Torsional Vibration and Resonance - Basic Theory and Issues

Discussion in 'General Auto Conversion Discussion' started by wsimpso1, Nov 20, 2012.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Apr 20, 2019 #61

    poormansairforce

    poormansairforce

    poormansairforce

    Well-Known Member

    Joined:
    Mar 28, 2017
    Messages:
    512
    Likes Received:
    105
    Location:
    Just an Ohioan
    Yes, there would be work involved but the key thing is to change the tension significantly over a very small travel length. Regular springs cannot do this. This gives me an idea I think might work and it would be so much easier to implement with zero special machining, parts, etc. IOW, off the shelf parts.
    Ever since setting up my son's 50 hp motorcycle engine in his go kart using 2 chains and a jackshaft I have been thinking about this issue.
     
  2. Apr 20, 2019 #62

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    Or harder because the catalog sizes were far off of ideal. I suspect that they had already worked out about what load vs travel and total capacity they wanted, and had to play with number of balls and pocket radii to get it. Still requires significant experiementation that is justifiable for a commercial product, but way too much for a one-off.

    Billski
     
  3. Apr 20, 2019 #63

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    Actually, the deal here is to both have enough travel to deal with normal thermal expansion and wear while also having enough travel to let the engine go through its decel without loading the rest of the system in coast. I suspect I could design coil springs to do that, but those balls will have a rising rate, which can let the whole thing be more compact and easy to build while adding some hysteresis, which are all advantages over coil springs...

    Billski
     
  4. Apr 20, 2019 #64

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    5,976
    Likes Received:
    1,994
    Location:
    Rocky Mountains
    First thoughts on using a hydraulic piston to tension the belt - in addition to a soft spring - similar to the some automotive tensioners?

    Once the engine is above the soft/low speed stage the hydraulic cylinder then takes over the job. It is also self adjusting just like a hydraulic valve lifter.
     
  5. Apr 20, 2019 #65

    poormansairforce

    poormansairforce

    poormansairforce

    Well-Known Member

    Joined:
    Mar 28, 2017
    Messages:
    512
    Likes Received:
    105
    Location:
    Just an Ohioan
  6. Apr 20, 2019 #66

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    If those pins are smaller diameter than the holes in the plate, and the pins can roll in the holes, that is an order absorber. That very type has been applied to radial engines and was available for V8 engines in a product called The Rattler. http://www.enginehistory.org/NoShortDays/TV.pdf If it is applied to electric motors, it is probably tuned to absorb the cogging frequency of the motor, will work across the speed range, and will be quite effective about reducing the magnitude of cogging.
     
    poormansairforce likes this.
  7. Apr 21, 2019 #67

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    5,976
    Likes Received:
    1,994
    Location:
    Rocky Mountains
    The stepper motor speed/freq changes a lot during the cut. Controllers do micro-stepping as well. The critical speed of the ball screw shaft can cause problems on machines that use that kind of drive system. When I designed my unit it was the factor that limited the travel speed. Now I see a way to change that..........On the next build.:cool:
     
    poormansairforce likes this.
  8. Apr 21, 2019 #68

    poormansairforce

    poormansairforce

    poormansairforce

    Well-Known Member

    Joined:
    Mar 28, 2017
    Messages:
    512
    Likes Received:
    105
    Location:
    Just an Ohioan
    Yes, that is its intent. The pins rattle around to change the tuning of the instrument, so to speak.

    There are videos showing the before and after effect. This one shows it in operation:
    https://www.google.com/url?sa=t&source=web&rct=j&url=#&ved=0ahUKEwj16ou6-d_hAhUFL6wKHc08CZUQxa8BCCwwAw&usg=AOvVaw0VTJzWxrN8ZFOD4xbU2nov

    They also make the collar ones as well.
     
  9. Apr 21, 2019 #69

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    If you have a motor that cogs at x times per revolution and the absorber is tuned to oscillate at x times per revolution, it absorbs that order vibration, as long as it is turning fast enough for the weights to be thrown "out" in their recesses. It does not care what the frequency is...

    Critical speed of a shaft is a different animal completely...

    Billski
     
  10. Apr 21, 2019 #70

    poormansairforce

    poormansairforce

    poormansairforce

    Well-Known Member

    Joined:
    Mar 28, 2017
    Messages:
    512
    Likes Received:
    105
    Location:
    Just an Ohioan
    Billski, I'd like to ask a question. What are your thoughts on chain redrives? Easier? Harder? About the same? I've been curious with my experience with the go kart and so many things use a chain drive. With so many individual elements that make up a chain I wonder about the (im)possibility of having a consistent resonance at each point in each revolution? Hope that makes sense.
     
  11. Apr 21, 2019 #71

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    5,976
    Likes Received:
    1,994
    Location:
    Rocky Mountains
    Ok, a bit more education?

    This video shows a stall of the stepper that, from what I understand, is directly related to the critical speed, or torsional compliance of the shaft. What is happening here is that the stepper advances one full step, but the system being in resonance has enough force to snap the stepper back one "tooth" to the original position.
    My understanding (which may be wrong?)is that while the critical speed of the shaft changes as the gantry is moved, due to the change of the distance between the 3 supports, the torsional frequency of the shaft remains relatively constant. Both phenomena are related to the same shaft variables?

     
  12. Apr 21, 2019 #72

    Ray Jacobelli

    Ray Jacobelli

    Ray Jacobelli

    New Member

    Joined:
    Apr 16, 2019
    Messages:
    1
    Likes Received:
    0
    VW PSRU a Very well thought out design - my primary concern is the support bracket mounting to the VW case. IMO there is too much induced moment between the studs securing the bracket and the center-line of the driven pulley. I understand there are limits to what can be done with a VW case.
     
  13. Apr 21, 2019 #73

    Vigilant1

    Vigilant1

    Vigilant1

    Well-Known Member

    Joined:
    Jan 24, 2011
    Messages:
    3,550
    Likes Received:
    1,527
    Location:
    US
    And limits on what can be done with VW Type 1 heads. Many people have tried to use a redrive to allow the engine to turn faster and develop more power while turning a reasonable size prop at reasonable top speeds. The issue is that the heads can only dump so much heat, generally limited to burning about 6 gph or making about 75 continuous HP. But using a redrive just to turn a larger prop within the hp limits of the VW should be no problem.
     
    pictsidhe and akwrencher like this.
  14. Apr 21, 2019 #74

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    They have been successfully done. Understand that many millions of front drive automatic transmissions and 4WD/AWD transfer cases use silent chains. Understand also that there is a soft and/or hydrodynamic drive element between the engine and the the next big inertia that drives lowest resonant vibration down below idle speed of the engine. The silent chain is down stream of that isolator and thus they see only modest input vibration. Like gears, they do make vibration at the tooth passing frequency, which is pretty high, so they generally do not pass much to the rest of the system either. They have to be well engineered, they wear, the sprockets have to be precisely spaced and heat treated to fairly well match the chains. Big Block Ford engines built on aluminum racing blocks and heads were being sold with a chain drive by Fred Geshwender, but has been gone awhile. The German word for RPM is geshwenderhiet. My German friends could not shed light on the orgin of the word. More here...
    http://users.owt.com/worden/alternateairpower/
     
    Last edited: Apr 21, 2019
  15. Apr 21, 2019 #75

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
  16. Apr 21, 2019 #76

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    This borders on thread drift, as CNC drives are not airplanes, but OK...

    Order absorbers are OLD news. I already cited the pioneering work done in the 1930's here http://www.enginehistory.org/NoShortDays/TV.pdf.

    The style shown in the video is the hockey puck style that presaged the various swinging counterweight styles that are still in use today on many four and six cylinder Lycoming and Continental engines. When the pendulums of whatever sort are loaded primarily by centrifugal forces, the natural frequency of the pendulum varies linearly with speed. So, tune it to oscillate at 2rd order of rotation and it sucks off 2rd order vibrations at all speeds. In detail, the motor or crankshaft is spinning, the weights are fully out in their pockets, the engine accels and decels 2 times per rev (4 cylinder even firing four stoke engines do this) and the pendulum lags during the accel and leads during the decel in time with firing. These are now being implemented in more and more automotive powertrains (mostly diesels and high powered four bangers)...

    If the problem in the video is natural frequency of the motor inertia on a springy shaft being excited by the pulse timing coming from the controller during startup, the rollers are only being loaded by gravity to fall down in the cavities. The natural frequency of the rollers when only influenced by gravity will have to be in the neighborhood of the natural frequency of the motor/shaft and/or the controller pulse frequency for this to work. That the man in the video with DIY CNC hit an appropriate swing radius (0.75 mm in his case) was either pretty good fortune or he knew more about what radius to use than he said... Once they are spinning at speeds high enough for centrifugal forces to take over, the device will be an order tuned absorber as discussed above. If they were tuned to absorb cogging of the motor at speed, they will do it. If instead they are tuned (by adjusting swing radii) to roughly coincide with the startup pulse rate, they will most likely NOT also coincide with the cogging. There are ways that will work to get both if needed. Lots of math in that.

    Billski
     
  17. Apr 21, 2019 #77

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    5,976
    Likes Received:
    1,994
    Location:
    Rocky Mountains
    Thanks for the info. Still attempting to to sort this part of harmonics out in my mind.

    I don't think it's really thread drift. A better understanding of this is as much a part of HBA as understanding the chemistry of adhesives or the absorption spectrum of paint. If we do end up going down the motor/battery path in large numbers then we may eventually have to deal with this kind of resonance problem.
     
  18. Apr 21, 2019 #78

    wsimpso1

    wsimpso1

    wsimpso1

    Well-Known Member

    Joined:
    Oct 19, 2003
    Messages:
    5,622
    Likes Received:
    2,866
    Location:
    Saline Michigan
    Long before we had CNC, we knew about critical speed of a shaft and that is what I thought you were talking about above. Big issue in transmission and driveline design of all types. https://en.wikipedia.org/wiki/Critical_speed As the wiki page describes, the excitation is simply whatever imbalance exists in the shaft when spun at the same frequency as fundamental bending frequency of the shaft.

    The stepping frequency could interact with the shaft whirling around like a jump rope, but it is hardly necessary for the phenomena to occur. In long drivelines, like driveshafts on trucks and for helo tail rotors, the standard response is intermediate support bearings and frequently Hooke joints near the supports. In a CNC, if you were spinning your lead screws that fast, you can always go up to the next diameter step or commission ball screws that have been gun drilled.

    The use of a tuned order absorber will help with a startup stall or when cogging frequency reaches some other resonant mode in the system. I kind of doubt it will help much at all on a shaft in jump-rope mode.

    Billski
     
  19. Apr 22, 2019 #79

    poormansairforce

    poormansairforce

    poormansairforce

    Well-Known Member

    Joined:
    Mar 28, 2017
    Messages:
    512
    Likes Received:
    105
    Location:
    Just an Ohioan
    Agreed, that's why I'll use a belt drive on the next CNC I build. It's good enough for my needs.
     
  20. Apr 22, 2019 #80

    Hot Wings

    Hot Wings

    Hot Wings

    Well-Known Member HBA Supporter

    Joined:
    Nov 14, 2009
    Messages:
    5,976
    Likes Received:
    1,994
    Location:
    Rocky Mountains
    This may be where I'm not 'in tune" with the theory? I've thought that an unbalance in the shaft was a given. I further assumed that since the definition of the exciting force is an angular frequency that the result was a torsional resonance. Apparently a bad assumption on my part.
    I'm familiar with the 'jump rope' deflection but just always thought it was a secondary phenomena of the torsional resonance, not the bending resonant frequency being excited by an angular input. I can visualize how that may happen.
    So it would appear that an observed 'jump rope' deflection can have 2 causes? The first being a simple out of balance force that bends the shaft, deflection increases with increasing speed, and a second type that is related to bending resonant frequency, not raw speed?

    Mechanical seems to have a lot more diversity then inductors, caps and resistors.
     

Share This Page

arrow_white