Sydney Dyno Day - October 31st

[parso_rex]

You can't measure torque accurately at the wheel because it's affected by gearing - lower gears produce higher torque figures. My understanding of what happens is that they do is measure the power, and then from knowledge of various aspects of the car's drivetrain, mathemetically derive the torque that the engine is producing.

One of my pet peeves with dyno graphs is power/torque plotted against km/h - it's meaningless because it only gives you figures for one particular gear. I much prefer seeing graphs where the power and torque are plotted against engine revs. Knowing the gear and km/h per 1,000rpm, you can relatively simply derive a power/torque vs rpm graph if you've got access to all the data points - something that only the dyno computer has...

Actually its a little different to what you might think in this case.
What you can measure accurately is torque at the rollers and power is actually back calculated from this. The can be some confusion depending on what you display, whether it be say roller torque derived torque derived power and then people get out their calculators and its all over the place . This isn't the case with Todds graphs though as he's done this about 10 zillion time before.
Now the kmhr thing thats another issue and very valid every dyno brand is different. A lot of people seem to forget things such as are the the km/hr figures you see the roller's speed or is it the wheel speed ?. Whats done with the mainline is pretty funky and you might have noticed Todd initially bringing the car up to a tacho indicated 3000 rpm before the power runs commenced. This is to let the dyno aquire an accurate vehicle indicated rpm v roller speed and its locked in for that car. From this you can derive the RPM and you'll see it appearing on the X axis on the printout - if it was enabled.
The other way is to use the Dyno's OBDII connectivity with the vehicle to obtain the RPM that way. You can also do this with Ignition advance, temps and manifold pressure (but not on a MKV and some others VAGCOM required there ) and so on. With cars carrying an aftermarket ECU such as Autronic theres even more options available but thats a topic in itself.
I think this might be something some of you might be interested in when comparing tunes. The downside is of course a small amount of extra time connecting things up which could be a PIA when you have 40 odd cars waiting.

The Mainlines are good and very repeatable even at say another workshop. They also can't easily be manipulated by relocating temp probes or choosing incorrect test modes and the like. Pretty good I reckon

Cheers

Mark

[Manaz]

You're right, I need to stop posting when I'm tired.

You can accurately measure the torque at the wheels (tractive effort) - but it varies across gears, and so if a run was done in first gear, it'd have sky-high torque figures when compared to a run done in say 4th, and I believe can also vary based on some other factors that may vary run to run. From the measured torque and rpm (which of course the dyno needs to know), power is calculated (since power is a function of torque and rpm).

I believe the 'derived' term comes from the calculations that are done to make sure that the results are repeatable.

My pet peeve relating to graphing against km/h is just a personal opinion - I'd prefer to have my outputs mapped against rpm is all.

The Mainlines do indeed have a good reputation - unlike some other dynos with such things as "shootout mode"...

[Guy_H]

Parso Rex is correct, its an indicative figure (but a very important one) - Tractive effort is a very, very old term used in locomotive engineering.

Lets just say that your car would not move without torque & torque decreases at a friction point where horsepower takes over.

Which will be quicker 0-60?

a car with 200kw & 800nm or a car with 300kw & 400nm's ?

Have a look at the big diesel AWD's (SUV's) - they are pushing 1000nm of torque & can propel a 3 ton monster 0-100 quicker than your average GTI which weighs half!

There is some info on MRT's site (or Ecutek's) about good remaps & avaerage ones - the torque plot is one of the main keys to it!

[windyzz]

To technical for me .

[Big Yellow]

how much torque did the 173kw mk4 r32 put out?

[gldgti]

You're right, I need to stop posting when I'm tired.

You can accurately measure the torque at the wheels (tractive effort) - but it varies across gears, and so if a run was done in first gear, it'd have sky-high torque figures when compared to a run done in say 4th, and I believe can also vary based on some other factors that may vary run to run. From the measured torque and rpm (which of course the dyno needs to know), power is calculated (since power is a function of torque and rpm).

I believe the 'derived' term comes from the calculations that are done to make sure that the results are repeatable.

My pet peeve relating to graphing against km/h is just a personal opinion - I'd prefer to have my outputs mapped against rpm is all.

The Mainlines do indeed have a good reputation - unlike some other dynos with such things as "shootout mode"...

i'll try to put this clearly for anyone that is unsure -

all dyno's measure torque you cant measure power directly, its a derived figure. the toruqe at the wheels is just fine, since youcant measure the torque from anywere else easily.

now, heres the important part - so long as you know what rpm the engine is doing for whatever road speed you have (gear ratio) you can back-calculate all the things you need to know. so, as was explained, before the run, the operator brings the vehicle (in the gear for the run) up to a predefined point - lets say 3000rpm for arguments sake) and at this point, the machine will look at the indicated road speed, and can work out a gear ratio (taking into consideration everything from the wheel to the gearbox input shaft - simply because it knows the roller diameter, equivalent road speed, and the engine rpm)

next, because the machine knows how fast the engine is going at any roller speed (because it has a gear ratio calibrated into it), the next thing is to do the 'run', and measure the torque as the rpm goes from start to finish.

the torque measured is "at the wheels" - but heres where the power part somes in - we know for a fact that the power transmitted to the wheels is whatever torque is measured at the wheels, multiplied by the engine speed at that point. P=Txw (power (watts) = torque (Newton.meters) x angular velocity (rotational speed, in radians/second). again, the dyno knows what angular velocity the rollers are doing at any particular point, and so the power that is being transmitted from the engine, through the gearbox, wheels, tyres to the 'rolling road' is known and exact. this power curve is representative of what is at the engine, minus losses... so sometimes we want to say "well, you lose 15% in the drivetrain" - which may or may not be accurate (varies for everything) but is at least more appropriate than to guess what kind of losses there are in torque. this is because the power is an energy rate, and generally speaking in mechanical systems, we can talk about losses as a pecentage of the energy rate more meaningfully than we can about the torque in a specific place.

so, anyway, since we also know an overally gear ratio for the system, its a piece of cake to stick some engine rpm values on the power graph in appropriate places, and use those rpm to find the "derived torque" for the engine. this is not innaccurate, there's no black magic here -its a simple rearranging of the P-Txw fomula - T=P/w! its just that instead of using the w for the rollers, we use the (known) engine speed.

i hope thats at least kind of clear for people.

one other thing - if you have rpm and km/h appearing on the same axis, this is useful since if you know some things about your car (gearbox ratio's, tyre sizes, etc) you can work out the overall gear ratio, and check it against the calibrated one before the run, that is used to find your torque curve).

some dyno's do provide a printout without a meaningful torque curve, as manaz has suggested. in the case of the mainline dyno's printout, this is not the case - it is indeed meaningful, since it is the 'derived' engine torque output, MINUS losses. infact, if you somehow got your engine dyno'd, you would then be able to work out your drivetrain losses, by taking the difference between the 2). however, even if you only get a printout with torque at wheels, its a simple mathematical excercise to again, figure out your gear ratio and use that to make a new torque curve (it will simply be scaled).

[gldgti]

Now the kmhr thing thats another issue and very valid every dyno brand is different. A lot of people seem to forget things such as are the the km/hr figures you see the roller's speed or is it the wheel speed ?

just to point out here- the km/h is both the wheel speed and roller speed - they must be the same, since there is no slip. the rollers must be going the same speed as the wheels. think about it

where confusion can set in is in the translation of all these different rates and measurements into something that can be understood, somehow. regardless, the only way to learn anything meaningful from any dyno, is to measure a torque, and know at that point what angular velocity (rotational speed) the measuring device (dyno) is doing. the indiated road speed may be 110km/h, but the point really is that the rollers are measuring a torque, and if the rollers are 200mm (0.2m) diameter, then they are doing:

110 km/h = 110,000 m/h = 30.55555555 m/s

roller circ = pi x d = 3.1416 x 0.2m = 0.628m

angular velocity roller = 30.5555555/0.628 = 48.6Hz (cycles/second)

48.6Hz x 2xpi = 305.7 radians/second (this is the number the dyno needs to know)

and if it takes a data point here, and measures that there is 230Nm driving the roller at that speed (through the wheels of the car), the the power transmitted to it is

P = T x w

= 230 x 305.7
= 70,311 Watts (or 70.311 KW)

[Guy_H]

To technical for me .

Lets just say you could have 1000 horsepower & no torque - your car wouldn't even move!

[Guy_H]

GTTom,

Found some of our calibration dyno's from the Mainline dyno.

Over 30 runs on 2 separate test cars we netted mean average :

Stock 103kw

Stage 1 121kw

Torque jumped from 3350 newtons to 3969 newtons.

There were more gains in the curve, but those are the peak numbers!

If you are ever up in Brisbane, drop in & have a look at all the plots

[johnc]

Guy, do you have any numbers for the Edition 30 on a mainline dyno? You said the kws look a little low. Wondering what you think I should be putting down.

[GTom]

GTTom,

Found some of our calibration dyno's from the Mainline dyno.

Over 30 runs on 2 separate test cars we netted mean average :

Stock 103kw

Stage 1 121kw

Torque jumped from 3350 newtons to 3969 newtons.

There were more gains in the curve, but those are the peak numbers!

If you are ever up in Brisbane, drop in & have a look at all the plots

Thanks Guy!

Might be up there next year sometime!

[Guy_H]

Hi John,

Sorry, never ran the ED30 on the mainline.

On Dyno dynamics (shootout) we see ~ 190 from a stage 1 Pirelli, and ~ 200 from a car equipped like yours.

From what I understand, the Mainline will read slightly less than those.

Only real way to tell would be to run your car on both. Mind you, you could run your car on twenty dyno's & end up with twenty figures as well.