Exclusive first Model 3 Dyno

[mongo]

Which is a rotating force- which is torque. By definition.
No, it's not.

work is force over distance. Which is what HP is.
You appear to have this exactly backward.
The definition of 1 horsepower is a unit of power equal to 550 foot-pounds per second
(which is yet more proof hp is computed from measured torque)
See my link above. It explains how each of the major dyno systems works.
all of them directly measure torque and rpm, and use those to compute HP.

Torque and work both have units of foot pound,(or pound feet) but torque is force at a distance and work is force through a distance. SI uses Nm for torque and J for work to reduce ambiguity,
Torque*RPM gives work via (lb*ft) (force at a distance) /(radius in ft)) * RPM * 2*pi*(radius ft) = lb through a distance in a time.


Somewhat tangential: I hope Tesla is using drive units in regen mode (or some generator device) for the test load when doing drive unit burn in (vs wasting the energy). Theoretically, you could get HP/kW directly from that output, but mounting the generator (or drive unit) on a load cell to measure torque and reading the RPM would be more accurate.

 

[SageBrush]

Torque*RPM gives work via (lb*ft) (force at a distance) /(radius in ft)) * RPM * 2*pi*(radius ft) = lb through a distance in a time.

For the life of me, I cannot follow imperial units but torque * rpm is a unit of power, not work.
I may have misunderstood you.

 

[mongo]

For the life of me, I cannot follow imperial units but torque * rpm is a unit of power, not work.
I may have misunderstood you.

Yah, that is the equation for Power (HP if you divide by 5252)
Don't know why I bothered to include it...

 

[PhaseWhite]

Can't wait to see the first Model 3 P dyno results. On a related note, thanks for the post of the Model S 75D dyno. This is the first/only Dyno of a non P model S I've seen. I have searched youtube long and hard but for the life of me cannot find dyno results for a Model S90D or Model S100D. All you find are P variants. Anyone else have any links?

 

[scaesare]

No, dynos do not measure torque. If they did, you’d be seeing numbers in the thousands. They have to reverse calculate the torque using the input gear ratio.

You have it backwards.

They measure torque at the wheels, and then have to apply a conversion based on the gear ratio you supply to derive torque at the shaft.

Think about it, the dyno can only measure what's moving the rollers... it has no idea what's upstream, including any differential or transmission gearing.

And given that you can measure torque, and RPM, but horsepower is the result of a mathematical formula with an arbitrary relationship between those two values, it also must be a derived value.

 

[Mysterylectric]

You have it backwards.

They measure torque at the wheels, and then have to apply a conversion based on the gear ratio you supply to derive torque at the shaft.

Think about it, the dyno can only measure what's moving the rollers... it has no idea what's upstream, including any differential or transmission gearing.

And given that you can measure torque, and RPM, but horsepower is the result of a mathematical formula with an arbitrary relationship between those two values, it also must be a derived value.

You are more than welcome to call dynojet research and tel the engineers they are wrong.

 

[Daniel in SD]

The laws of physics state that acceleration rate is directly proportional to how much power the tires place on the heavy roller to get it to rotate.

From your link. That sounds like a power measurement to me derived from how much energy it takes to change the speed of the roller and how long it takes to do it. The point is that as a tool the dyno measures power because in order for it to calculate the torque of what you're testing you need to give it information about gearing and tire radius. Nominally the horsepower is independent of gearing and tire size.

(bold added- or to paraphrase, hp (which is calculated from measured torque) sells cars, but torque wins races)

Horsepower wins races.
Energy = Mass * Velocity^2
Energy = Power * Time
Time = Mass * Velocity^2 / Power
If torque won races then diesel cars would be the quickest models.

 

[Mysterylectric]

From your link. That sounds like a power measurement to me derived from how much energy it takes to change the speed of the roller and how long it takes to do it. The point is that as a tool the dyno measures power because in order for it to calculate the torque of what you're testing you need to give it information about gearing and tire radius. Nominally the horsepower is independent of gearing and tire size.

Horsepower wins races.
Energy = Mass * Velocity^2
Energy = Power * Time
Time = Mass * Velocity^2 / Power
If torque won races then diesel cars would be the quickest models.

And since the rollers are weighted, and the dyno measures how long it takes to change velocity, it’s able to calculate the power without the need for a gearing input to measure torque.

 

[Knightshade]

You are more than welcome to call dynojet research and tel the engineers they are wrong.

One of the approaching dozens sources I cited proving you wrong used Dynojet as one of its sources on how their own dynos work.

They measure torque. And then compute horsepower.


I'd suggest you call dynojet so they too can explain like every other source has why you are wrong.

The point is that as a tool the dyno measures power because in order for it to calculate the torque of what you're testing you need to give it information about gearing and tire radius. Nominally the horsepower is independent of gearing and tire size.

That is exactly backward.

It measures torque and computes horsepower.

As literally every single link provided states.

Again- HP=Torque x RPM/5252

The very formula tells you HP is computed by measuring 2 other things. Torque and RPM. Which are the 2 things a dyno measures.

I'd suggest you go back to the Wiki link that gives the history of the term horsepower.

The unit of 1 hp is defined as- a unit of power equal to 550 foot-pounds per second. (very originally it was defined by James Watt as 32,572 ft⋅lbf/min)

Because it's a computed number, derived from a measurement of torque and rpms. Terms which both existed, and were measured, before horsepower ever was

That'd be impossible if you need to measure HP to compute torque.

Horsepower wins races.
Energy = Mass * Velocity^2
Energy = Power * Time
Time = Mass * Velocity^2 / Power
If torque won races then diesel cars would be the quickest models.

I mean... "horsepower sells cars, torque wins races" is a quote from Enzo Ferrari.... but what did he know about cars, right?

The best explanation of the quote I've seen is this, and even addresses your Diesel remark:

Mr. Ferrari believed that torque was most important of the two provided that both were present in sufficient amounts.

With lots of turns, acceleration, and braking, torque wins hands down IF there is sufficient horsepower. Without the horsepower, diesel trucks would win auto races. Without torque, jet-powered cars would win all the races


Obviously Mr. Ferrari was also discussing the type of racing where turns, braking, and re-accelerating are important... a drag race would be quite different for example (where indeed jet powered cars hold all the records)

 

[scaesare]

You are more than welcome to call dynojet research and tel the engineers they are wrong.

For grins n' giggles I Googled "what does a dynamometer measure" and looked at every single link in the initial results page that discussed the type of dyno in question. They all refer to measuring torque (or force) and recording RPM. they also all refer to calculating (or "figuring out" or "determining") power based on that.

So I don't believe the DynoJet engineers are wrong. I suspect you understand them incorrectly.

Thought experiment:

A torque wrench is placed on a lug nut. The nut experiences 100 ft/lbs. of force (torque). How much force is being exerted on the end of the wrench handle?

Answer: You have insufficient information to know.

I now tell you the wrench handle is 1 foot long. You now may calculate the force exerted on the handle end is 100 lbs.

I now replace this with a wrench that is 2 feet long. The lug still experiences 100 ft/lbs of torque. You may now calculate the force at the handle must now 50 lbs.


This is analogous to the dyno... it measures the end result after any mechanical leverage is applied through gears. Only by manually supplying a gear ratio may you calculate the input force at the shaft.

Adding speed to all of this simply allows the mathematical division necessary to derive the arbitrary power value of "horsepower".

 

[mongo]

This is analogous to the dyno... it measures the end result after any mechanical leverage is applied through gears. Only by manually supplying a gear ratio may you calculate the input force at the shaft.

Adding speed to all of this simply allows the mathematical division necessary to derive the arbitrary power value of "horsepower".

And this is why they use HP as the reference unit, it doesn't change (other than losses) when going through a transmission/ differential.
Speed_out = 1/n * speed_in
Torque_out = n * torque_in
Speed_out*Torque_out = 1/n * speed_in * n * torque_in = speed_in * torque_in

Given two motors and an ideal gearbox (infinite range zero loss) the motor with higher HP performs better since the output at the same speed will have more torque/ force/ ability to accelerate the vehicle mass....

 

[scaesare]

And this is why they use HP as the reference unit, it doesn't change (other than losses) when going through a transmission/ differential.
Speed_out = 1/n * speed_in
Torque_out = n * torque_in
Speed_out*Torque_out = 1/n * speed_in * n * torque_in = speed_in * torque_in

Given two motors and an ideal gearbox (infinite range zero loss) the motor with higher HP performs better since the output at the same speed will have more torque/ force/ ability to accelerate the vehicle mass....

Agreed.

And to derive horsepower, you must perform a calculation based on 2 measurable values. One of those is torque, which is measured at the point it is presented to the measuring device, which for a chassis dyno is at the wheels.

 

[diamond.g]

If the dyno measures power, none of this explains why one poster saw 394 of them (the power) instead of the 335 the BMS is supposed to limit you to. This goes back to the question that has died down, did Tesla nerf the LR 3?

 

[Krash]

The original dyno torque numbers are incorrect. I don’t know if the gear ratio was set wrong, or something else was configured improperly or if the machine was malfunctioning, but the dyno numbers given match neither the car’s self reported numbers nor do they match the drag numbers. When I asked the physicists here on the forum as to the discrepancy they simply stated that the dyno numbers were a mathematical impossibility.

Nice to see Mysterylectric redoing the job properly.

My only request is that dyno day participants run the API performance test runlog (iOS Powertools or Android Dashboard for Tesla) at the same time as the dyno, or better yet gather the CANBus data at the same time, to compare. None of these work on the 3 at the moment as the API for the 3 is mysteriously broken and the CANBus on the 3 has yet to be cracked.

 

[Mysterylectric]

I just spent 31 minutes on the phone with Dynojet Research regarding the 424xLC2 dyno we used for the 75D run. This dyno DOES NOT MEASURE TORQUE in an acceleration run. It uses an acceleration delta based off LINEAR force applied to the rollers, and calculates power based off that and the weight of the rollers. If we are to measure torque, and calculate horsepower, we need to do a manual step run using what is called an Eddy brake, that applies load to the rollers to hold a user specified speed, and it measures in direct torque to the rollers, and is capable of holding up to 1800ft/lbs of direct torque per drum, and uses the force applied to the brake, and the user input RPM, to calculate motor torque and horsepower. In an acceleration run, it uses the weight of the rollers, and the acceleration delta to measure horsepower, and then uses the user input RPM to calculate motor torque. The telephone number for DynoJet Research is (800)992-4993, press option 2 for chassis dyno support.

 

[mongo]

I just spent 31 minutes on the phone with Dynojet Research regarding the 424xLC2 dyno we used for the 75D run. This dyno DOES NOT MEASURE TORQUE in an acceleration run. It uses an acceleration delta based off LINEAR force applied to the rollers, and calculates power based off that and the weight of the rollers. If we are to measure torque, and calculate horsepower, we need to do a manual step run using what is called an Eddy brake, that applies load to the rollers to hold a user specified speed, and it measures in direct torque to the rollers, and is capable of holding up to 1800ft/lbs of direct torque per drum, and uses the force applied to the brake, and the user input RPM, to calculate motor torque and horsepower. In an acceleration run, it uses the weight of the rollers, and the acceleration delta to measure horsepower, and then uses the user input RPM to calculate motor torque. The telephone number for DynoJet Research is (800)992-4993, press option 2 for chassis dyno support.

I'm agreeing, but attempting to clarify a couple points.

In pure inertia mode it measures torque, but as a derived value, not directly.

Physical side:
The drum is round, the wheels apply a force tangential to the drum's center of rotation, that is a torque (sort of like linear, but everything is round). The torque produces an acceleration in the drum proportional to the rotation inertia of the drum (calibrated).

Dyno side:
The system measures the speed change of the drum along with absolute speed v.s. time. Based on that, it calculates the change in kinetic energy in the drum. That is the power input to the drum (HP at rear wheels). Based on the drum radius and RPM, the dyno can also calculate what the force (torque) was at its surface. Then with vehicle gear ratios information, it can calculate torque/ RPM at the motor/ engine (HP is just efficiency correction).

With the eddy brake, the torque applied by the brake is added to inertial change of the drum to get total power allowing for longer runs or fixed operation points. At a fixed speed, torque could be directly read.

Dynojet Automotive Chassis Dynamometer Model 424xLC2 All Wheel Drive

 

[Mysterylectric]

I'm agreeing, but attempting to clarify a couple points.

In pure inertia mode it measures torque, but as a derived value, not directly.

Yes, this exactly. My wording is poor lol.

 

[mongo]

Yes, this exactly. My wording is poor lol.

Your wording was fine, I was mostly redundant and pedantic.

 

[Mysterylectric]

Your wording was fine, I was mostly redundant and pedantic.

I appreciate the added clarification.

 

[scaesare]

I just spent 31 minutes on the phone with Dynojet Research regarding the 424xLC2 dyno we used for the 75D run. This dyno DOES NOT MEASURE TORQUE in an acceleration run. It uses an acceleration delta based off LINEAR force applied to the rollers, and calculates power based off that and the weight of the rollers. If we are to measure torque, and calculate horsepower, we need to do a manual step run using what is called an Eddy brake, that applies load to the rollers to hold a user specified speed, and it measures in direct torque to the rollers, and is capable of holding up to 1800ft/lbs of direct torque per drum, and uses the force applied to the brake, and the user input RPM, to calculate motor torque and horsepower. In an acceleration run, it uses the weight of the rollers, and the acceleration delta to measure horsepower, and then uses the user input RPM to calculate motor torque. The telephone number for DynoJet Research is (800)992-4993, press option 2 for chassis dyno support.

Sure.... that's another way to approximate the force at the wheels.... or in this case derive it, which you then use that derived value to then in turn derive HP.

My point was that when you said: "No, dynos do not measure torque. If they did, you’d be seeing numbers in the thousands", many dynos do exactly do this. The fact that DynoJet estimates does not change the primes that the roller experience the torque applied to them after the multiplication factor of the gearing.

So the rollers DO see thousands of pounds of force. The fact that they don't display those numbers is because the computer takes in to account the separately specified gear ratio to determine what the shaft torque (and thus HP) is, and that's the value presented.