Continuing the Thrust Stand Controversy

[hinote]

Hello all:

This thread is an effort to divert the ongoing thrust stand discussion (!?!) to a more suitable venue.

May I also reiterate what might be called "rules", here?:

1. NO name calling!

2. No belittling of another person

3. Present only the facts or questions you may have--or respond only to the facts or questions others may have presented.

4. If you disagree you may state such--but only in a respectful way!!

So, fellow kiddies--if we're all willling to play by the rules...................

I have some thoughts regarding the status of Viv and Luc's thrust stand.

It appears that considerable thought, effort and money has gone into their effort. I'm certainly not going to be the one to place any doubt upon their expertise. Also, please note that I don't have the engineering expertise to review or criticize the methods they're using, to get their data.

However--that said, I can't help but think there must be something else going on. It just doesn't seem right that their fabulous thrust (and other data) stand should repeatedly indicate an increase of thrust performance for multiple engines, compared to other (however primitive) measuring devices. Statistical interpretation of the results alone would indicate the need for further evaluation.

The problem here (as I see it) is there is no CONTROL for Viv and Luc's effort.

I would like to see them seek outside assistance in the form of an independent expert (or experts) who is willing to review their effort--with a very careful and skeptical eye toward trying to find something that might be wrong.

This would have no value to us poor pulsejet enthusiasts, on this Forum. The value would lie in providing Luc and Viv with solid correlation for their claims.

I appeal to logic and the scientific method for my discussion. I make no accusations--please note.

FWIW

Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

[Viv]

However--that said, I can't help but think there must be something else going on. It just doesn't seem right that their fabulous thrust (and other data) stand should repeatedly indicate an increase of thrust performance for multiple engines, compared to other (however primitive) measuring devices. Statistical interpretation of the results alone would indicate the need for further evaluation. "

Can we clear up this misconception first.

The thrust improvements are not down to the stand they are down to improvements in the engines, how many times does this have to be said?

We have been working with the pressure jets and due to the original design they were not producing maximum thrust, we have redesigned them and improved there performance and measured there new and old thrust accuratly.

So no more of "any engine reads more thrust on our stand" statements please, can we have instead any engine reads accuratly on our stand:-)

Viv

[Viv]

The problem here (as I see it) is there is no CONTROL for Viv and Luc's effort.

I would like to see them seek outside assistance in the form of an independent expert (or experts) who is willing to review their effort--with a very careful and skeptical eye toward trying to find something that might be wrong.

This would have no value to us poor pulsejet enthusiasts, on this Forum. The value would lie in providing Luc and Viv with solid correlation for their claims.

I appeal to logic and the scientific method for my discussion. I make no accusations--please note.

FWIW

Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

Outside expert! rofl:-)

Do you consider the national research council of Canada an outside expert with the nessecery expertise to validate our work? or how about the natural gas reasearch laboratory of Canada?

I think we can take the outside expert question as answered:-)

Viv

[Eric]

Did they run pulsating combustion engines on this device, and then compare the same engines on other accurate thrust measuring devices?

[marksteamnz]

Hmm questions that I hadn't asked and it may be to late now that lips are sealed.
How do you calibrate a dynamic load cell system? I'm familiar with the use and abuse of load cells for static, rate increasing and peak hold uses and we or the calibration company always slung weights on them to check calibration. So what is done for a dynamic load cell system (Google is a bit vague)
One of the reasons spring systems are bad is interacting resonaces can build up. But could that not also give a false "better" than actual static reading?

[Mark]

Some general stuff, click the topics on the left of the screen.
Mark
http://www.omega.com/literature/transac ... tory3.html

[Mark]

I remember taking a radio production course at WUWF which is located at the university here. For some reason I thought of all the different types of microphones, each one having a particular characteristic which best fits the need.
I guess measuring thrust or weight is kind of analogous to choosing the right microphone. There are more designs than you would think. I wonder if Graham might have something to say about the many types of microphones.
Waves and pressure,
Mark

[Mike Everman]

There is little doubt that Viv and Luc's set up is as good as it gets. Every thing else we try will pale in comparison. Believe it. It even had a high enough frequency response that they could tune the acoustics with the help of it's feedback. Given unlimited resources, I would do it exactly this way. Really, believe me or don't, this speculation about GLC's numbers is a wasteful diversion. The main point needs to be "how to get just about as good feedback without all that tech", IMNSHO.

Most other approaches suffer from sticktion and friction which make it very un-reliable to calibrate statically, then believe what you're looking at during high vibration.

Bill's setup has a great deal of friction that for the most part goes away in the face of the vibration environment. The mechanics are OK, but rather than attempt to calibrate a secondary readout statically, his force gage needs to directly couple to the load, in dead parallel fashion for the numbers to have meaning; not having cosine losses due to the angle of action. A spring/damper inserted between will eliminate the needle bounce and make a reliable average thrust reading. Still very difficult to calibrate out the line of action error.

A flexure or knife-edge based bearing system would be much better; insignificant friction allowing repeatable calibration statically.

My set up will not nearly have the frequency response of GLC's, but I will have a real-time SFC as an output, which is great for optimization too. Unfortunately I blew up the audio inputs on both laptops at Burning Grape, so acoustics will have to wait!

[steve]

It looks top me like larry's trapeeze thrust stand would probably be one of the most accurate and simple out there right now, and one of the cheapest. It should give an almost 100% accurate reading for average thrust produced if friction is minimized (suspend the platform from very thin cable perhaps?)

speaking of which- larry, did you ever finish that thing?
;-)

[hinote]

Most other approaches suffer from sticktion and friction which make it very un-reliable to calibrate statically, then believe what you're looking at during high vibration.

Bill's setup has a great deal of friction that for the most part goes away in the face of the vibration environment. The mechanics are OK, but rather than attempt to calibrate a secondary readout statically, his force gage needs to directly couple to the load, in dead parallel fashion for the numbers to have meaning; not having cosine losses due to the angle of action.

My response to this is, I'm pulling on the moving portion of the stand (that the engine is mounted on); the angular irregularities in the linkage are thus discounted because a given force is accompanied by a given deflection position--and the fact that the incremental values on the scale card aren't linear is of no consequence. Application of force with the gauge is accomplished using a guide, to eliminate any possibility of off-axis deflection. Friction is stimulated out of the system with a rubber mallet, and the deflection is measured with multiple observations, for consistency.

A spring/damper inserted between will eliminate the needle bounce and make a reliable average thrust reading. Still very difficult to calibrate out the line of action error.

I'm afraid to use any elastic medium as suggested, because of potential resonance. In fact, I've made every effort to overbuild all components to eliminate any flexing at all.

A flexure or knife-edge based bearing system would be much better; insignificant friction allowing repeatable calibration statically.

I estimate the system static friction at something like 2 lbf max, which is unacceptabe to me. I'm in the process of refining it so the value is consistently below 1/2 lbf. I believe frictional errors should be related as a ratio to the operating range the user desires; since I'm measuring thrust values from fairly large engines, even 2 lbf doesn't really cause any major error.

BTW a pulsejet makes an excellent instrument shaker, so static friction issues aren't really an issue in actual operation.


Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

[Viv]

There is little doubt that Viv and Luc's set up is as good as it gets. Every thing else we try will pale in comparison. Believe it. It even had a high enough frequency response that they could tune the acoustics with the help of it's feedback. Given unlimited resources, I would do it exactly this way. Really, believe me or don't, this speculation about GLC's numbers is a wasteful diversion. The main point needs to be "how to get just about as good feedback without all that tech", IMNSHO.

Most other approaches suffer from sticktion and friction which make it very un-reliable to calibrate statically, then believe what you're looking at during high vibration.

Bill's setup has a great deal of friction that for the most part goes away in the face of the vibration environment. The mechanics are OK, but rather than attempt to calibrate a secondary readout statically, his force gage needs to directly couple to the load, in dead parallel fashion for the numbers to have meaning; not having cosine losses due to the angle of action. A spring/damper inserted between will eliminate the needle bounce and make a reliable average thrust reading. Still very difficult to calibrate out the line of action error.

A flexure or knife-edge based bearing system would be much better; insignificant friction allowing repeatable calibration statically.

My set up will not nearly have the frequency response of GLC's, but I will have a real-time SFC as an output, which is great for optimization too. Unfortunately I blew up the audio inputs on both laptops at Burning Grape, so acoustics will have to wait!

Just an interesting question about the spring/damper thingy:-) if the wave form is a sine wave then yes I can see a reliable average but the waveform we have to deal with is any thing but a sine wave, nor does it have equal plus or minus amplitudes.

So what sort of an average value is a spring/damper or any other mechanical damper going to give? just looking at a series of cycles from an engine shows no two cycles are exactly the same or with a simple enough structure to average in a simple way.

The time constant of the damper will decide the output so what constant do you choose?

some one asked about calibration of a dynamic cell, easy drop some thing on it:-) as long as you know how heavy it was and from how high you droped it you can then work out how much force it generated when it hit the cell, simple high school physics:-) or you can use a calibrated force hammer and compare the two readings.

Viv the heratic:-)

[hinote]

Bill's thrust stand is the same principle as Larry's.

Ben:

That was true, when you saw the thrust stand in operation at Burning Grape.

However, I found the yaw stability inadequate for operation with larger engines--so I've changed the hanging parallelogram over to a trolley-on-track system, using ball-bearing wheels; the rising-rate weight arm and connector linkage remain the same.

Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

[Mike Everman]

Viv,
If the damper is good enough to steady the needle, then the gage will give the reading we want. Single cycle asymmetries are not relevant to this, and the moving mass in the gage is so low as to eliminate significant effect of it's dynamics.

My force gage just requires a few kinks in the welding wire I pull the trolley with to steady the needle. There is no possibility that I am reading more or less average thrust than is going on.

Dang, this subject is gathering voodoo status. It's just not.