Dyno graph technical information inside

[JJ@WMS]

Since we own a Dynojet dyno and use it often I figured I would take the time to show and explain a few things about graphs.

First is smoothing.

Smoothing is a way to display the graph in a manner which is more appealing to the eye, it also gets rid of the rough jagged points of the dyno plot. Those jagged points come from the signal from a magnetic pickup attached to a drum where the drivetrain affects how smoothly the drum is rotated. If a vehicle has an out of round tire or an ignition miss then those things will severely affect how smooth the dyno plot is shown. There are also other factors that can affect the plot.

Standard Smoothing in the industry is set at 5 to show a standard smooth graph and to remove any anomalies from the graph/plot. Moving smoothing down to 1 will show more of those anomalies and usually give you a few more hp and tq along the way.

Above is our car completely stock with the SAE correction and smoothing at 5. SAE is the standard correction for temp and atmospheric conditions. 276hp/286tq are the numbers.

Above is the same dyno graph but this time the smoothing is set at 1, the correction factor is still SAE. Now its 281hp/290tq A 5hp/4tq gain from nothing other then changing the smoothing.

Now lets look at STD vs SAE correction factors. The following borrowed from LS1 tech;

SAE:
"SAE (Society of Automotive Engineers), USA. Power is corrected to reference conditions of 29.23 InHg (99 kPa) of dry air and 77 F (25°C). This SAE standard requires a correction for friction torque.

STD:
STD is Another power correction standard determined by the SAE. Power is corrected to reference conditions of 29.92 InHg (103.3 kPa) of dry air and 60 F (15.5°C). Because the reference conditions include higher pressure and cooler air than the SAE standard, these corrected power numbers will always be about 4 % higher than the SAE power numbers. Friction torque is handled in the same way as in the SAE standard."

The debate will rage on forever about which one is more accurate, I tend to lead with the SAE numbers but have used STD numbers before as well, its really up to the end user as to what numbers you feel represent the actual hp/tq of the vehicle.

Now here is our same stock car with STD numbers with the smoothing at 5;

283hp/293tq

Now with the same graph with the smoothing at 1;

288hp/296tq A 5hp/3tq gain just from changing the smoothing.

Lastly I'd like to make a short statment about tq spikes which will show great tq numbers. When you see those spikes look beyond that spike and further down the curve to see what the actual/average tq is. That will give you a better idea of the power that vehicle and/or modification is making.

Also keep in mind that dyno's calculate hp based off of a measured torque. HP is not measured, torque is and there is a calculation to determine hp based off of the tq the vehicle made over a given amount of time.


All of the above information is posted to show you the differences between different correction factors and different ways of showing the dyno plots in order for you to be a bit more knowledgeable when it comes to this subject.

I hope this information helps in some way and if I can be of any help dont hesitate to give me a call.

Thanks for your time,

JJ@WMS

Sponsored

 

[dragonacc]

Good post. Would you mind also including information on the differences between unloaded and loaded dynos too?

 

[Juben]

^Yep, I'd like to see info on that as well.

 

[BigCatDan]

Thanks JJ, this is great info. I was wondering about this yesterday when I had seen a graph with jagged points. Looked odd, but now I know! Thank you again!

JJ the small HP/TQ gains when the graph is smoothed.. are they not real gains, or are they real gains? That part i'm unsure of.

 

[arghx7]

Some of what appears to be noise in the power and torque trace are actual torque fluctuations from combustion. There are variations in torque from cycle to cycle, especially at WOT when you have a lot of spark retard.

A few points on correction factors:

The actual measurement is the uncorrected value. Applying a correction factor doesn't make your engine generate any more torque/power. It's sort of like the answer to a philosophical question of what the engine "really" makes.

Correction factors were really designed for engine dyno labs, for a case where your supply air conditioner maybe can't hold the temperature and pressure as tightly as it you would like.

The STD correction factor comes from an old standard (j607) for small engines like you would find on lawn mowers. The standard was superceded by j1349, what we would call the "SAE" correction factor. The main difference is that a STD test would set the lab supply air conditioner to have cooler air with higher pressure. STD test conditions is like a cool fall day at sea level with a high pressure weather system in the area. "SAE" conditions is more like a warm late spring day.

Good post. Would you mind also including information on the differences between unloaded and loaded dynos too?

"unloaded" dynos don't directly measure torque (they calculate it based on acceleration) and don't have a strong enough brake to hold a steady engine speed. They are common for performance chassis dyno tuning but that's pretty much it. That being said, there are a ton of unloaded dynos out there (they are also called inertia dynos) used for horsepower pulls and quick testing.

Pretty much any engine dyno is a "loading" dyno, and a large number of chassis dynos are loading dynos. The brake is strong enough to hold an engine speed steady while you're making max torque at a given speed. The torque is measured directly, and the power is calculated.

You can calibrate a "loading" dyno just like you calibrate a torque wrench. If I measure 100 lb/ft of torque and confirm that with a calibration, that's 100 lb/ft of torque and whatever power that makes at a given speed.

A more expensive type of loading dyno would be called a motoring dyno. You can actually put the car on there with the engine in neutral and it will speed the wheels up to whatever speed you want. The same type of thing exists with engine dynos. These types of dynos can measure friction. They measure the amount of torque it takes to hold the vehicle (or engine) to a specific speed without torque going to the wheels (or flywheel). That's called the friction torque.

 

[Juben]

^Now that's the kind of posts I love reading! Well said.

 

[dragonacc]

Thanks for the info arghx7.

 

[Herr_Poopschitz]

How about differences in how long it takes a turbo to spool?

Are the revs shown from dyno to dyno for where a turbo makes full boost a reliable indicator for what a customer can expect on the street?

 

[arghx7]

^ That mostly depends on how fast the engine speed ramps. So for most dynos, that's controlled by what gear you dyno the car in--unless your set a speed damp with the dyno loading.

On an inertia dyno (unloaded dyno that measures acceleration and backcalculates power and then torque) there's so little load on the vehicle that it's almost like flooring the car on a slight downhill on the street. But it still also depends on the roller design and the gear you are in.

The flip side of that is what you see in advertised manufacturer torque curves, hitting peak torque at silly engine speeds like 1250rpm (like BMW will advertise). Unless it's some kind of stationary generator engine or something, that condition really only happens in an engine dyno lab, because the engine speed is controlled by a dyno and ramps very slowly... it's slower than doing a 6th gear pull in a vehicle on a dyno.

The other thing to keep in mind is the torque converter behavior if it's an automatic. It depends when and how it locks and unlocks.

Sponsored