Mods can we perhaps sticky this?
Preface:
Ok I'm going to supply some information based upon my 13+years of working with chassis dynos. I'm going to keep this as factual as possible w/ as little bias as possible, I invite any of the other dyno reps to posts or shop owners who have a dyno.

First off I will state my current position. I am the southeastern USA sales rep for Dyno Dynamics. In the past I have worked w/ dyno from various mfgs. All of these dynos I have had extensive experience with. Dyno Dynamics, Dynojet, Superflow, and Mustang. I have working knowledge of Roto-Pak, Dyna-Pak, Bosch, and Maha.

This is key, an number is just that, a number.
It doesn't matter wether a dyno reads 400hp on 1 mfg's dyno or 3 apples, 2 oranges, and a kumquat on some other mfg's dyno. A dyno is a tuning tool. The purpose of a dyno is to simulate real world conditions to help tuners. What is important is to see how change made to the vehicle affect it's power output. So your tuner makes a change and now you make 405hp or 3 apples, 2 oranges, and 2 kumquats.
A gain is a gain, a loss is a loss, period.
Numbers are only good for pissing matches and bench racing.

Now on to the comparison:
Here is how the various mfg dynos relate to one another. This is just a comparison of numbers output assuming the same vehicle on the dyno.

The industry leader in number output by far is Dynojet. Their marketing in the mid-90's specifically used the larger output numbers as a selling item.

Next is Mustang, generally about 7% less than Dynojet numbers.

Superflow is about 3% less than Mustang.

Dyna-Pak and Dyno Dynamics are about 3% less than Superflow.

Ok lets make this point again to make sure it's clear. The SAME CAR on the DIFFERENT DYNOS. So our test vehicle output on the various dynos are as follows:

Dyno Dynamics 200hp
Superflow 206hp
Mustang 214hp
Dynojet 226hp

This is assuming the dyno operator has not changed any of the parameters of the dyno. All of the dynos software incorporates parameters that are operator definable. These adjust the displayed power output. In other words the operator can "adjust" the output of the dyno to display any number desired. By adjusting weather station parameters, external corrections factors, inertia compensation, or any one of a dozen different factors the output number can be modified. The number can be modified to display anything from 10% of measure power to 300%. So our 200hp example vehicle can after a pull display peak power of 20hp or 600hp w/ just a couple of keystrokes.

Key Fact: Just because the dyno spits out a number, it doesn't means squat without knowing ALL of the correction factors.

Some dyno operators adjust the output of their dynos to closely match the highest numbers in the industry, Dynojet. This is not an attempt to defraud the customer. It is just that customers have a false notion that a bigger number means more power. But as you have seen, numbers mean nothing.

If anyone has any questions about dyno operations or which type of dyno is better or worse please start another thread & I will be more than happy to answer any questions.

 

Awesome information. Thank you.

 

The Mustang dyno I dyno on....the operator [ 25 years ] Says 10 to12 % lower numbers than a Dynojet . But your right...Operator can spit out any number he wants

 

TTT
So, mods is this sticky worthy?

 

I've always noticed that the DD have always read lower than the Dynapacks.

 

The DynaPack dyno I work with reads higher than DynoJets w. a 10s ramp-up load setting.

 

I think that Church's Automotive is largely responsible for giving the DynaPack a reputation around here of being inflated. In fact, my local tuner owns a DynaPack and does not frequent this board yet when I told him that some people were seeing some pretty high numbers on the DynaPack he said "is there any chance those results were from a place called Church's Automotive in California?" Imagine my surprise!! I guess their reputation for being, uh, HP friendly has made it all the way across the continent.

It just goes to show (and to the point of the original post) that dyno numbers are really only helpful for measuring increases and decreases relative to the same car on the same dyno.

--Steve

 

That may be true for most rigs, but not necessary for others. In the case of the DynaPack dynamometer, it is entirely possible to reproduce results that are within the machine’s built-in margin of error time after time and machine to machine. The single largest variable is the ‘load’ setting for which I specifically pointed out to be a 10-second ramp up (I personally used). As such, the result will read higher than a standard DynoJet configuration. I will explain why the 10-seconds setting in a bit.


The reputation of Church Automotive Testing is more readily appreicated among the Honda/Acura camp. It did not arrive as the result of the ‘inflated’ numbers. Instead, it was the result of many things that had transpired over many years. Shawn Church, who founded one of the oldest car enthusiast site on the internet (Temple of VTEC, www.vtec.net) has a strong presence on various technical forums and his willingness to help and educate others shines. He's managed good relationships with Hondata, DynaPack as well as North America Honda and Acura as the result.

Jim Wolf (of Jim Wolf Technology), whom holds more water in the Nissan camp, also owns and operates a DynaPack for very good reasons. He uses an 8-second ramp-up load setting and also achieves results that are higher than the typical DynoJet figures.

Ramp up time setting of between 8-10 seconds is ideal for various reasons. It is just long enough to minimize drivetrain inertial loss, yet short enough to keep heat soaking factors to a minimum. I suspect your DynaPack facility is using load settings that is much lower; maybe 6 seconds or less. That will most certainly net you less power to the rear hub as more energy is lost trying to 'quickly' accelerate all the mass up to redline. At the same time, if one were to use settings that is much higher, say 15 seconds or more, you'd find rotational loss to be minimal but results offsetted by the heat build up in the intake tract that robs valuable power.

It is possible to achieve figures on the DynaPack that is very close to the DynoJet’s if one were to use ramp-up time that is equivalent to the amount of time it takes to do a DynoJet run. This setting will vary with different cars, output and gearing, however.

 

What I was refering to was that any dyno has parameteres that can be adjusted by the operator. The adjustment affect how that specific dyno reads relative to other dynos both those of the same and different mfg.

As long as the variables that are used to derive power figures are the same amongst machines, the power output levels will be similar.

I was not refering to the specific repeatability of a specific dyno from run to run.


Gearing should not affect the power output of the engine. Imagine a dyno is loading the same level across pulls in multiple gears. The engine does know know or care that the ratio betweeen it and the drive wheels has changed. The power output of the engine does not change. A dyno that shows a change from gear to gear w/ a contant load is... not so good. The dyno is then making some sort of correction and guessing @ power output as opposed to actually measuring power.

Load affects engine power output only up to a certain level. On NA and supercharged engines load has a minimal effect. Once a sufficant load to simulate WOT is reached, more applied load will minimally affect power output unless the thermal capacity of the cooling/oiling system is exceeded. Then, of course, power output will drop due to the additional heat.

On turbo engines load that exceeds real world conditions will cause increased cylinder pressures, as it does w/ NA and Supercharged, thus spooling up the turbos sooner and @ a faster rate.

 

Agreed. No arguement there.


While those statements are not false, they are very misleading in the context discussed here regarding ‘chassis’ dynamometers, such as the DynoJet, DynaPack and the likes. Unless dealing with an ‘engine’ dynamometers, where there are no drivetrain attached, the above statements are not technically correct. Indeed the power generated *at* the engine does not change, but the same power having been transmitted through the drivetrain is subject to a non-constant mechanical and thermal lost as observed by the ‘chassis’ dynamometer at the vantage point of the rear axle or wheel.

In fact, any dyno that show zero delta between dynoing different gears is doing some funny math to offset the results. It gets down to the the 'ramp-up' time for the test run. Short ramp-up time (from using low gears) should reflect higher drivetrain inertial loss than long ramp-up time using (higher gears). However, too long of a ramp-up time also leads to power loss from heat soaking. DynoJets do not have a way to control ramp-up time, but load based dynos do. The most extreme case study would be to test 1st gear vs. 6th gear.


This is where the ‘settling time’ parameter (found on DynaPacks and perhaps other load based dyno) is an extremely invaluable tool. Full throttle is applied shortly after engaging the clutch and the dyno does the rest. A variable load (in real-time) is exerted, enough to keep engine rpm at a specified starting point (e.g. 2000 rpm) and held for a predetermined length of time (e.g. 2 seconds) before ramping up and start of measurement. This takes away all the guess works and decreases operator induced errors and/or cheats.

 

Clint..

Ah - Looks like you did your home work.. I'll Read again

Next time you are going over to Tony's let me know.. I'm an "Old" school guy
but an Ex "Edwards AFB Aircraft Test Guy."

You can then translate this for me.. Ha !

Cheers Amy -

 

bump for a very informative thread

 

What about "load based dynos"?

 

What about them? Do you have a specific question about them?

IMHO Load based dynos are the only way to tune a vehicle. Tuning can be done on an inertia dyno but not recommended.

 

Yes, how do their HP numbers compare to other dynos? i.e. 10% more 10% less, etc.

 

See above.

Really there is no hard and fast rule for what a dyno is going to read. The numbers I gave above are the generally accepted industry standards of comparison. An inertia DynoJet vs an eddy current DynoJet are going to read the relativly the same. Since they came from the same mfg. Mustang is probably the only mfg who's inertia vs eddy numbers vary. Mustang inertia is ~7% less than DJ. While Musang eddy is ~10-12% less than DJ.

Generally the load based dynos read lower than DJ's because of their measuring standards.

 

I found an excellent article about how dynos work and how Load Bearing dynos are more accurate at estimating flywheel HP than Inertia type (ie. Dynojets)

http://www.ktrperformance.com/servic..._faq.htm#power

enjoy,
--mike

 

Intresting!

 

Thanks for the information.

Warmly,

Michael
VRT

www.ViolentRacing.com

 

I see from your website, you've been on quite a few different dynos. Any opinions or observations to contribute?