Vibration Damping

Table of Contents

Vibration Damping and Isolation of the Flight Control Boards



Why vibration damping is important:

  • The APM and PX4 Flight Control boards have accelerometers built into the board that are sensitive to vibrations.
  • APM:Copter (the firmware) combines the data from these accelerometers with barometer and GPS data to calculate an estimate of its position.
  • This position estimate is critical when in AltHold, Loiter, RTL, Guided, Position and AUTO flight modes.
  • With excessive vibrations, the estimate can be thrown off and lead to very bad performance in these modes (i.e. the copter becomes jumpy or flies off).

Measuring your vibration levels:

  • Vibration needs to be less than 0.3 G in the X and Y axes and less than 0.5 G in the Z axis.
  • You should strive to get in the region of + and – 1/10 G in all axes, the information provided here will generally achieve that.
  • Please refer to the Measuring Vibration wiki page for details of how to measure whether your copter’s levels are within the acceptable range.

Check frame motors, props and prop adapters:

For copters vibration comes mostly from frame flex and motors and propellers and often you can reduce vibration significantly:

  • Frame flex especially arm flex is a big cause of asynchronous vibration, Frame arms should be as rigid as possible.
  • The real DJI Flamewheel copters have sufficiently rigid injection molded arms, the many off brand clones do not.
  • Generally carbon fiber armed copters have sufficient anti twist and anti-bend characteristics.
  • Heavier aluminum armed copters like the Arducopter are flex free, cheaper Chinese ones often not.
  • Most purpose designed commercial copters with injection molded exoskeletons or arms like the Iris or Phantom are sufficiently rigid.
  • Cheap, light frames tend to flex more than high quality stronger ones and the heavier you load the copter the more “Flexi” it gets.
  • Motor to frame arm and frame arm to central hub mounts need to be secure and flex free (sometimes a problem for carbon tube arms).
  • Motors need to run smoothly (bearings not worn-out or “screeching”).
  • Prop adapters connecting the propellers to the motors need to be concentric and very straight.
  • Propellers should be fully balanced using a good manual prop balancer
  • Motor balancing (or really well factory balanced motors like T-Motor) can have a major effect.
  • Propellers that are not well matched to the frame and weight or do not have the same flex for CCW and CW are very problematic.
  • Get Good propellers, the new APC multirotor propellers are a good choice as often are GemFan carbon filled propellers.
  • Carbon Fiber propellers are best, but the main company, Graupner has just gone out of business and cheap CF props are problematic.
  • And carbon fiber props are expensive, rigid strong and as sharp as a razor so they are a major safety hazard.
  • Large slow propellers will definitely induce more vibration than small fast ones, but the big slow props are a lot more efficient.
  • And large slow propellers will also make frame flex a lot worse as the motor units will twist around the props axis from unbalanced lift.
  • If you really optimize all of the above characteristics, your flight controller will likely need only a bit of Gel or foam for vibration isolation.

Isolate the APM/PX4 from the frame:

  • Double sided foam tape or Velcro has traditionally been used to attach the APM/PX4 to the frame.
  • In many cases foam tape or Velcro does not provide adequate vibration isolation because the mass of the flight control board is so small.
  • Below are 4 possible alternatives that have been tested and shown to perform better:

Dubro Foam

  • 1/4″ dubro foam can be purchased directly from DuBro or from other RC retailers.
  • Use Krazy Glue (or equivalent) to attach the foam to one of the levels of your frame.
  • Then glue the bottom of the APM/PX4 case (containing the APM/PX4 of course) directly to the foam.
  • The Glued on Foam method means it will be difficult to remove the flight controller.
  • You can purchase extra cases or frame parts to allow you to swap the flight controller between frames or use one of the other methods.
  • The APM 2.5 and 2.6 and Pixhawk case can be unscrewed from the top and the controller swapped to another glued down case back.
  • Unfortunately APM 2.0 case screws are on the bottom making swapping using this method problematic.

IMG_0267


Gel pads

1. Put off-the-shelf vibration damping gel pads of 1/2″ ~ 1″ (1cm ~ 2cm) on each corner of the flight control board.   Possible gels include:

2. Secure the board with a 1/4″ ~ 1/2″ (~1cm) wide velcro retaining strap or a rubber band.  Be careful the strap does not hold down the controller so securely that it interferes with the damping of the pads.  Consider putting a layer of soft foam between the strap and the flight controller.FlameWheel F330 with Kyosho Zeal Pads

FlameWheel F330 with Kyosho Zeal Pads


FlameWheel F330 With PX4 Mounted on Intermediate platform

FlameWheel F330 With PX4 Mounted on Intermediate platform


FlameWheel F330 With PX4 on Zeal Pads with Protective Hard Top

FlameWheel F330 With PX4 on Zeal Pads with Protective Hard Top


PX4F330accels_5_9_13The Zeal Mounted FlameWheel has Plus and Minus one vibration all axes = 1/10 G

This video demonstrates vibration isolation using Moon Gel on a Go-Pro camera (Here!).


O-ring Suspension Mount:

  1. Create a platform upon which to mount your flight control board with holes or screws on the four corners.  Mount your flight controller on this board with double sided foam tape.
  2. Mount 4 standoffs on the top of your frame spaced 1/10″ to 1/8″ further apart than the width of the platform upon which the control board has been mounted.
  3. Insert 1/16″ nylon O-rings through each corner of the flight control board and the standoffs so that the flight controller has no hard connections to the frame.  Link (Here!)
      • The overall O-ring diameter should be chosen to firmly retain the board while providing for light to moderate initial but rapidly snubbed movement of the board (generally 1/2″ to 3/4″ OD) and Silicone O-rings should generally damp better than Buna-N O-rings (Sizes 15 – 21) if you can acquire them.

FlameWheel F450 O-Ring Suspension Platform Mount

 FlameWheel F450 O-Ring Suspension Platform Mount Raw Accel Values

FlameWheel O-Ring Suspension MountGraph of vibration levels shows they are under 0.1G on all axis

Warning! If you are mounting your flight control board to the power distribution board it might be better to mount the standoffs for the Flight control board to a separate piece of fiberglass cut to size that can subsequently be bolted through existing holes in the power distribution board. Only drill through the power distribution board for mounting the standoffs if you are completely certain you will not cause a short and use threaded nylon machine screws or standoff studs.

Vibrations are short coupled, so all that leaving excess corner clearance does is to require higher initial O-ring tension which reduces vibration damping responsiveness and allows the board to physically tilt more (which is undesirable as it throws the sensor to airframe relationship off).

The disadvantage to O-ring suspension versus Gel pads is that it is mechanically more complex and it requires tuning of both of O-ring diameter and cross section.

You can combine O-ring and gel pad design by using an intermediate plate and benefit from dual rate damping.


Ear Plug Mount

  1. Purchase slow response silicon or urethane foam or PVC foam earplugs such as these from 3M
  2. Create a platform upon which to mount your flight control board with holes at the four corners.  The holes should be large enough to allow the ear plugs to be inserted into them but not so loose that the board comes loose during hard landings.  Ensure the holes are smooth so they do not cut into the ear plugs.  Also keep the holes near the corners of your electronic module plate as possible to minimize unnecessary module movement.
  3. Mount your flight controller on this board with double sided foam tape.  Extra mass added to the board may improve vibration damping.
  4. Squeeze the earplugs through existing holes in the frame (or cut new holes) and the holes in the board upon which the flight controller is mounted.  “Tuning” is possible by varying the amount of earplug left exposed in the middle.

Vibration_EarPlug

Ear Plug Vibration Mount

Example Ear Plug vibration reducing mount


Bulb Damper + Ear Plug .05G Ultra Low Vibration Mount

  1.  Robert Lefebvre developed this extremely high performance mount which actually performed way better than expectations.
  2. It consists of a mounting plate with a 100g soft rubber bulb type “gimbal” damper at each corner and a half a urethane foam earplug placed inside each one.
  3. Gimbal bulb type dampers themselves can work in tension or compression.
  4. The earplugs provide an additional damping medium with a different frequency damping range than the bulb dampers by themselves.
  5. The ear plugs also stiffen the bulb mounts up a bit preventing excessive free motion being caused by normal flight maneuvers.
  6. This was successful at damping a Flamewheel clone with flexible arms and over size 12″ propellers into the .05 G range.
    DSC01059
  7. The APM flight controller is also mounted on anti-vibration grommets available from McMaster Carr package of 25 each part #9311K64 recommended.
  8. This was successful at damping a FlameWheel clone with flexible arms and over size 12″ propellers into the .05 G range.
    F450 Quad Vibes 1
  9. Note that the X & Y vibrations are less than + and – 0.05G, the Z vibrations are a bit higher due to Flamewheel clone arm flex, high vibration motors and the inordinately large 12″ propellers.
  10. Robert describes this Copter as a real “paint shaker” due to the inadequacies listed above making the results acheived even more remarkable.
    DSC01056
  11. The 100G bulb type gimbal vibration dampers can be ordered direct from China from a variety of vendors: foxtechfpv
  12. This can be considered to be an excellent solution for many types of airframes as it seems to have a very wide dynamic range.
  13. Clearly some fine tuning would be possible by varying the type and size of the earplug section in the dampers.
  14. There are also 200G and 300G dampers but those would be best used on higher mass modules perhaps including a battery.

An Excellent 3D Printed Anti-Vibration Platform

By GuyMcCaldin with the original Blog article here: http://diydrones.com/profiles/blogs/3d-printed-anti-vibration-mount

Over the last few weeks, I’ve been prototyping different anti vibration mount designs using an Up Mini 3D printer.  I wanted something that performed well, using affordable and readily available components.

The first design started with anti vibration dampeners in a simple vertical configuration.  This suffered from too much play in the horizontal plane, which might have caused instability in quick accelerations.  The next design rotated the dampeners 45°, which resulted in much more even resistance across the horizontal and vertical planes.

The large rubber dampeners provide outstanding isolation.  Too much in fact.  The ideal anti vibration mount isolates the controller from high frequency vibrations, but conducts lower frequency vibrations that might represent small changes in attitude.  For the next design, I moved to smaller and slightly stiffer dampeners available from Hobby King.

I’ve been testing them on a Turnigy Talon tricopter, using an APM 2.5.  The GPS is mounted to the top of the APM to increase the moving weight, which assists slightly in reducing higher frequency vibrations.

The results so far suggest that it’s working very well:

The above graph is taken from a two minute hover in loiter mode.  All axes are below ±1g, which exceeds the specifications set out in Arducopter guidelines.

Using a tool (Link) that Forrest Frantz developed, we can see these results from another perspective:

The Acceleration results are still excellent, but the pitch and roll stability shows a bit of oscillation.  I’m using default PID values which would definitely benefit from tuning, and flying in slightly windy conditions.  I’m also new to rotary wing flight, so I’m still honing my flying skills.  I’m able to keep the tricopter in a 1m x 1m x 1m cube, but it slowly drifts around if there is a change in wind.

These are the most likely causes, but in the interests of full disclosure, I wanted to mention that it could be caused by the APM not responding to small changes in attitude if the mount is creating too much isolation.  It certainly doesn’t appear to be affecting flying performance, the tricopter is impressively stable in loiter mode.  I won’t know if it’s an issue or not until more people try Forrest’s tool out, so I have more data to compare my results to (you can find the forum thread (Here!)

If you have access to a 3D printer, you can download the STL files here: Omnimac APM Mount v1.5.stl

Alternatively, I’ve uploaded the design to Shapeways.  You can purchase it starting from $22.92 depending on material:

I’ve include a 30% margin in the price (~$5) that will go towards a trip to Africa in November, where I’ll be donating time and resources using 3DR powered UAS in wildlife research and conservation.

To complete the mount, you’ll need these Dampers which cost $1.20 from Hobby King.

The mount can be installed using double sided tape, or M3 screws spaced at 45mm x 45mm.  If you’d like me to modify the file to suit the mounting options of your airframe, just let me know.

The log file that was used to produce the results shown above can be downloaded from here: Turnigy Talon 2min Loiter (default PIDs)


Additional Vibration Reduction Considerations

  1.  These grommets can provide sufficient or supplemental vibration reduction: Hard Disk Drive Anti-Vibration Grommets
  2. Significant gains in vibration isolation can also be realized by using a high flex wire and strain relief approach to all wires connected to the Flight control board (and using the minimum number of wires necessary as well.)
  3. Some frames have lower than normal vibration characteristics due to frame stiffness / flex and isolated centralized mass can greatly influence motor/prop vibration transfer to the central fight controller.
  4. Isolation and damping can be improved somewhat by sandwiching the flight control board / enclosure between damping pads on both sides in about twenty percent compression. 30 durometer Sorbothane is actually specified at 15 to 20 percent compression for optimal damping.
  5. Although 30 durometer Sorbothane seems an excellent candidate, personal experience indicates that it becomes permanently compressed and is not as effective at vibration reduction as the Gel solutions.
  6. A link to a Blog about the first APM anti-vibration mounting system to achieve 0.05 G damping (2/20/2013 improved to 0.02 G), a dual zone isolation system, combining Oring suspension and silicone pad is (Here!)
  7. Motor balancing can also reduce vibration and especially so for cheaper or larger motors.  Balancing involves:
    • Tightly fasten a small tie wrap around a motor (WITH NO PROP), trim off the extended tab and spin it up.
    • Try multiple times, each time turning the tie wrap on the motor housing a bit until the vibration reduces or goes away.
    • A small piece of Scotch tape can be re-positioned instead of the tie wrap if desired or for smaller motors.
    • When you locate the spot where there is the least vibration (and you should be able to hear it), mark the spot directly under the clasp of the tie-wrap with a felt pen.
    • Add a small dot of hot glue gun glue where the Tie-Wrap clasp was and increase the glue a bit at a time till the vibration is minimized.
    • If you put too much glue on it can be removed with an X-acto knife.
  8. Vibration damping motor mounts like this or this may or may not reduce vibration.
  9. Camera Mounts also need to be effectively isolated and damped from vibration, but they already have a number of “soft” mounting solutions.
  10. The camera servos need to be vibration isolated as well, either in the isolated camera mount itself or with their own vibration reduction solution.
  11. You should use high quality ball joints on your camera servo arms and adequate bearings or bushings in the mount itself with zero free play to prevent inertial slop.
  12. Quality servos without free play are also a must for precision camera work.
  13. At this point in time it seems that the more rigid the frame the better because frame flex introduces undesirable mechanical delay (hysteresis) in translating motor induced actions to the centrally located flight control board. (Do NOT shock mount the motor Arms).
  14. The amount and type of damping medium needs to be carefully matched to the weight (mass) of the item we are trying to isolate as well as the frequency and amplitude of the vibrations we are seeking to damp.  We are trying to isolate a flight control board that weighs less than 2 ounces and this is a very small mass.
  15. Virtually all off the shelf solutions (either pad or stud type) are designed for an isolated mass that would weigh at least 5 to 10 times what an APM or PX4 / IO board weighs for optimal effectiveness.  This includes all pre-made Sorbothane, Alpha gel, EAR, memory foam or other silicone or urethane gel or foam mounts as well as Lord Micro mounts.
  16. A threaded stud or sleeve type mount gel mount properly designed for the mass of our flight control board or electronics module undergoing the stress’s of normal flight would be a much better long term solution.

A Summary of the particular vibration characteristics we need to damp:

  1. The vibration frequency and amplitude we primarily need to reduce is a characteristic of the motor / prop units turning at flight speed.
  2. That is, it is a fairly high frequency with fairly low amplitude.
  3. This requires that we provide a short coupled damping and isolation range.
  4. The board itself does not need to have nor benefit from a range of motion that exceeds the amplitude of the vibration.
  5. Because the board does not apply any force to the airframe, the only thing we need to be concerned with Damping / Isolating is the weight (mass) of the board itself plus the forces applied to it by airframe’s normal flight maneuvering.
  6. Since excellent broad frequency range, high damping materials are available our biggest concern will be to use the proper amount of them to optimally damp our flight control board. (Too much is just as bad as too little.)
  7. Combining the Flight control board and receiver onto a separate vibration damped electronics module “plate” or enclosure can increase the mass of the module making it easier to damp effectively as well as reducing the interconnecting wiring and making the whole system more modular.

Note on terminology:

Isolation is simple undamped (spring or rubber band support) which allows the movement of the isolated object largely separate from the containing object (Automobile spring for instance).
Damping is the conversion of vibration into heat energy by a shock absorbing medium (Automobile shock absorber for instance.)
Our ultimate goal here is to provide the most high and medium frequency reduction while still allowing low frequency actual board movement to take place in concert with the airframe with a minimum of delay so realistically our methods incorporate both damping and isolation.


Links to related discussions:


Questions about this page? Comments? Suggestions? Post to APM Forum! Use the platform specific to your query, and make sure to include the name of the page you are referring to.

Comments are closed.