Compression versus Tension - Gimbal

So here is my much delayed reply. I wrote most of this up a couple days ago, but ran out of time. So without further ado...

First let me just say that I would never call myself a video Pro, but I appreciate the comment.

I'd also like to say that even though I stabilize everything in post, that's NOT a complete solution. There is only so much you can do in post to stabilize. The smoother the raw footage is, the better. I've had to scrap many many many many shots because we couldn't smooth out the vibrations in post. For personal stuff, I don't have a very high standard and will just use a cool shot even though it might have some shakes here or there. It's just easier and faster - and it's pretty much just my family watching (and maybe some of you guys if I post a video here). Even for the personal stuff it's still frustrating to have vibrations. It bothers me when I see it. So yeah, I'm always willing to try something new. When I look at your footage, I do see some vibrations here and there, but it's very obvious that the few vibrations there are could easily be fixed with a light pass of stabilization. An example is in the first video when you throttle up at about 20 seconds in. Easily corrected to cinema grade stability.

As an aside, the reason I stabilize everything is because it's painless, and often times footage that appears totally stable on a small screen will have some apparent vibrations on big screens. You'd be surprised. Footage that seems totally locked down usually has some vibration. Even footage shot with a camera on a tripod indoors can have vibrations from people walking across the room. A camera on a tripod outdoors will shake a tiny bit just from the wind. Most don't notice it, but you don't need a trained eye to see it - you just have to tell someone to look for it. Anyway, that's why you stabilize EVERYTHING.

I'd probably be willing to pay $20 to $80. Maybe more depending on what it is. However, if it's just a different set of replacement dampeners, then I'd feel taken advantage of.

When it come to dampeners, temperature matters.
I often fly in cold to extremely cold environments. A set that is fantastic in 70F might be total crap at -10F. So that's something to keep in mind. If you'd like me to do some testing for you at different temps, I can probably do that over the course of the winter. Just PM me. I'm spending a lot of time in Vermont and will undoubtedly get a few -10F days. And I'll get lots of days between 0F and 30F. I also will spend a lot of time in PA where temperatures range from the 20s up to 60F.

The question is how to do the testing scientifically. I have 3 Solos, but the best method would be to compare results on the same Solo/props/motor configuration with and without your system.
 
Mostly the stabilizer built into Final Cut. Contrary to others, I find it better than warp stabilizer. For problem issues I go into Motion which has more control. Tried core melts Lock and Load, which is supposed to be better, but just isn't. Not anymore anyway. I don't think they've updated it in a long time. I'm looking into using reel steady (an after effects plugin) to stabilize super aggressive flying. ReelSteady

Are you using something like Neat Video to clean up the H.264 block noise?
 
Are you using something like Neat Video to clean up the H.264 block noise?
I use Neat Video for cmos noise when it shows up in low light settings. Amazing how it can salvage footage that seems unusable. I'm taking about noise that looks like little sparkles, not compression artifacts (h.264 block noise).

To reduce compression artifacts, I NEVER shoot above 30fps. GoPro footage at 2.7K 60 fps is just to artifacty for my taste. If I want to shoot action for slow motion I might use 60fps, but I try not to. Depending on the subject and the lighting, I can get away with 30fps and then in final cut use optical flow frame interpolation to slow things down (as opposed to frame blending). Seems to work pretty well for aerial footage.
 
  • Like
Reactions: RichWest and J R
So here is my much delayed reply. I wrote most of this up a couple days ago, but ran out of time. So without further ado...
Very informative and honest, thanks for taking the time to explain.

Several here are on my list as beta tester just because they're in cold climates. That was one of my major hurdles in sourcing and testing materials, how would it work in very cold environments. My wife became accustomed to finding things other than frozen foods and ice within our freezer. For my needs as a beta tester, it's how hot can it handle.

The system I've devised is not perfect by any means. Certain vibrations can't be contained only because they are random in appearance. Your reference to the 20sec mark with throttle up is a prime example of what the irregular motor spin up can inject. Motor or ESC aberration can be tuned or remedied, so I've ignored those by listening to the motors when evaluating the progression in finding a solution. Wind was less of a factor while testing, since the gimbal and isolators worked as they should.

As to the end cost, I've really tried to ignore that aspect while searching out a solution. Otherwise, I'd been done and flying my solution weeks ago. Further, we wouldn't be having this conversation today. To answer your question, this is a two part system. So its not just a new isolator design, even though it would work well by itself.

In the end Silicone seems the answer, as it has been the choice by many for a reason. It has a long list of positives. The only glaring negative is the mold and production costs for limited runs. Not like I have a million parts needed at this point. So I'm being conservative and projecting small runs in multiples of 1000, which is 250 complete solutions. Obviously I don't have hard costs at this point, but you're starting to see why I was anguishing if others saw benefit from the solution as presented.

Anyway, I still have a lot of work to do. My world revolves around metal, so this is a long stretch from where I reside.
 
To reduce compression artifacts, I NEVER shoot above 30fps. GoPro footage at 2.7K 60 fps is just to artifacty for my taste.
I just went out and tested shooting at 30fps and you are correct in that it reduces the artifacts. Really interesting.

However I encountered an issue. Slowing down the frame rate revealed a slight vertical oscillation in the video that didn't show up at 60fps. It wasn't horrible, but it was there. This was during the same flight for the two videos, so nothing physically changed. Stepping frame to frame the images are clear I'm assuming I'm not able to see the motion as well at the higher frame rate.....crap...need to think this over.
 
I just went out and tested shooting at 30fps and you are correct in that it reduces the artifacts. Really interesting.

However I encountered an issue. Slowing down the frame rate revealed a slight vertical oscillation in the video that didn't show up at 60fps. It wasn't horrible, but it was there. This was during the same flight for the two videos, so nothing physically changed. Stepping frame to frame the images are clear I'm assuming I'm not able to see the motion as well at the higher frame rate.....crap...need to think this over.
Ah. That's right. Higher frame rates help reduce jello at the expense of quality. You should definetly test at lower frame rates, even 24 fps.

Just keep in mind that jello is not nessisarily caused by vibration. With certain subjects (like crashing surf on a beach) you can get jello on a camera that is sitting on a stable tripod with no wind. People always use vibration and jello interchangeably, but that's not technically correct. While high frequency vibration can cause Jello, it can also occur on a rolling shutter (like the GoPro has) when the timing of shutter speed (NOT frame rate) is out of step with the subject matter.

That's why you can reduce jello by either uping frame rate, which on a GoPro forces a faster shutter speed, or by using slower frame rates with an ND filter that actually slows the shutter speed down. It's all about timing.

Slower frame rates with slower shutter speeds results in a higher quality image with a cinematic motion blur. I prefer that to fixing jello with higher frame rates / shutter speed.

Anyway, I'm headed into a low coverage area for awhile. I'm pretty sure I describe this under the frame rate section of the GoPro settings sticky. Check it out.

At the end of the day, if you've been testing at 60 fps this whole time and your video has less vibration than it did before, then you've succeeded in reducing vibration, regardless of what you saw with 30 fps today. Maybe it needs more tweaking, but your on the right track.
 
Last edited:
So here is the latest test, ignore the jello, as that was not the goal for this flight. This is just a video to show off what the Solo really is capable of with the "right equipment" and a very simple gimbal mod. You can refer to the 2216-12 motor thread for details on the motor aspect using APC props.

Again no stab applied, 2.7k 60fps med, converted in GPStudio to 29.97fps ( ;) ) and then uploaded to YT. Even the 1080 quality video looks good from my end.

I think I'm heading in the right direction now...;)

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
  • Like
Reactions: erikgraham
Wish I knew how to control a GP better, results could possible appear better than presented.

Anyway this is the results of a DIY solution of about 3 weeks. I'm sure it could be improved upon further, but my mind is spent and am looking forward to others progressing these efforts. It a simple idea and requires minimal skills or monies.

I'll release the mod later this weekend and hopefully a video to show the process to replicate.

Wind, with minimal vibes. Again no post stabilization....it's just hardware @erikgraham ...;)

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
Rich,

Looks really solid. I'll be looking forward to your next posts. Thanks so much for all of the work you've done on this stuff. Really, thank you very much!

Jerry
 
Hello all. I’m finally at a point to share my efforts in reducing gimbal vibration. I have devised three separate solution, but my goal has remained to make it a DIY solution. The first is a redesigned isolator, molding cost make this endeavor expensive. The second is using a standard material and shape cutting it, low up front cost but high material cost. And finally the third solution modifies the isolator itself, very cheap solution.

So let’s discuss the problem, the isolator balls that support the gimbal and attempt to isolate the vibration.

For Macro vibes, gimbals work best if they can swing or sway, like a pendulum. Any gimbal’s design works best on arcs of movement. In our case the gimbal does not ROLL very well, it typically moves back and forth on the same plane. This is pretty much the same issue for all compression type damper systems. What is needed is for one pair of isolators to give (compress) and the opposing pair to elongate (extend) and then do the same in the opposite direction.

Micro vibes are from the bounce of the ball. It almost takes on a life of its own, the bounce. There’s a catch 22 for how hard or soft an isolator is for the application and the weight of the item to be isolated. You can actually get to the point that isolator’s designs are the cause for the micro vibes. Too soft and you risk the isolator to collapse, negating any dampening abilities. It’s a fine line for how soft you need to dampen vibes and then the hardness to support the structure.

About 8 weeks ago I had decided to put everything on hold with the first two options. Needed to stop Brain f’ing myself about how to move forward. Wasn’t sure what NAB was going to produce for us, great more software…yippee.

The needed time away from the project was great. About 3 weeks ago I decided to try out one of my original ideas, I had purchased 10 sets of stock isolators back in December, just to play with the idea at some point. Anyway I punched a few 1/8” holes in the isolator and tested, no improvement. Went to 3/16” holes, still no change. Finally I punched ¼” holes and bam everything was stable. Really solid video, macro vibes were virtually gone. And then the micro vibes were down to a level I had never had before.

The problem however was when I pitched the camera below 45 degrees and below, jello city. I assume the camera offset the pitch weight to the gimbal and had collapsed the front pair of isolators. The remedy it appears was by placing a Confor foam (memory foam) cube into the isolator. This supported the isolator from collapsing or just reduce the instability of the isolator. It’s possible an ear plug could work equally as well, didn’t try, so I have no clue.

Obviously each test required the gimbal to be removed and tweaks have to be done. In this round alone I removed the gimbal more than 10 times, over the last six months I’d venture to bet over 400 times. I know the gimbal fairly well, it’s really rock solid…;)

Now is this “THE” solution? Not sure, but it will give everyone a new opportunity to resolve the vibration. I do believe my two other ideas will playout in the future for some application. I’ve really learned quite a bit about vibration, damping materials and gimbals these last few months.

Below I’ll layout the hole placement on the isolators and the best orientation within the gimbal. Also some suggested tools for the task and a source for the Confor foam.

I would recommend buying an extra set of isolators if you screw up or it doesn’t work out for you. It is imperative to have made an effort to route the HDMI cable properly to the gimbal so that it is of minimal impact to the gimbal’s movement.

Finally, I would hope we have some adventurers here willing to try the mod. As far as I’m concerned this is the best solution to date, but will require further tweaking to eke out as much vibration as possible. I do believe temperatures will effect this mod, Confor is temp sensitive, so your solution may require adjusting to suit your needs. It is possible I’ll have a universal solution sometime in the future, but then you could as well with a little experimenting.
___________________

Punch 1/4" holes through the isolator wall at 90⁰ to each other around the circumference. I suggest using the molding line as a reference as it will help to align the isolators when installed, this will be important in how the mod works. You can use an unsharpened pencil inserted into the mounted isolator to turn so that the punched holes are oriented as shown in the video.

I used an actual leather punch, the silicone material is easily punched by hand pressure. You could improvise the punch if you have a 5.56 rifle case. Use a board or popsicle stick to inserted into the isolator as backup to the punch process, nothing more than 5/16” wide. Cutting with a straight edge razor of sorts will end in failure at some point, hole punching the material works best.

The nylon fishing line will help aid installing the isolators into the gimbal mounts. Fold a length of line, press to a point. You should end up with a pointed loop of about 3 inches long.

Once isolators are installed, align the newly punched holes of all four isolators so that they align to the gimbal’s hub. This is shown in the video, a pencil is used to turn the isolators.

The confor foam is really for support rather than actual damping material, I’m using the medium (blue) material at the moment. (in the video I used yellow for the contrast) The correct size at this point is about an 18mm ball, which allows for some compression once inserted. It can be cut with a razor knife with some success, freeze it to cut with a saw. Seeing that it is too soft to shape round, a cube of 18mm should not be an issue as you can always whittle it down to suit your needs. I used a 5/8” punch and made two cuts to fashion a semi-round cubed inserts.

1/4" Leather Punch or similar
Short length of 50 lbs fishing line or similar
Slow Recovery Super-Cushioning Foam Selector Block, 1" x 5" x 5" 86195K1– McMaster.com

The GoPro Frame is used in the video to demonstrate, this mod is intended for the Solo gimbal.
All FOUR Isolators should be modified as described and installed onto the Solo gimbal.


To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Good luck!

Edit to Add Drawing:

Solo_Gimbal_Isolator_MOD.JPG
 
Last edited:
Hello all. I’m finally at a point to share my efforts in reducing gimbal vibration. I have devised three separate solution, but my goal has remained to make it a DIY solution. The first is a redesigned isolator, molding cost make this endeavor expensive. The second is using a standard material and shape cutting it, low up front cost but high material cost. And finally the third solution modifies the isolator itself, very cheap solution.

So let’s discuss the problem, the isolator balls that support the gimbal and attempt to isolate the vibration.

For Macro vibes, gimbals work best if they can swing or sway, like a pendulum. Any gimbal’s design works best on arcs of movement. In our case the gimbal does not ROLL very well, it typically moves back and forth on the same plane. This is pretty much the same issue for all compression type damper systems. What is needed is for one pair of isolators to give (compress) and the opposing pair to elongate (extend) and then do the same in the opposite direction.

Micro vibes are from the bounce of the ball. It almost takes on a life of its own, the bounce. There’s a catch 22 for how hard or soft an isolator is for the application and the weight of the item to be isolated. You can actually get to the point that isolator’s designs are the cause for the micro vibes. Too soft and you risk the isolator to collapse, negating any dampening abilities. It’s a fine line for how soft you need to dampen vibes and then the hardness to support the structure.

About 8 weeks ago I had decided to put everything on hold with the first two options. Needed to stop Brain f’ing myself about how to move forward. Wasn’t sure what NAB was going to produce for us, great more software…yippee.

The needed time away from the project was great. About 3 weeks ago I decided to try out one of my original ideas, I had purchased 10 sets of stock isolators back in December, just to play with the idea at some point. Anyway I punched a few 1/8” holes in the isolator and tested, no improvement. Went to 3/16” holes, still no change. Finally I punched ¼” holes and bam everything was stable. Really solid video, macro vibes were virtually gone. And then the micro vibes were down to a level I had never had before.

The problem however was when I pitched the camera below 45 degrees and below, jello city. I assume the camera offset the pitch weight to the gimbal and had collapsed the front pair of isolators. The remedy it appears was by placing a Confor foam (memory foam) cube into the isolator. This supported the isolator from collapsing or just reduce the instability of the isolator. It’s possible an ear plug could work equally as well, didn’t try, so I have no clue.

Obviously each test required the gimbal to be removed and tweaks have to be done. In this round alone I removed the gimbal more than 10 times, over the last six months I’d venture to bet over 400 times. I know the gimbal fairly well, it’s really rock solid…;)

Now is this “THE” solution? Not sure, but it will give everyone a new opportunity to resolve the vibration. I do believe my two other ideas will playout in the future for some application. I’ve really learned quite a bit about vibration, damping materials and gimbals these last few months.

Below I’ll layout the hole placement on the isolators and the best orientation within the gimbal. Also some suggested tools for the task and a source for the Confor foam.

I would recommend buying an extra set of isolators if you screw up or it doesn’t work out for you. It is imperative to have made an effort to route the HDMI cable properly to the gimbal so that it is of minimal impact to the gimbal’s movement.

Finally, I would hope we have some adventurers here willing to try the mod. As far as I’m concerned this is the best solution to date, but will require further tweaking to eke out as much vibration as possible. I do believe temperatures will effect this mod, Confor is temp sensitive, so your solution may require adjusting to suit your needs. It is possible I’ll have a universal solution sometime in the future, but then you could as well with a little experimenting.
___________________

Punch 1/4" holes through the isolator wall at 90⁰ to each other around the circumference. I suggest using the molding line as a reference as it will help to align the isolators when installed, this will be important in how the mod works. You can use an unsharpened pencil inserted into the mounted isolator to turn so that the punched holes are oriented as shown in the video.

I used an actual leather punch, the silicone material is easily punched by hand pressure. You could improvise the punch if you have a 5.56 rifle case. Use a board or popsicle stick to inserted into the isolator as backup to the punch process, nothing more than 5/16” wide. Cutting with a straight edge razor of sorts will end in failure at some point, hole punching the material works best.

The nylon fishing line will help aid installing the isolators into the gimbal mounts. Fold a length of line, press to a point. You should end up with a pointed loop of about 3 inches long.

Once isolators are installed, align the newly punched holes of all four isolators so that they align to the gimbal’s hub. This is shown in the video, a pencil is used to turn the isolators.

The confor foam is really for support rather than actual damping material, I’m using the medium (blue) material at the moment. (in the video I used yellow for the contrast) The correct size at this point is about an 18mm ball, which allows for some compression once inserted. It can be cut with a razor knife with some success, freeze it to cut with a saw. Seeing that it is too soft to shape round, a cube of 18mm should not be an issue as you can always whittle it down to suit your needs. I used a 5/8” punch and made two cuts to fashion a semi-round cubed inserts.

1/4" Leather Punch or similar
Short length of 50 lbs fishing line or similar
Slow Recovery Super-Cushioning Foam Selector Block, 1" x 5" x 5" 86195K1– McMaster.com

The GoPro Frame is used in the video to demonstrate, this mod is intended for the gimbal.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Good luck!
This operation should be done only 2 isolators and all 4?
 
  • Like
Reactions: RichWest
Good luck!

Rich,

I'm really looking forward to giving this a try! I love the fishing line trick! I have the same question as Dmitry, we're only modifying the two front isolators, correct?

Thanks for sharing this with us. Having followed your progress I can't even imagine the time that you invested in this possible solution.

Jerry
 
  • Like
Reactions: RichWest
This operation should be done only 2 isolators and all 4?
Yes, sorry, it should be all four isolators. I was just demonstrating the install portion...
Very nice, Rich! And Yello for background ta boot!
Well it was actually brown paper background, I really didn't want anyone to see my mess of a work table. Be glad you got a video...I shot it in one take, otherwise you'd just read the description...;)
 
Yes, sorry, it should be all four isolators. I was just demonstrating the install portion...

Well it was actually brown paper background, I really didn't want anyone to see my mess of a work table. Be glad you got a video...I shot it in one take, otherwise you'd just read the description...;)
I was referring to the music! lol
 
I was referring to the music! lol
Actually it was Art of Noise, but sure I see what you're saying now...haven't thought of Yello in a long while. I guess I need to dust off some albums and try some "new" music... That's half the fun of producing any video, what music people haven't heard in a while....or never. I was about to insert Good Vibration by the Beach Boys...
 
Amazing work there Rick. Thanks a lot. Next thing I'll do when I come back home is try it.
 

New Posts

Members online

Forum statistics

Threads
13,095
Messages
147,750
Members
16,065
Latest member
alan r pfennig