Vibration Ruins Video
Getting clean video with a multicopter can be a difficult and frustrating task. The reason, of course, is that slightly out of balance motors and propellers cause significant amounts of vibration. If you’re only taking stills, then you can use a high shutter speed to get a clean picture. So with stills you can tolerate a certain amount of vibration without affecting picture quality.
However, with video you usually don’t have the option of using high shutter speeds. And video cameras and DSLR’s that have CMOS sensors are particularly sensitive to vibration. Even after you’ve bought the best motors and props, there’s still vibration that needs to be damped out.
Most often, it’s a trial and error process to get good, low vibration. You add some damping between the camera mount and the helicopter, go out and shoot some video with the new damping, and then process the video to see how good it looks But it’s often hard to judge whether the change you made is heading in the right direction. You may have lowered the vibration a little, but you can’t really tell when the video hasn’t changed much. You have no accurate reference against which to measure. You may need to do many experiments just to see if you’re damping material is going to work out
Wouldn’t it be easier and quicker if there were a way to measure the vibration when you spin up the helicopter and “see” even a slight improvement? That’s what vibration meters are for. But vibration meters are expensive and heavy. Or are they?
Measuring Vibration with Your iPhone
I’ve found two iPhone apps that turn your iPhone into a vibration meter. These apps work extremely well. They remove much of the guesswork and trial-and-error experimentation. One of these apps is called “Vibration” and the other is called “Seismometer.” Here are the websites for Vibration and Seismometer.
Though they both measure vibration, these two apps are quite different. Here are some points of comparison.
1. Vibration measures vibration in all three directions: x, y, z. It gives you 3 separate graphs, capturing the vibration in each direction. On the other hand, Seismometer gives you a one dimensional graph that captures the RMS energy in all directions. It might seem that one dimension would be limiting. But I found it to be quite good enough (see example below).
2. Vibration will only sample for 10 seconds. You can delay this sample up to 100 seconds. Seismometer, on the other hand, starts sampling as soon as you hit the button, but continues until you press the button to stop. So, it will sample indefinitely. As you can imagine, Seismometer is more convenient to use, because you can start it and then take off, without worrying that it’s going to measure vibration at the wrong moment. With Seismometer, you can monitor the entire flight.
3. The little red dot that tells you that you are recording with Seismometer is so small that it’s difficult to see. Vibration, on the other hand, has big countdown digits as it prepares to start a recording. You can’t miss it.
4. Both tools let you email yourself a CSV (comma separated values) file with the recorded data.
5. Vibration lets you take a screen shot. Whereas, Seismometer saves each data file in your phone as long as you like. You can go back and look at them at any time. Vibration let’s you email it anywhere you’d like. However, Vibration only stores the current sample until you email it. Then it’s gone from the phone.
6. You can’t name your files something meaningful with Seismometer. Seismometer names them with the time and date. However, with Vibration you can name the file anything you like.
7A. To use Vibration on a helicopter, you have to strap the iPhone in, set the delay and then push the “Sample” button. I usually set a 20 second delay, which gives me time to spin up the helicopter and get it into a hover before Vibration starts its measuring. This is a bit tricky, but you get used to it. As far as I can tell, neither tool allows you to attach a note to a file to give it a meaningful description.
7B. To use Seismometer, you just push the start button, step back and spin up the helicopter. It records the whole time. When you land, you just turn off the recording. While Seismometer is easier to use than Vibration, for the most part, it does have one problem. The little red “dot” that tells you that you are recording is extremely small and difficult to see when your iPhone is strapped into a camera mount. This “dot” is barely the size of a period. More than once, I thought I was recording, only to discover after landing that I hadn’t pushed the record button hard enough.
8. Vibration is $4.99 at the Apple Store. Seismometer is $0.99. Both are bargains at these prices.
Tuning a Multicopter with Seismometer
In order to get clean videos from my hexacopter (6 motors, 6 propellars), I tried many different props (discuss different props and vibration). After balancing, I was still getting a bit of vibration. This was probably the result of imperfect balance and slightly out of balance motors. I thought that putting some rubber dampers between the helicopter and the camera mount might help. I had just purchased Seismometer, so I decided to measure the vibration at the camera mounting plate before adding the rubber dampers. Here’s the setup.
Here’s a picture of my iPhone strapped on to a Photoshipone MktrPro camera mount for testing.
Here’s a screen shot of the Seismometer output during a fairly tight hover.
After adding four 10mm rubber damper between the camera mount and the bottom of the helicopeter, here’s the resulting vibration. I put the helicopter into a similar hover as in the first sample. In this configuration, the camera mount is hanging by the rubber dampers (picture). I found a similar level of damping with 8 rubber dampers, which safely supports more weight.
As you can see, the rubber dampers lowered the vibration considerably. Seismometer displays with a log scale. To my eye, it looks like about a 5x reduction in vibration amplitude.
Don’t Forget the Propellers
After adding the rubber dampers, I was still getting some jitters on cameras with CMOS sensors. I had been using APC propellers. These are inexpensive, composite props, which I balanced using a static balancing tool. I was using 12 x 3.8 slow flyer props. These props have considerable flex. This can create excessive vibration under heavy load (especially when descending into the prop wash, etc.). I switched to Xoar wooden props at roughly the same size, 12×5. The Xoar props are much stiffer and don’t tend to flex under varying loads. The videos were dramatically smoother. In addition, the ESC’s were running cooler as well.