.
I did east and west tests for the multiple runs for a given modification expecting that there would be little noise and the 2 runs for the 1 modification would indicate that the data was reliable.

I believe the variation has to do with the very few cars that were passing every now and then.
It was tempting not to post the results but maybe someone might get something out of it or inspire someone to test one of these configurations themselves.

I am hoping to follow up with some reliable data when i get time to retest.