Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

A Frequency of a spring?

  1. Nov 1, 2018 #1
    I need to calculate the frequency of a flat spring. I need to make reed valves for a pulse jet engine.

    I read online to use Hooks's Law but this is not working. F = force. x = the amount spring moves. k = spring constant.

    I have a flat spring made of .010" spring steel. Spring is .375" wide by 1.250" long.

    I have no clue what k is for flat spring steel?

    It seems to me Hooke's law does not tell me the frequency of a spring.

    If I tie a string to the ceiling that hands down 6' with weight on the end I can get the weight swinging and count the swings per min. No matter how much the weight is changed Hz is the same. If I change the distance of the swing Hz stays the same.

    If string is shortened to 3' frequency changes. Now I can do math to calculate frequency if string is 1.250" long.

    I don't think my experiment & math applies to metal that is spring steel.

    If I cut a sample piece of spring steel, clamp it in the vise, flip the end it is too fast for me to count the Hz I can hear is buzz like a tuning fork.

    NEXT the pulse jet engine is 40" long I can use the formula to find the frequency of a tube to get Hz of the tube. It has been 7 years since I worked on this project I don't remember the frequency of the pulse jet engine body which is a tube closed on the valve end.

    If the frequency of the tube is 180 Hz the spring steel reed valves also need to be 180 Hz also.

    I could probably use some type of electric device that listens to tuning forks to listen to my valve to tell me what the Hz is if there is suck a thing?

    After a lot of trial and error I have an engine that runs very well but maybe it could be better. Watch the YouTube video of my engine running. I have run this engine for many hours the valves have no damage and no signs of getting hot. This engine produces 20 lbs of thrust running on propane or gasoline. A mix of 80% gas + 20% diesel engine produces 24 lbs of thrust. Running on methanol 34 lbs thrust. Methanol + 5% water = 34 lbs thrust.

     
  2. jcsd
  3. Nov 1, 2018 #2

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I don't quite understand the problem here. Are you unable to measure the spring constant of your spring, or are you saying that you do not know how to find the spring constant of your spring?

    And once you find the spring constant, do you know how to find the natural frequency of the mass-spring system?

    Zz.
     
  4. Nov 1, 2018 #3

    berkeman

    User Avatar

    Staff: Mentor

    There are cell phone apps for that... :smile:
     
  5. Nov 1, 2018 #4

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Is there a mass attached to the spring or is it just like a reed, vibrating due to its own mass and its spring constant?
    I tried google and eventually found a search term that gives a number of hits that are sort of relevant. I used the terms "cantilever vibration" and this link was one that has what you need to know. There is a formula that gives the natural vibration frequency in terms of the dimensions of the spring and the Modulus of the material. Have a look at it an see if it makes sense.
     
  6. Nov 1, 2018 #5

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The OP is talking about a reed valve, which is, at its simplest, just a cantilever which vibrates. Its period is not 2π√(m/k) but a more complicated formula - found in the link I gave above. It would not behave linearly, I suspect, because of the valve action on the fluid it regulates. (more like a relaxation oscillator) We'd need much more detail from the OP. I don't quite see how the original post contains stuff about a coil spring and a mass . . . . . ? Perhaps the teacher hasn't caught onto what the OP actually has in mind.
     
  7. Nov 1, 2018 #6

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I agree that we went off on a tangent, but I was replying to arman777 since Post #5.

    Moderator's note: that discussion has been moved to https://www.physicsforums.com/threads/hooks-law.959131/

    Zz.
     
    Last edited by a moderator: Nov 2, 2018
  8. Nov 1, 2018 #7
    I don't know how to measure spring constant?

    I have to make my own spring from spring steel, how can i find k if spring is not make yet?

    Spring Hz has to hatch the Hz of the engine.

    I need a way to make a spring the correct Hz?

    There use to be online calculators for Hz of a closed 40" tube can not find 1 anymore?
     
    Last edited: Nov 1, 2018
  9. Nov 1, 2018 #8

    CWatters

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

  10. Nov 1, 2018 #9

    Nugatory

    User Avatar

    Staff: Mentor

    That's easy: measure the force required to bend the spring by a given amount. You will find that for sufficiently small deflections the relationship between force and deflection is linear; the constant of proportionality is the spring constant.

    What's not easy is calculating the natural frequency from that constant. If you were working with a mass attached to a spring of negligible weight, there would be simple and beautiful differential equation (google for "harmonic oscillator differential equation") for the frequency; but if the mass of the spring is a significant fraction of the total mass of the reed valve, the equation becomes hairy and ugly and intractable. You might find it in some handbook of engineering formulas... maybe... but I wouldn't count on it.
     
  11. Nov 2, 2018 #10

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    What spring are we actually discussing here? Is it a common helical spring or is it a leaf type reed spring? If you want results for a leaf spring then make one and measure the displace for a given load (a graph) and then work back from that to find the Modulus. Otherwise, use the available information about steel (the place you buy the steel from can tell you the grade it it).
    I reckon that an arbitrary choice of modulus value from a list of 'typical' steel grades will give you an accurate value to work with because there will be many other variables involved in your final design of a reed valve and the dimensions of the spring will need to be adjusted somewhere down the line. You can't rely on making just one spring and for it to work as you expected.
     
  12. Nov 2, 2018 #11
  13. Nov 2, 2018 #12
    This is a picture of reed valves made from spring steel these are flat springs. If you flip the tip end of the spring with your finger it does exactly like a tuning fork it goes back and forth at a certain Hz.. The engine runs at 291 Hz so the spring Hz needs to be 291 also.

    005712.jpg

    1-s2.0-S0140700716302304-jijr3392-fig-0001.jpg

    379-0809-parent.jpg
     
  14. Nov 2, 2018 #13

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Right. So the coil spring with a mass hanging on it is not relevant.
    Without the extra shaping near the anchor point, it would be fairly simple to calculate the frequency. As it is, you are in much deeper water if you expect to calculate the characteristic. The result will be somewhere between the simple rectangular case (a high frequency) and treating it as a short length of much more 'bendy' reed ( about 2/3 of the width, loaded by a long, massive piece that doesn't flex significantly but will have significant angular momentum, as well as linear movement. At what actual level are you treating this problem? You have to expect to do some experimental measurements.
    Also, I wouldn't expect the natural frequency of the reed to be exactly the same as the engine frequency because the motion is truncated each time the valve closes. It's probably not as critical, actually because there will never be any 'resonance' with only a half cycle involved each time.
     
  15. Nov 2, 2018 #14
    Here is instruction and animation video how a pulse jet engine runs. With spark plug on continuous spark air & fuel goes into the engine at the same time. Fuel air mixture ignites in the combustion chamber then blows out the exhaust. Internal air pressure goes from 0 to 9 psi in .003436 / 2 = .001718 seconds as exhaust blows out the exhaust pipe. About 1/2 way down the exhaust pipe internal air pressure drops to 0. As exhaust continues out the end of the exhaust pipe it sucks a vacuum inside the engine negative air pressure is about -8 psi that allows fuel & air to flow into the engine again. Once the engine starts the cycle continues automatic and the spark plug is turned off. Each new fuel air mixture is ignited by the hot burning gasses in the exhaust pipe. Engine starts very easy at 25% thrust then is throttled up to 99% throttle. This engine will not tolerate being over throttled on a test stand more than 100.8% but as the engine picks up speed on a airplane or missile ram air allows the engine to burn more fuel at 400 mph engine thrust has doubled and tripled. A speed indicator automatically increases fuel as the engine goes faster.

    When reed valves are tuned to 291 Hz to match the frequency of the engine high speed camera shows the reed valves open and close 291 times per second. The reed valves look like rockers on a rocking chair as they open & close.

    NOTE the engine does not run at exactly 291Hz it fluctuates plus & minus 2, 3 & 4 Hz = 287 to 295 Hz about 94% of the time. Sometimes it fluctuates from 286 to 297 and some times 1% of the time as much as 284 to 299 Hz but this is not much of a problem for the reed valves. If valve Hz does not match engine Hz fairly close engine will fire before valves have completely closed.

    Argus Reed Valves only need to open .186" where a STOP prevents the reed valve from opening more than .186".

    https://en.wikipedia.org/wiki/Pulsejet#Valved_pulsejets
     
    Last edited: Nov 2, 2018
  16. Nov 2, 2018 #15

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Where are you going with this now?
    Do you have the engine without the valves or is it all at the planning stage at the moment? If you use the link I gave at the top, you can design a reed that will oscillate at any frequency you want. The length and width are presumably fixed by the port size so you just need to choose a grade of steel (read around for the best choice) and that will lead to the thickness of a simple rectangular reed. You can adjust the natural frequency in the same way that they tune harmonica reeds - shave off the reed at the end to increase frequency and shave off at the base to lower the frequency.
    If, as you say, you want to restrict the movement, the period of the reed will not be the inverse of the frequency. It's going to have to be more suck it and see than you seem to be hoping but that's no problem for a practical project like this one.
     
  17. Nov 2, 2018 #16
    I did not see the link above. That will be helpful. Like a tuning fork reed valve Hz does not change with shorter or longer stroke. I want to build a larger engine.
     
  18. Nov 2, 2018 #17

    Mister T

    User Avatar
    Science Advisor
    Gold Member

    All you have to do is play that same buzzing noise on your phone (using an app) while listening to the piece of metal vibrate. Adjust the frequency of the buzz on your phone, matching it by ear. Humans are quite good at this skill, especially when we learn how to listen for the beats.
     
  19. Nov 3, 2018 #18

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    If you already have some working reed valves, you can use the same design and just increase the number of ports. Presumably the engine will have a bigger cross sectional area so there should be room for them. I think you just to need some hands on with this problem. See if you can make a replacement working valve for the existing engine first.
     
  20. Nov 4, 2018 #19
    I was thinking the same thing but not sure that will really work I don't want to waste a lot of machine time building something that does not work. The way I originally designed the best working reed valve was to make sure 8 psi vacuum will suck them open. Metal thickness has to be .010" thick for the valve to not, chips, crack, break off, or bend.

    I already have several good running engines, 5 lbs of thrust up to 100 lbs of thrust. Time to build a larger engine.



    '
     
    Last edited: Nov 4, 2018
  21. Nov 4, 2018 #20

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    I think you need a bit more confidence with this project.
    So why not try making a design that you already know will work? Have you not made reeds before, for your existing engines? I don't understand the back story of this.
    You can buy 10thou shim material so you only need to make up a stamp (or you could even use a chisel on a block of hardwood) to mimic the shape you know will work. It strikes me that you expect to do some calculations and end up with a working engine. That is just not realistic; you need to progress from a point that is achievable now. You won't find anyone who will give you the right dimensions for your reeds. There is very little machine time needed for the actual reed cutting. Making the ports the right size and shape is far more time consuming.
    Also,
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted