Dyno Correction Factors

When you use the Dynojet software to graph dyno runs you get some options as to how you interpret the data.  Because atmospheric conditions (ambient air temperature and pressure) affect air density they therefore affect power output.  The influence of atmospheric variations needs to be accounted for in some way to make comparing dyno figures achieved under different conditions valid.  There are international standards set down by organisations such as the SAE (Society of Automotive Engineers), the European Community and whoever is responsible for the DIN correction system (I’m not sure who that is, TUV maybe?) for this.  Exactly how each standard does this is up to the body that puts it forward.

According to Edition 4 (1996) of the Bosch Automotive Handbook (an amazingly fact packed piece of literature) all the standards listed are the same save the DIN standard.  The DIN 70 020 system uses a different ambient air pressure and temperature to the other standards – they use 1013mBar and 20 degrees Celsius as the nominal ‘zero correction’ conditions, whereas the SAE J1349, ISO 1585, JIS D1001 and EWG 80/1269 all use 990mBar and 25 degrees Celsius as the nominal conditions.  The way the ambient pressure is measured is also different, with the DIN system using ‘absolute’ air pressure and the others using ‘dry’.  As far as I can remember, ‘dry’ is ‘absolute’ with the vapour pressure subtracted, which is how relative humidity is accounted for.  If I had a thermodynamics book on my bookshelf I’d be able to tell more, but for some reason I only have materials and mechanics/vibration books, which is odd given I majored in Fluid Mechanics and Thermodynamics.

The DIN system also uses a different equation, so it varies quite a bit.

The Dynojet software gives 4 correction options – SAE, STD, EEC and DIN.  I haven’t seen the STD or EEC correction formulas, but the EEC gives a very similar result to the SAE system which I usually use.  Of the 4, STD gives the highest readings by far with DIN not too far behind.  The table below shows power outputs according to the 4 systems and the uncorrected power as measured for one of my ST2 runs.

 RPM UNCORR SAE STD DIN EEC 2500 19.5 19.9 20.3 20.2 19.9 3500 39.1 39.8 40.6 40.3 39.8 4500 54.6 55.7 56.8 56.4 55.7 5500 71.8 73.2 74.7 74.1 73.2 6500 85.5 87.1 88.9 88.2 87.2 7500 92.7 94.5 96.5 95.7 94.6 8500 91.4 93.2 95.1 94.4 93.2 9500 83.5 85.2 86.9 86.2 85.2

If we didn’t have this correcting system comparing runs would be fairly worthless.  This in particular came to my attention when I showed a graph for my ST2 showing the influence of losing about 1kg of flywheel weight.  Although it made more power, it actually took longer to accelerate the roller.  This seemed rather strange to me, but I really couldn’t think of a reason for it at the time.  I didn’t consider the atmospheric conditions having an impact on the dyno runs, as they’re usually corrected.  However, it’s the power and torque that are corrected.  The acceleration time and air/fuel are recorded ‘as was’ so they can show something the corrected power curves don’t – the impact of the day.  When I realised this I felt a little silly for it not being obvious to me before.

The next few graphs show those ST2 runs again uncorrected and with SAE, STD and DIN corrections.  Because the ‘before light flywheel run’ was run on a day with conditions more suited to making power the correction factor for those runs is less that 1 – in this case 0.98 – meaning it is reduced.  For the ‘after’ runs the factor is greater than 1 – 1.02 - meaning it is increased.  This is because the after runs were made on a day that was hotter and had lower ambient air pressure.

So, if you’re comparing runs, keep in mind what the corrections are and what they’re correcting.  I might have to go back and edit that ”Acceleration time” report based on this, although that sounds like way too much effort.