View Single Post
  #1 (permalink)  
Old 09-11-2007, 11:56 AM
bwilson4web's Avatar
bwilson4web bwilson4web is offline
Engineering first
 
Real Name: Bob
Location: Huntsville, AL
Hybrids: Prius Classic 03
Posts: 4,902
Unhappy Engineer vents -- methodology

Hi folks,

Recently someone posted a rule of thumb for expected MPG vs miles per hour for a Honda Civic Hybrid and it 'set me off' about the absence of MPG vs miles per hour charts for all of our hybrids. Fundamental to all science is to have a benchmark, a baseline that establishes exactly how something works. This leads to insights when the measured values don't match theory and allows us to experiment and know if we are seeing a real change or some artifact. It is how engineering works and what attracts me to GreenHybrid.com.

What gauls me is the apparent lack of standard day (or any day) charts of MPG vs miles per hour for all hybrids like I do for the NHW11 Prius (see below.) With these charts, I've found notches where there are local peaks and valleys. I've used it to quantify:
  1. tire pressure effects
  2. oil fill level effects
  3. 42 miles per hour transition
  4. bump in MPG from 55-70 miles per hour
  5. temperature effects
It is incomprehensible to do any experimentation without having a baseline to compare and contrast with experimental data. Yet this very simple exercise, collecting and plotting the data, has yet to be done for other hybrids (unless someone has a link.) We are stuck with at best, two point, EPA curves, which are not curves.

Empirical knowledge, the grist of engineering and science, takes more. It isn't that hard to baseline a system but the benefits for subsequent experiments far and away outweighs the cost. The alternative are anecdotal reports and 'one time' experiments without a control and these are the feedstock of TV infomercials and advertising.

The GreenHybrid.com database is better than not having a database. We have gained insights about vehicles and within each model, temperature and trip duration effects on mileage. In the case of the Prius, we've also gained insights about the variable capacity fuel tank. What is missing is a simple way to map miles per hour vs MPG for each model hybrid.

Engineering teaches that the experimental methodology is critical to avoiding serious errors and mistakes. It is what lets us quantify effects and make plans that have a chance of success. So that is why I remain frustrated by the absence of MPG vs miles per hour charts for other hybrids. A 'rule of thumb' is better than none but begs the question of why not do the job right and actually measure MPG vs MPH and plot out the data.

All it takes is to get on the highway and when the temperature is 65-75F (or close enough) on a low wind day (or close enough) and just set the cruise control at one speed for an hour or so and then read out the MPG. That is all it takes to get one data point in an MPG vs miles per hour curve. When two or more observations like these give the same value, we have credible data and gain insight about how our various vehicles perform.

I just get so frustrated because short of buying one-each of every hybrid that exists and 'doing the experiment,' I don't see any effort to baseline our vehicles. The one HCH report was better than none but my frustration is how much better would a full MPG vs miles per hour curve. Having such data is so much better than just anecdotal 'buzz'.

Bob Wilson

ps. All of these charts used standard instruments found in every hybrid and in some cases, just gas receipts and trip meter mileage. All it takes is the patience to record the data and plot it on a graph:

First MPG vs miles per hour Chart


Improved Chart With Theory Drag

Hill Climbing Chart



.

Operation Iraqi Oil Freedom:

Automatic, stock, project car.

My
other 1500 cc car:

Automatic, stock, backup car.
Free speech, dialog and knowledge thrives without the poison of SPAM.
Reply With Quote