What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Requesting help with torque measurement geometry

hevansrv7a

hevansrv7a

Well Known Member
I am struggling with how to calibrate my test club. I won't bore you with all the things that did not work. And this may not, either.

The present problem is how to correct the reading on the scale for the geometry including the offset between the pivot on which the airplane rests and the centerline of the crankshaft.
TorqueTestSketch.jpg
[SIZE=+1]

I already understand there is an issue with the sideways force on the ball or rod on which the airplane rests, but the analog scale under the wing is where I can measure force and convert it to a torque reading.

I am considering taking a tangent off the arc, the center of which is the crankshaft and correcting the down-force reading with normal trig, assuming the actual torque is the tangent to the arc where the arc intersects the hardpoint that presses on the scale.

I already know there is a vibration issue, so don't even go there. I'm looking for a geometry solution at this time.

Thanks in advance
[/SIZE]
 
I don't understand the picture, or what you're trying to do, at all. But the physics formula for torque is:


Torque = R x F where R is the length of a straight line from the pivot point about which you want to know the torque, to the point where the force is being applied (by your hand, or by the scale, or whatever). F is the amount of force (lbs) you are pushing with.
Here's the tricky part: "x" means you multiply these two number together, then multiply by the sine of the angle formed by the intersection of R and F (e.g., the directions in which R is running and the direction in which you push).

You may notice that with a torque wrench in normal use, torquing a bolt, the angle is 90 degrees and the sine of 90 is one.

Also in practice it is often easier to break down R into components parallel to and perpendicular to F. The cross product is easier, since the sine of 90 is one, and the sine of zero is zero.
 
If my memory of statics class is correct, I believe the correction for the geometry you show is zero.

The torque on the prop shaft will be equal to the product of the horizontal distance from the craft centerline to the scale center and the net weight on the scale.

I built a club dyno for ultralight engines many years ago as a much younger engineer and seem to remember the prop cl being different from the torque arm cl. (since it did not matter)

*EDIT* Checked my MARKs Handbook this AM at the office. My memory is correct.
 
Last edited:
I too found your drawing confusing, but torque calculations are easy.
The basic formula is:
Tw= (Ta * L) / (L +A)
Tw= Indicated torque
Ta=Actual (desired) torque
L= length of torque wrench (from center of handle to center of drive head)
A= added length

If you 9" torque wrench and put an 1" extension on in line with the wrench, you add the full lenght of the extension, so the total length is now 10".

If you put the extention at 90 deg to the wrench, you haven't added anything to the lenght so A=0, so total length is still 9".

If you put the extension at an angle between 0 & 90, the extention is a positive number less than 1. If you put the extension beyond 90 degrees, you now are subtracting from the total lenght. "A"= a negative number.

You could do some math, to get the lengths to add/subtract if you put the extention at something other than 0 or 90. Or you could just measure it.
 
If I understand what you are trying to do, the correction is zero. The torque applied at the point of your arrow ( prop shaft?) equals the horizontal distance from the point of torque application to the location where the scale touches the beam, times the reading on the scale. The vertical distance is not important here, as long as the force applied to the scale is precisely vertical.

Also, there are no horizontal forces on your free body diagram provided. Only offsetting torque and vertical forces. The overall setup should not want to move right or left.

Aaron
 
misc. rebuttals

I do appreciate the attempts to help, but they all seem off-target to me.

The drawing is a front view (looking aft) of the airplane, only one wing is shown, a scale is on the ground under the port wing. A rod connects the scale to the hard point on the wing. The airplane fuse is resting on a large dowel which is supported by a saw horse. The "prop" is a test club that produces only drag, no fore-aft thrust. The air is all moved radially outwards and the torque is the equal and opposite reaction.

1. I already understand the basics of torque - a force applied at right angles to the radius (a tangent to the arc) such that the radius times the force is the torque expressed, in this case, as pound-feet.

2. The engine applies the torque because it uses the club to push air sideways. The torque is being applied to the crankshaft. As shown.

3. Because the torque is at right angles to the radius there IS sideways force on the rod (shown as a circle) under the airplane and on which the airplane teeters.

4. The force applied to the scale (which is on the ground) will be vertical. However, the issue is how to determine the torque from the reading on the scale. If the airplane were suspended on a rotisserie exactly through the center of the crankshaft, then the torque would be applied to the hard point on the wing at a small angle and there would be an easily determined correction using simple trig.

5. The real problem, then, is that the airplane teeters on the rod underneath it and even though I will cage the rod so that it does not actually move sideways relative to either the airplane or the main support, there is still a sideways force. How is that to be figured in the total? If the force on the scale is 100 pounds and the vertical distance between the center of the rod and the center of the crank is 1.5 feet and the hardpoint is 3 feet out from the center of the rod, then what is the torque? Values for convenience only.
 
You have two engineers telling you that the correction is zero.

In Engineering terms, a "couple" represented by the prop shaft, may be assumed to be applied ANYWHERE on a solid body (the airframe). This is not intuitive but it is true.

The vertical distance between the pivot bar and the prop shaft is irrelevant.
The torque of the prop shaft can be assumed to occur at the pivot bar. The rest is easy via the moment arm calcs you already know.

Also, I see no net sideways force. Oscillation perhaps, if the club blades are not mass or aerodynamically ballanced, but no continuous sideways force to impose an additional constant load on the scale.
 
Last edited:
That's four engineers now, unless you've got a side-force generator that we don't know about.

Nothing so far described constitutes a side-force generator. But we're interested in learning about one if there is one.

We're all assuming that the club pushes the air radially or axially or circumferentially or all of these, that the crankshaft is parallel to the rod, and that there's no airframe obstruction to that flow in the plane of the test club. If these assumptions are incorrect then you might indeed have a side force generator, but these assumptions are consistent with your sketch and description.

Another issue is that if you're balancing the plane on the rod centered under it, is that local point beefed up enough to handle that load?

And what about a reaction under the tail (how is the tail supported)? You've described a two-dimensional situation, and actually it's more complicated than that. If you have a rod under the tail, and it's not in line (coaxial) with the rod under the belly, then you can react torque through that couple with a side load.

The support system which you are sort of developing might carry axial, side and vertical forces (no torques) at the belly; side and vertical forces (no torques) at the tail; and vertical force (no torques) under a wing. If the test club doesn't make net forces in any direction, just torque, and the rods are coaxial and the weight of the airframe is symmetric, then the force measured under the wing will give you pure engine torque.

If the direction of the crankshaft isn't parallel to the rods then the situation becomes more complicated and you might need two more scales. For example, if the crankshaft were parallel to the wings, you'd need a scale under the tail. It it were vertical you'd need a scale to measure the side force at the tail.

Dave
An old aerospace P.E.
 
Just curious: What will you learn by measuring the torque from the engine/test club combination?

I thought the purpose of a test club (prop) was to absorb enough energy to allow full throttle operation without overspeeding the engine?

OTOH, if you are comparing different propellers, then you would be interested in the thrust being generated by the different combinations.
 
PCHunt said:
OTOH, if you are comparing different propellers, then you would be interested in the thrust being generated by the different combinations.
Click to expand...

A bit off topic, but static thrust is not of much value in evaluating propeller performance. True performance comparisons can only be made with flight testing (the propeller actually moving through the air, not sitting static, trying to pull air through it).
 
I'm sure that's correct. But what does one learn by measuring the torque on the airframe from a test club, (per the OP's original question.) That's what I don't understand.

Trying to learn something................... not easy for an old man!!! :rolleyes:
 
Pete,
This setup will allow measurement of actual engine horsepower.
H Evans will have RPM from a tach and torque from the scale under the wing.
Those are the two numbers needed to calculate horsepower.
 
Last edited:
If you are using a propeller as a test club, the interaction of the slip stream with the wing/tail will cause significant torque errors. The vortex effect is quite large for the static thrust condition and will cause asymmetric lift on all surfaces.

Roger Bloomfield
RV-9A flying
RV-8 painted
 
Sounds like you have experience with this Roger.
When I did my dyno years ago, the engine and standard prop I was testing was off the airframe. I had a torque arm like the initial image and the whole lot was mounted on a shaft running in a pair of ball bushings so I could read static thrust too. I did not have airframe aerodynamic effects to contend with.
 
What it Measures and Why

First, the "club" is calibrated per a formula provided on a spreadsheet by Jan Carlsson in Sweden and, he says, originally developed by Mr. Eiffel of "tower" fame and later refined by M. Columban of Cri-Cri fame. Mine was built for me by Lonnie Prince of Prince Aircraft. I showed a picture of it during my presentation at AirVenture2012. That presentation is on my website. I also verified the underlying math with Gary Robinson, the aerodynamicist who did the improvements on the Sherpa.

OK, so what is the point? The point is that we don't really KNOW our horsepower and without knowing BHP, then even if you know THP (for which see my presentations on my website) you cannot accurately determine propeller efficiency in the real world. Manufacturer charts are notorious for their unknown factors and can usually be assumed to be optimistic in their errors.

Using the "club", which, I repeat, does not generate any thrust, I can determine BHP from RPM alone, using the spreadsheet provided by Jan and knowing the actual dimensions of it as built.

But what if the club is not accurate? That's what I'm trying to do now. I am trying to calibrate the club against actual torque readings.

Once calibrated, the club can be used to determine BHP for a given MAP and RPM and/or for a given fuel flow. It also follows that this would enable a calculation of SFC.

Oh, btw - I checked my RPM by running at a speed that should provide a picture of a motionless prop in a video camera. Such cameras take 30 frames per second. The math is easy. My iPhone did the job. My GRT tach was, of course, dead-on.
 
hevansrv7a said:

4. The force applied to the scale (which is on the ground) will be vertical. However, the issue is how to determine the torque from the reading on the scale.
Click to expand...

Did the question arise because the scale values seem a bit excessive?
 
Sorry, I missed the zero thrust for the club. . . . The weight on the pivot should change directly with the measured force at the arm. Any thing different and you are seeing aerodynamic effects. The club disk will have a large "donut" of recirculating air, and also a large inflow of air from both front and back. Because of the interaction with both the ground and the air frame you will see some "P-Factor" or influence from the flow field.

If you see a lateral force at the pivot you are most likely seeing a shift of the cg from the vertical alignment, but also possibly some aerodynamic effects.

I think a more accurate way of calibrating the club would be to run it on a known engine that has been Dyno tested. From a known data point, the power can be calculated with good accuracy as long as the RPM spread is not large.

Interesting project, and I think people really confuse propeller efficiency with thrust efficiency. Can't have good thrust efficiency at low speed unless you have a propeller that is the size of a helicopter rotor!









Roger Bloomfield
 
Last edited:
mtnclimber said:
snip..

I think a more accurate way of calibrating the club would be to run it on a known engine that has been Dyno tested. From a known data point, the power can be calculated with good accuracy as long as the RPM spread is not large.
..snip..

Roger Bloomfield
Click to expand...

That is a GREAT idea. Does anyone have a 180 or 200 hp lyco or clone that would be willing to help? I'll pay the shipping. Caveat: the club is not able to absorb more than about 140 hp at 2700.
 
hevansrv7a said:

But what if the club is not accurate? That's what I'm trying to do now. I am trying to calibrate the club against actual torque readings.
Click to expand...

It seems to me that the club only needs calibrating enough to make sure that it generates enough torque at maximum RPM to keep the engine from going faster. If the engine can't get to max RPM then you're not making enough power to fully turn up the club and the club is too coarse. If you can exceed max RPM then the club is too small.

It's a bit like selecting a propeller for cruise or climb.

Unfortunately your test run won't be a standard day seal level conditions. Do you have a means for correcting the results?

Accessories will have an effect on the power too. For example, 27 amps from an alternator is 1/2 HP right there, ignoring efficiency losses while generating it.

One thing I'm wondering, is what you'll be able to use the information for?

Dave
 
More answering and clarifying

David Paule said:
It seems to me that the club only needs calibrating enough to make sure that it generates enough torque at maximum RPM to keep the engine from going faster. If the engine can't get to max RPM then you're not making enough power to fully turn up the club and the club is too coarse. If you can exceed max RPM then the club is too small.
Click to expand...
The engine is not developing full power because the club prevents normal airflow into the airbox. Therefore your analysis fails. Anyhow, that would not help calibrate it.
David Paule said:
...
Unfortunately your test run won't be a standard day seal level conditions. Do you have a means for correcting the results?
Click to expand...
Yes. Lycoming or Superior charts are assumed accurate as to the effects of those variables.
David Paule said:
Accessories will have an effect on the power too. For example, 27 amps from an alternator is 1/2 HP right there, ignoring efficiency losses while generating it.
Click to expand...
Well, that is the point of doing this on the airplane instead of in a manufacturer's test cell where they don't specify what was mounted or not.
David Paule said:
One thing I'm wondering, is what you'll be able to use the information for?

Dave
Click to expand...

From my previous post: "OK, so what is the point? The point is that we don't really KNOW our horsepower and without knowing BHP, then even if you know THP (for which see my presentations on my website) you cannot accurately determine propeller efficiency in the real world. Manufacturer charts are notorious for their unknown factors and can usually be assumed to be optimistic in their errors. Using the "club", ... I can determine BHP from RPM alone... Once calibrated, the club can be used to determine BHP for a given MAP and RPM and/or for a given fuel flow. It also follows that this would enable a calculation of SFC." So besides propeller efficiency, true BHP and SFC, what would you want to know?
 
Another engineer votes for "zero."

As previously stated, if the rollers supporting the fuselage truly offer no lateral force, then the only reaction is a vector that passes through the axis of rotation, and the correction is zero. But then, the fuselage is free to roll of the edge and onto the floor with the engine running. :eek:

On the other hand, if you determine the horsepower through flight test, all of those 'less than efficiencies' get thrown into the bucket of stuff that didn't make the airplane climb - alternators, propeller efficiency, shop rag in the intake. If it doesn't show up on my VVI, why do I care where it came from? :cool:
 
trying to reply

flynwx said:
As previously stated, if the rollers supporting the fuselage truly offer no lateral force, then the only reaction is a vector that passes through the axis of rotation, and the correction is zero. But then, the fuselage is free to roll of the edge and onto the floor with the engine running. :eek:
Click to expand...

The single roller will be caged below and above so that it will act as a pivot for the fuse. It will not roll anywhere.

flynwx said:
On the other hand, if you determine the horsepower through flight test, all of those 'less than efficiencies' get thrown into the bucket of stuff that didn't make the airplane climb - alternators, propeller efficiency, shop rag in the intake. If it doesn't show up on my VVI, why do I care where it came from? :cool:
Click to expand...

I believe, based on a lot of study (which you can find in my various presentations) that you cannot determine BHP from any flight test. As stated, if you know THP and BHP then you do know net-installed prop efficiency. If you don't know them both then you don't know prop net-installed efficiency. Good prop design must include taking into account the airframe. Isolated wind-tunnel tests, even if you could do them, would not give a correct or (thus) usable result.

Your vertical speed indicator (or better, a WAAS-GPS) times your weight can tell you the net surplus horsepower that can be used for climb after subtracting that needed for flight at that CAS. That is still THP, not BHP.

If you could get 5% more thrust at 8000' on a 180 HP engine that would be about 6.75 BHP. If that could be obtained by changing the prop, then it would be, for example, 6.75 HP for about $2,000. And without affecting engine reliability, longevity, etc. It could also be used to reduce fuel consumption for a given speed. It seems obvious to me that being able to correctly and accurately determine prop efficiency as installed in a flying airplane is worthwhile. But others may disagree.
 
You must log in or register to reply here.
Back
Top