PERFORMANCE COMMENTS: Dyno Testing Issues.
This page will be added to and updated over the next several months.
I've always considered dyno tests (as well as any other forms of testing) one of the more interesting parts of a magazine. As a "techie" sort of guy as well as someone who actually runs tests, I'm always interested in the effects of design change. However, as the years have gone by, as I have learned more about header design as well as other areas of performance improvement, I find myself less and less impressed by the dyno tests provided in magazines. Now, I often find myself more disappointed, disillusioned, and even disgusted with what is being said. Instead of reading a well thought out design discussion that is accompanied by well executed backup tests, I often feel like I'm reading information provided more to motivate somebody to spend money, incomplete discussions that left out information that really needed to be known or created more questions in my mind than gave me answers, discussions that, to me, were "tilted" or biased rather than neutral in position and sometimes, in my mind, even determined or affected the tests' outcome, comments that obscured truths, procedures that protected a product that had one or more design flaws, opinions that could be as misleading to some as they might be helpful to others, etc. The following discussion is not based on any one immediate situation but on hundreds that I've been witnessing for decades. This following discussion is also going to disturb many who read it so you are being warned, up front, about that possibility.
Problems with dyno testing.
There are no industry standards on dyno testing procedures to assure consistency of results that allow accurate comparison. While the data from sensor readings taken during a dyno test are usually CORRECTED to a specific temperature and barometric pressure to make comparisons easier, the manner in which the test(s) is/are run can vary considerable from one dyno cell to another, from one dyno operator to another, even reasons the why test(s) were run in the first place, etc. - can create enough difference in outcome that just comparing numbers from different tests might end up being very misleading.
For example, while the best way of running a dyno test might be with open collectors (assuming the engine is going to be used that way), the reality many shops have to face is that they are in or near residential areas so dyno testing has to be conducted with some sort of exhaust system. Now one must be concerned not only about how the basic exhaust systems differ from shop to shop (what mufflers are used as well as how many), how different in size the flexible tubing BETWEEN the headers and the rest of dyno cell's exhaust system might be, but also HOW different the connection between the headers and the rest of the dyno cell's exhaust system might be. (These differences might not seem important to many but I remember arguments between some fellows in this area about "so and so's" dyno reading 25 HP higher than some others' dynos and that, to some, meant he was using a different CORRECTION (altitude and barometric pressure) factor to make his engines look like they were making more horsepower. When I pointed out to them that I thought "so and so's" exhaust system may be better thought out than others, these fellows said that they had never thought of that possibility. So, in effect, they were accusing him of cheating when that may not have been the case at all.)
Notes to us for future comments:
Dyno results are factored (recalculated) to a certain altitude and temperature and, in the process, generate what can be called misleading data.
Talk about a really good exhaust hookup to improve testing consistency.
Dyno racing?! (NOT racing in vehicles?)
Dyno testing done predominantly with bad/incorrect header designs.
Dyno testing done at full throttle. Street driving is NOT done at full throttle - so, do dyno tests generate less than useful (misleading) information relative to street use?
Dyno tests often start at rpms ABOVE rpms engines commonly run at on the street - does this mean, again, that the tests results of a dyno test are misleading when applied to use on the street?
Minimal or NO tuning being done on dyno tests run in magazines. Depending on various designs and/or design errors and their effect on tuning, is it possible to run a header test and have the worst header look the best because of tuning "errors"?
Explain that at one point we were doing cylinder head work and were planning on expanding into doing engine work and, part of that expansion, was to include a dyno cell so we were "studying" dynos for a number of years - possibly in preparation of buying one.
Show headers we've made for dyno testing that address bad hookups to dyno exhaust systems.
Also: E-mail us with suggestions you would like us to comment on?
Dyno Testing Issues