Horsepower Dyno Readings, USA compared to AUS, et. al.

This came up in a forum discussion the other day. Now I know that all dyno numbers are rather subjective based on a multitude of different factors and ultimately the numbers should be less a focus and we need to look at gains/losses pertaining to tuning changes et cetera.

Nevertheless, let's explore. In the thread, dyno numbers were disputed based on the idea that a certain engine with certain upgrades couldn't achieve the claimed wheel horsepower. The discussion included members from the USA, Great Britain, and Australia. As it progressed, the apparent resolve was that the engine combination could in fact make the claimed horsepower based on dyno readings in the USA, but it's universally known that dynos in the USA are calibrated to read an average of 20% higher than they are in Australia and some other countries. Now, to be clear, we were not discussing the conversion factor from kW to HP, nor were we talking about different brands or types of chassis dynos. We were talking HP in the USA to HP in Australia and others.

So I ask here, is it an accepted fact that chassis dynos in the USA are calibrated to read 20% higher than the chassis dynos in other countries, namely Australia (since that's who was making the claim)?

I don't think you can say that this is Australia vs USA. It's really down to the particular dyno. With that in mind the most common dynos in Australia tend to be Mainline and DD which conveniently seem to read somewhat similar. In the US the dominant dyno if you want big numbers is the Dynojet. I agree 100% that there is a massive difference between a Mainline and Dynojet number. I put it between 15-18% myself but that's a moving target and in some instances could easily be 20% or even more.

Before our Mainline dyno I owned a Dynapack hub dyno and I'd previously tested and found that this read almost identical to a Dynojet. No surprise that our Mainline reads much lower and this took a bit to get used to. The key point is that dyno figures are (at least in my opinion) relatively meaningless without some context. You can dyno the same car on 5 dynos and get 5 different numbers. The important part is to look at where you started and where you ended for a tuning session or part upgrade, or alternatively what a similar combination is known to produce on the dyno you're using.

Thanks for the reply, and like stated, I agree completely that the numbers themselves aren't what's important.

Yes, the Dynojet is very popular in the USA and a high reading dyno at that. That being said, the Mustang dyno is also very popular here in the states and is dubbed the "Heartbreaker" because of it's low readings which seem to be in line with the Mainline readings. Where the heated discussion came in was when it was revealed that claimed numbers were in fact achieved on the low reading Mustang dyno.

But based on what you said, it's more clear where they were coming from so I appreciate you providing some context.

A curly one - as Andre said, about the only thing you can be assured of is when you use the same dyno', under the same test conditions, to check changes.

While one can say power is power and a horsepower has a strictly defined value, so one should be able to compare like with like, it isn't that simple.

How often the dyno' is calibrated for accuracy, the speed/rpm increment rate and gear used, heat soak from repeated runs, rotational inertia from diferent wheels and tyres, tyre pressures and tread type, etc, will all play a part in the final figures given.

Even 'pure' engine dyno's are controversial as some will run an engine with no air filter, open exhausts - even using dyno headers instead of those to be used, electric water pumps when mechanical will be used in vehicle, no ancilliaries, etc. whereas others will have as close to installed as practical... Think SAE vs DIN. This is one of my personal bitches as some 'tuners' will spend a lot of time doing the first to get it 'right' when installed will be quite different. That is without taking into account fuels that may be different and accelerating rates or steady speed - the latter will be best and some 'tuners' were notorious for 'flash' readings where they effectively stall the engine with sudden loadings, but I hope that is a think of the past. Then there is the use, and abuse, of 'correction' factors.

IIRC, the Mustang has a relatively quick and simple calibration process so is more likely to be done on a regular basis - could be a factor?

Oh, forgot - since most dyno's used in the the UK and Australia would also seem to be US made, they may need to be taken out of the equation and perhaps the focus put on some people being a little less than honest?

OP can you tell us what engine, vehicle, and mods you are debating? The others in this thread have already brought up the points about Dynojet vs others, and the correction factors. Boosted engines add another dimension to it.

Another thing to keep in mind is altitude. It's common for example to see some absurd "corrected" power numbers in high altitude areas. For example, Subarus in Colorado (USA, thousands of kilometers in altitude). If you read the actual SAE J1349 document (I've read it many times), any correction over 3% is consisted questionable. A manufacturer isn't allowed to certify a given power or torque number if the correction factor is greater than 3%.

When a Formula Atlantic engine got built and dynoed in the UK they would make 218-220BHP. When they got shipped to the US they gained 20BHP somewhere on the trip over. Go figure.

The debate actually centered around the Subaru platform. It was alleged that a stock long block Subaru EJ engine could not make 400whp, no matter what mods with an emphasis on "EVER". The particular vehicle in question was a 2005 Legacy GT running a stock location GT30 family style turbo and supporting mods/fueling, 24psi tapering to 21psi to redline running E85 fuel. Altitude of about 1000ft ASL. I've seen plenty go beyond. In fact, a personal friend just put his on a Mustang dyno a few weeks back. Similar mods and boost, 2008 WRX, stock long block except for head studs, and he hit 422whp.

While too much emphasis is often put on dyno numbers instead of what a dyno's main purpose is (tuning), I still like to see the figures reported somewhat in the realms of plausible. For instance, some years back, a debate was sparked that went something like this. A low reading dyno (can't remember what kind) was used to monitor a build. Some baseline ramp runs were put down in the range of 130whp. Now mind you, this was on a bone stock Subaru rated at 250hp from the factory. That's what sparked the debate especially since it was claimed that they've had many of the same platform put down similar numbers in bone stock trim. To me, that just sounded low since they are claiming nearly a 50% drivetrain loss when it's generally accepted that manual transmission Subaru's experience about a 15-20% drivetrain loss. In stock form, those 250hp Subarus put down about 210-220 on the high reading dynojet, but that's at least in the ballpark of where it should be. Nevertheless, I don't think anyone was being dishonest because the HP GAINS they reported were in line with the particular modifications they made to the vehicle. The final numbers did seem low, but nobody was disputing the gains.

I'm a numbers guy so when the numbers don't make sense, it bothers me. Likewise, there's a notorious, big name tuner I'm familiar with that has done some local vehicles (not naming names), and his dyno numbers always seem inflated. Nobody dares dispute those numbers because of this guy's reputation, but certain turbos, certain boost, and some reported numbers just seems erroneous to me. I won't get into the the specifics. In some instances, those engines haven't lasted long, blew on the dyno, or shortly after being tuned (I can think of 3 in my small circle of acquaintances). Never have they questioned the tuner. They always chalk it up to certain mechanical failures. Maybe his numbers are accurate, but stretching engine components beyond their limits is rather counter-productive and needs closer attention. It's for reasons like this that I'm here learning for myself and hoping one day becoming proficient enough, that I can provide this service to my local community

At this point you would have to check if the dyno itself has been calibrated and maintained. I'll give you an example. OEM grade chassis dynos are meant for fuel economy and emissions certification. They have to follow a rigid sschedule in order to be valid for certification with governments. For aftermarket performance tuning, anything goes in terms of accuracy and maintenance.

As for whether an EJ255 can make 400ish horsepower, well if you're stuck with stock heads and cams you need a lot of boost, low restriction intake and exhaust, and high octane fuel like E85. However that's going to stress components due to high cylinder pressures. So making that kind of power and then breaking something soon afterwards doesn't surprise me.

One thing to consoder is operator error, either intentional or not. I went to another shop for a dyno day and on their Mustang AWD dyno a stock 1990 miata made 123WHP. It was suspicious as OEM they are rated at 116HP crank and since the car is over 25 years old I'd figure it would lose power. Now on my Mainline 2wd600L a stock NA Miata makes about 87-92WHP. I have attached a link to video made by AMS which shows how easily a Mustang dyno can be manipulated. On most dynos you can see the Correction factor to make sure it was in line as what Arghx7 Said. But in mustangs you can play with the calibration settings that can be used to provide some very optimistic numbers.

Here is the link:

https://youtu.be/sDwjfZvmPHg

While I understand this is a 6 year old thread, something recently came up in a conversation I was having with DynoTech about some of the settings in our dyno.

Our DynoDynamics single retarder dyno has a power readout setting called "comparison mode" and when I asked the rep from DynoTech what that was for, he said "Oh, it's so you can show power numbers that are calibrated to US standards, so you can share dyno numbers with people in America"

From my testing/calculations - it's ~10% higher than my normal "power at the wheels" measurement

Just thought that was worth adding, even though this thread is old as hell :)

Pretty much matches my experience.

Our horses are smaller, so it takes more of them to make the same power.

That's why they call them pony cars.

It's one of my favourite topics) Just can't get my head around the situation when universal physic lows aren't producing same numbers because of different measurement systems. We wouldn't accept the same approach in electricity, for instance, right? You have X voltage Y resistance but come up with different current result depending on measurement tool although Omh's low is the same) It would be very strange if we accepted that , wouldn't it?) Yet we perfectly accept it in automotive world despite the physical lows backing up only one correct figure...

As of US dyno numbers being often 10-20 percent higher I believe it has a lot to do with Dynojet hystory. They became official dyno for NASCAR and Daytona races and that made them quite famous in US. As far as I remember when the first Dynojet was built they could not properly measure horsepower of one motorbike that was somehow claimed to have 100 something HP (I don't remember the exact fugure) - their pulls were coming out always short on power. Rather than properly dealing with that issue they decided to change the software to match the claimed number. And since then that factor is inbuilt in their system resulting in higher dyno numbers. When I ask Dynojet numbers people why their HP numbers are usually so high comparing to Mustang and aren't supported by air mass consumption they usually respond that Mustang dyno can also be manipulated- I just love it))

It may, in part, be down to Europe using DIN (dyno' installed net) which is with the full intake, exhaust and ancilliaries fitted, whereas the US uses SAE, which is with no air filter, exhaust, or ancilliaries - often even the water pump is electrically driven rather than by the engine.

This can make a considerable difference to the "correction" factors, and horsepower numbers. There may also be a difference in the air temperature, pressure, etc, it's "corrected" to.

Personally, I'd say it's the power/torque to the wheels that is the important thing, under the conditions it's being run in.

It's worth noting that Mustang dyno numbers have changed over the years depending on software version. Early/mid 2000s Mustang software read lower than software for the updated PCI card setups of the late 2000s, early 2010s. The updated PCI units and newer Powerdyne software needed about 12-13% added to get to Dynojet numbers when I back to back tested in 2012, and confirmed again around 2020 on 2 different pairs of dynos.

I've not compared Mustang with Powerdyne software to Mustang with Holeshot software, but I am curious if numbers are different.

Also keep in mind some use math channels to provide Dynojet numbers on a Mustang so customers have something they're more accustomed to, or in an R&D setting I've done it because a business replaced a Dynojet with a Mustang and needed a way to loosely compare to old runs. This can be done in a way that doesn't skew raw data recording, so native Mustang dyno numbers are saved and can be viewed as well.

I feel as long as the operator is clear about what's being shown, this is a reasonable approach.

On low HP cars especially, an apparent percent difference between dynos often comes down to parasitic loss not being measured properly. For example someone above mentioned a miata making 120+ HP on a Mustang dyno. Perhaps their correction factor is high, but often it just means the did the parasitic loss test with a vehicle that has higher parasitic losses. That test involves accelerating the dyno up to a speed a stock Miata can't achieve, so perhaps they did it with a big v8 vehicle.