curious, but what correction factor was used on the dyno, and what altitude was this at? our shop is at 4,700 feet and our dyno reads a little high on boosted cars when using sae or stp correction factors.

I made this writeup on another forum, points out some interesting findings

As you all know the altitude of Idaho is rather high (about 4,700 feet in Idaho Falls) and detrimental to performance of internal combustion engines. The question has been raised several times at our dyno shop regarding a cars HP number being higher than expected. I thought I would take some time and share our findings.

All dynamometers use one form of correction or another so the numbers recorded match what the car would do at sea level on a 77* day. Depending on humidity, temperature and barometric pressure of a particular day, the correction factor in Idaho Falls (we use SAE) can be anywhere from 1.17 to 1.22, meaning the measured wheel power number is multiplied by the correction factor and that is what is displayed on a readout. On a naturally aspirated car the correction factor works perfectly. To demonstrate we took a bone stock 2011 Infinity G37x (N/A) and placed it on our dyno. The engines are factory rated at 328hp (at sea level) so one would expect a wheel power number to be somewhere in the 280 range. From what I have found, this is what most people report their bone stock cars will do, and with a correction factor of 1.2 our Infinity laid down a run of 280whp just as expected.

Things get a more complicated when you have a boosted engine however. Everyone knows a forced induction engine has smaller losses at altitude. Think of piston engine airplanes n/a vs. forced induction. In the First World War (I think, my history may be off) airplane manufacturers started playing with forced induction to compensate for the power losses and altitude ceilings of a naturally aspirated airplane. It’s been proven many times that forced induction makes more power at altitude.

We continued our experimentation with several boosted vehicles, diesel and gasoline, all bone stock. First up is a 2008 ford F-350 6.4 turbo diesel in 100% factory trim. Stock the 6.4 powerstroke is rated at 350hp and 650tq (at the crank of course, sea level.) On a chassis dyno one might expect numbers to be in the 300 range, and that is what many owners report (at low altitude). But what happened with ours? Uncorrected numbers come in at about 280hp which is pretty close to what it should be. But when we apply the 1.2 correction factor the number jumps to 336 which would translate into roughly 380 crank hp, which is 30 more hp than stock… What? Example #2, a bone stock 2005 WRX STI: Factory rated at 300hp (again crank power at sea level) one would expect a chassis number of around 240 due to awd losses. What happens when we dyno a bone stock example at Idaho Falls altitude? Un-corrected we get a number of 220 and SAE corrected we get 270…. Wait a minute you might think, that’s way high. And you would be correct. But if you take only a 1.08 correction factor you actually get 237 which is believable for a stock vehicle. We have tested several other vehicles and found the same results.

Here’s the thing: I do not know exactly what correction factor would yield most accurate results (it varies with boost pressure, higher boost=less loss) but this is what I have found. If you take about half of the correction factor of the day you will get pretty close to believable number.

So why do we use SAE corrections on our dyno? Well mostly due to customer requests. We have people say “all other shops use SAE” or “My car dyno’d XXX hp on this other shop’s dyno, and they used SAE.” I personally do not think the SAE numbers on turbo or supercharged engines are accurate at this altitude, I think they are artificially inflated by about 10% to 12%. But what do I do? People want bragging rights, and a higher number gives them that, but I try to explain to them what the numbers really mean. I just wanted to start a discussion to let the community know our thoughts and findings. Hopefully I don't burst to many people's bubble lol.