Testing Methods and Procedures

We all know that the quality of the components that you install on your vehicle is only as good as the R&D that went into their design. Claims of horsepower gains are a major selling point and this article will explore where those numbers come from and how they can be manipulated so that you can be a more informed consumer.

The majority of the numbers that you see advertised on a component will be derived from the use of a flow bench, chassis dyno or engine dyno. All three methods can produce viable results in the hands of a properly educated operator. Accurate test data is the result of controlling all possible variables through proper test procedures. More often than not we see, especially on the forums where HP numbers are as good as gold and quoted as gospel, that certain testing procedures are improperly applied or thrown out the window. Industry standards have been put into place over the years for  reason and deviation from them results in invalid results that do not equate to a legitimate gain in HP when installed on your car.

Correction Factors

The most commonly overlooked procedure is the application of correction factors which are described as follows;

“Correction factors and arithmetic manipulation are necessary to characterize the data to some standard condition so you can compare it against data gathered under different atmospheric conditions.” (Bettes & Hancock 2008)

All too often we hear conversations where someone is saying that their operator didn’t use a correction factor during testing because it makes no difference in the results. It does make a difference and what the use of a CF allows the comparison of tests that were performed at different time under different weather conditions.

Let’s say for example that you were to flow test an intake manifold on a hot day in August and do not apply a correction factor. You’re not so happy with the results so you remove it from the flow bench and make some modifications. It’s the middle of winter before you’re able to find time to flow it again and you don’t use a CF. Low and behold the manifold flows 20% more air and you attribute it to the changes you made to it. In actuality, a good bit of that gain is a result of the cooler, dryer air that is more dense. The proper application of a correction factor to both sets of original test data would provide a more accurate representation of the actual gains or losses by characterizing the data to a standard condition.

Correction factors are also necessary for tests that are performed on the same day. Ambient conditions change constantly throughout the day and it is essential that they be monitored, recorded and applied during all testing. It is imperative that all test conditions be carefully documented so that other tests can be performed under the same conditions for a TRUE comparison of results.

There are two current SAE (Society of Automotive Engineers) standards in practice within the industry. The first is SAE J1349 standard which corrects to 29.53″Hg, 77*F and 1% water vapor pressure. The second is SAE J607 which corrects to 29.92″Hg, 60*F and dry air. Also known as the “Hot Rod” method, SAEJ607 will show 4% higher numbers when applied to the same data.

Correction factors are recorded as a percentage. A test performed under conditions that match SAE J1349 would show a CF of 1.000. If the same test was conducted during less than ideal conditions and a 2% correction was applied the CF would be recorded as 1.020 which means that the CF formula is adding 2% to the HP and Torque readings. Conversely, a test conducted under very good conditions (cool, dense air) equaling a -2% correction would be recorded as .980 meaning 2% is being taken away from the HP and Torque readings.


Where CHP = Corrected Horsepower

CF = Correction Factor

BHP = Base Horsepower

CHP = 400 x 1.020

CHP = 408

Since correction is only an approximation, the chance for errors is significantly greater when more than +/-2% correction is applied. If the calculated CF is greater than this, then it is recommended that you perform your testing under better conditions to get the CF as close to 1.000 as possible.

The use of a correction factor is only as good as the accuracy of the instrumentation used to measure the conditions. The CF probe needs to be placed in the air intake tract so that it is reading the same conditions that the engine is operating under. Placing the probe in an area that is warmer, for example, would create a situation where the numbers are over-corrected. This means that the HP numbers would be falsely inflated.

Accuracy, Instrumentation and Equipment Calibration

Valid results in all forms of testing require accuracy which is a result of the quality of equipment that is used. Acquiring dependable instrumentation is vital to valid testing. Grabbing a meat thermometer from your wife’s kitchen and getting barometric pressure readings from the local news station isn’t going to cut it. You need to have accurate methods of measurement  to verify the results of the primary equipment. It’s not likely for both tools to malfunction simultaneously, so a drastic difference in their readings would raise a red flag.

All measurements should be taken at the site of testing as conditions can be drastically different even just outside the shop. Readings from a weather station down the road are useless.

In addition to back ups, it’s also necessary to have your equipment regularly calibrated. This includes both data acquisition and test equipment. Many manufacturers of high quality instruments include calibration instructions with the tool and/or offer a service where you can send the instrument in to be serviced. Consistency in data acquisition equals consistency in testing which means valid results!

All dynos must be periodically calibrated, both chassis and engine types. Most will have a calibration arm upon which the operator will place one or more reference weights. Let’s say for example that a 50lb weight is placed on the arm. This amount is then multiplied by the length of the arm itself. Let’s assume a length of 3ft. for a total weight of 150lbs. To this you must add the weight of the arm itself unless it is a permanent fixture of the dyno. Assuming a weight of 20 lbs, this means that the torque scale should now read 170 ft/lbs. If this reading varies, proper adjustments must be made to the scale so that it shows the correct torque value. A small error here will manifest itself exponentially during testing.

Flow benches also require calibration to provide accurate results. This involves the use of an orifice plate that flows a known about of air. This is a “standard”. The operator will easure the amount of air that it flows and compare it to a calibration sheet that was supplied with the flow bench. Adjustments are then made to the instrumentation so that their readings match the standard.

Your operator should have a detailed record of when their equipment was serviced and calibrated. When attending a dyno session or witnessing a flow test, ask the operator to show you their methods of calibration and data acquisition. If they can’t or wont, it would be a wise decision to chose a different shop to perform your testing. Beware of any person or facility that cannot verify this for you or says, “trust me” or ” that doesn’t matter”.

Testing Location

The location of the testing within the shop is just as important as the procedures. The dyno or flow bench needs to be placed in an area of the shop that has the most stable conditions available.  Competent testing facilities have dedicated rooms or “cells” for their equipment. This minimizes the variables and maximizes consistency. It would not be wise for a shop to place their flow bench near an outside door, for example where constantly changing conditions would hamper results.

Flow Bench Testing

In addition to correction and location, there are a few specific parameters associated with flow testing in particular. All of the above information applies as well.

Setup of the component being tested is crucial to accuracy. If a cylinder head is being tested, it needs to be placed on a bore size that matches the size of the engine the head is going on.

The intake side of the head should have a radiused entry to direct the air into the runner. The actual intake manifold could also be bolted to the head for testing and would provide accurate results relevant to how the setup will perform as a whole once installed on the engine.

Baseline Testing

An accurate comparison cannot be made without first establishing a baseline. In order to tell how much the airflow of an intake manifold improved, you must first flow test it in it’s stock configuration on the same bench and apply the same correction using the same testing methods. Every component will flow/perform slightly different. Each flow bench or dyno will read slightly different. Baseline testing eliminates these variables.

The baseline consists of a series of three tests whose results are very close, then averaged. The changes are made and a second series of three tests with a repeatability of <1% are performed and averaged. Both sets of averaged data can then be compared. To take it a step further, when possible, is to reconfigure the component back to it’s original specifications and perform another series of tests to re-establish and confirm the original baseline. This step which may seem redundant helps to confirm the repeatability of the test. This is called ABA testing. This may seem pointless but the redundant testing verifies the results and confirms that there were no errors within the testing procedures.

Back Up Testing and Repeatability

Another method of maintaining accuracy involves the use of back up testing. Also known as repeatability, this is just a simple back-to-back test with no modifications made to the component being tested or the actual test procedures. This minimizes the chances for errors that result from mistakes made during the test. Repeatable test methods and also adds validity to the procedures. To be considered valid and comparable, an operator must be able to perform back-to-back tests within <1% repeatability.

Dyno Parameters and Manipulation

Testing an engine on and engine or chassis dyno presents some additional parameters that need to be monitored to ensure accuracy. All of these variables need to remain constant in order to provide accurate results.

  1. Oil temperature needs to be accurately monitored and controlled. Hotter oil is thinner oil which means less friction. This will result in a higher HP reading.
  2. Water temperature affects the expansion rate of all of the components which affects HP. Ring seal, clearances and charge temps are all affected by this and will alter the HP output of the engine.
  3. Tire inflation pressure and temperature affects the amount of rolling resistance and should remain constant for all testing.
  4. Transmission temperature should be treated just the same as engine oil temp.
  5. The amount of tension applied to the tie down straps that hold the car on the rollers must remain constant. It is also important that the angle of the straps remain constant between tests. Both of these affect the amount of drag which will skew the readings.
  6. Differential fluids, just like transmission and engine oil must be constant
  7. The brake system can create drag on the drive wheels. Some operators will remove the brake calipers and fully compress the brake pistons before re-installing them so that they don’t drag
  8. Type and source of fuel. The quality of fuel can vary from day to day and from pump to pump. All testing should be conducted using fuel from the same batch to eliminate variables introduced by the inconsistency of the available fuel.

Any combination of these parameters can be easily manipulated to show false HP gains which is why it is imperative that you find yourself a trustworthy operator and facility. There are infinite ways from them to provide you with optimistic readings to make themselves look better.

Let’s say that this operator straps your car to the dyno and makes a pull with all of the fluids cold. A subsequent pull is made after some tuning changes but now the engine and drive-train fluids are hot. Now of course this second pull is going to show higher numbers simply because of the decreased friction from the hotter oils. The only way to validate the results is to revert back to the original tune and make a third pull with oil temperatures identical to what they were during the second pull.

The operator could also use the correction factors against you unless you know what to look for. One simple way is for them to make a pull using SAE J1349. A change is made but SAE J607 is applied to the data from the second pull. This pull is going to automatically show a difference of 4% and will nullify your results. The same formula must be applied to both sets of data in order to be considered valid.

Another way is to use no CF at all. There have been cases where an operator places an air conditioner in front of the car during a dyno pull and spouting of new, “record setting” numbers. This cooler, denser air is certainly going to show falsely high readings. Even if a CF is applie, going back to our +/-2% rule we would still have invalid results.

Reading and Validating Test Results

Until you have several test sessions under your belt you may be overwhelmed by the amount of information on your dyno sheet. The first thing that you are going to look at is how much peak power you made. Granted, this is one of the reasons for the testing but there are other things you need to look at to validate those numbers.

First you need to verify that all of the parameters remained as constant as possible through all tests. Look at the temperatures and pressured between the different pulls and make sure they don’t vary. This way you know you’re comparing valid data.

Second, verify that the proper correction factor was applied to all of the pulls. Also make sure that the current atmospheric conditions were entered at the time of each test to make sure the CF is accurate. Remember that a CF greater than 1.020 or less than .980 are INVALID!

The third thing you need to do is see how long the pull took. All of the pulls should be timed and you can view  how many seconds it took for the engine to sweep between to set RPM points. The faster the engine sweeps during a pull on the dyno, the faster it’s going to be at the track, period. Take for example an engine that makes 400 peak HP and sweeps from 3500-7000 RPM in 5 seconds. It will run faster than an identical car that makes 450 peak HP but takes 7 seconds to sweep through the same RPM band.

Did you ever see someone that claimed their car put down 700HP and wonder why it only runs 12’s?There is a simple formula that can be used to verify the numbers from a chassis dyno session and it involves going to the drag strip. Because MPH is a reflection of the amount of HP produced, the following formula uses it and the vehicles weight to estimate the amount of power being generated at the wheels. Compare this number to the one given to you by your dyno operator. If the numbers are drastically different, then the operators methods were flawed and the results should be considered null and void. Keep in mind that just like dyno or flow testing, weather correction factors must be applied to track ET and MPH to compensate for differing conditions. For the purposes of this article we will assume that the numbers below are corrected

HP = (.0040 x MPH)^3 x W

Where MPH = Corrected trap speed measured in MPH

W= Vehicle weight measured in lbs.

Assume that the corrected trap speed is 118 MPH and the car weighs 3500lbs.

HP= (.0040 x 118)^3 x 3500

HP= (.472)^3 x 3500


Based on that trap speed and weight, the vehicle is putting approximately 368HP to the ground. Now if you had just come from a dyno session that measured your WHP at 400+, then you should question their practices and find another operator. This is a simple formula and will keep your operator honest.


We expect that you have found this article to be informative and insightful and hope that it has opened your eyes as to the proper way testing is to be carried out. There are a lot of shops out there that are trying to too their own horn that will use improper, questionable procedures to boost their ego and sales. The bottom line is that you need to be an informed consumer so that the time and effort you put into your project is not a waste. The next time you’re in the market for the newest and greatest bolt-on part, ask the vendor for legitimate test documentation.

Comments are closed.