Thanks to @fpvsteve for the "Gopro Hero 5/6/7 and the Solo Gimbal" post. Coincidentally, I've been working on the same project of fitting a gimbalized Hero 5/6/7 to the Solo Drone. I seriously considered four approaches; (1) like fpvsteve hack a Solo gimbal; (2) mount a Tarot TL3T05 (purpose built Hero 5/6/7) gimbal; (3) mount a Zhiyun Rider-M wearable gimbal; and (4) mount a GoPro Karma Stabilizing Grip (don't laugh, some guy on YouTube successfully pulled this off).
With preserving the tilt function a high priority, I actually made a fair amount of progress with the first two approaches. Here's a couple of photos of my Solo Gimbal hack.
The blue pad is Moon Gel. Notable differences with fpvsteve's approach are the Hero 4 connector, I just cut it off with a rotary tool, the blue Moon Gel pad to provide a small amount of offset and some vibration dampening, the micro HDMI cable, I cut it off with some scissors and replaced it with a 30cm micro HDMI ribbon extension cable, and the counter weight on the back. Regarding this counter weight, although it looks crude and ugly, it's actually the result of countless hours spent futzing with the placement and attachment of many different sizes, shapes, and masses of counter weights. The reason for all this futzing is both my hacked Solo gimbals (yes, I hacked two of them, one is now deceased, and the other is permanently brain damaged, but more on this later) proved to be extremely sensitive to counter-balancing issues. I could finesse them into smoothly rotating through 90-degrees of tilt on the bench, but then lose that smoothness once in the air.
Here's an early video of my first hacked Solo Gimbal in the air. Note, flying low and slow without any in-camera stabilization was done deliberately to better expose all the warts of gimbal performance. To be fair, some (but not all) of the pervasive judder is due to air turbulence on the free-hanging HDMI ribbon cable, but bottom line, try as I might, I simply failed to produce any commercial-grade stabilized airborne footage from my hacked Solo gimbals.
By contrast, here's an early video of the Tarot TL3T05 gimbal in the air, and the difference in gimbal performance is dramatic. This gimbal is virtually out of the box--no endless hours spent hacking, trouble-shooting, and futzing with counter weights. I simply T-spliced the TL3T05 red and black power leads into the red and black wires in the Solo gimbal control cable, mounted the TL3T05 gimbal to the Solo drone, and voila--near 100% commercial-grade stabilized footage--eazy peazy. Again, though, to be fair to the hacked Solo gimbal, this particular flight test had no HDMI ribbon cable attached, so no possibility of air-turbulence induced vibrations, and the TL3T05 tilt function was not connected.
Thus, these two videos are far from a scientific, apples-to-apples test. Nevertheless, they present enough evidence to embolden me to stand up and be a wet rag on the whole hacked Solo gimbal movement. Permit me to take a couple of steps back here: when I got my first hacked Solo gimbal working on the bench, I was wildly excited. So much so that I immediately began hacking my second Solo gimbal, and it too performed brilliantly on the bench. Unfortunately, once I got around to flight test, the illusion melted and the dream became a nightmare. Worse yet, even though I hacked both gimbals in an identical manner, they (mis)behaved radically differently in the air--so no consistency, and more importantly, no repeatability.
In an ever-increasing effort to discover the golden recipe, I somehow damaged the first gimbal and killed it dead (possibly while trying to desolder the Hero 4 port connector). When I moved the second hacked gimbal to my primary Solo drone, it inexplicably lost all linear movement capabilities--it would not, could not move in a straight line. Worse, when I moved this gimbal back to its mother drone, the wonk behavior moved with it. Two stick calibrations and one gimbal calibration (via Mission Planner) later, and it's still busted. I've given up on it.
I'm ok with being a wet rag here because the results of the TL3T05 gimbal are so dang much better. So to pivot to the
TL3T05 here are some detailed photos:
The T-spliced power connector for the TL3T05 is on the right.
The TL3T05 is mounted to the undercarriage via two 3D printed struts. I've published the "TL3T05 Gimbal Strut.stl" file on thingiverse.com. ABS 100% infill. Strut dimensions are 6x12x21mm. Four screw holes are 7mm OC. If memory serves, diameter is 1.25mm (tap drill diameter for M1.6 screws) by 6mm deep. The screws are the second largest/longest in this kit. At one point, I exerted considerable effort to mount the the gimbal up inside the drone, but discovered there simply isn't enough space to allow the dampening balls full range of movement. Even if the top motor is placed upwards by a few millimeters, it could potentially crash the side of the mounting plate.
This photo documents two important points. The first is the placement of the strut screw holes on the mounting plate. The mounting plate is 75mm wide (l to r) and the gimbal mounting holes are 54mm OC, thus the screw holes are on parallel lines 10.5mm from each edge. I used the four vacant corner holes to measure a horizontal line half way in between, and the screw holes are located 3.5mm above and below this line (7mm OC total). For better accuracy, I made all markings on the other (smooth) side of the plate. I used the smallest tip in this soldering kit to "drill" the holes, then used an Xacto knife to clean up the melted edges. Keep the diameter of these holes small (< 2mm).
The second important point of this photo is the pinout diagram on the top of the TL3T05. This is where I could use some help back from the community. Based on Googlefu alone, I'm about 97% certain the default mode of the TL3T05 is PWM (as opposed to SBUS). I've also gleaned from multiple sources that the Solo drone's tilt signal is located on pin 14. My plan is to enable tilt by running a wire from pin 14 to one of the two PWM connections on the TL3T05 header. Can anyone out there positively confirm or deny the soundness of this plan?
One final tip, I recommend the 30CM or even 40CM version of this HDMI ribbon cable. The extra length is needed to preserve the full range of motion of the gimbal while at the same time routing the cable with the least amount of air turbulence.
Good luck everyone, and please let us know if you have any information to add.
With preserving the tilt function a high priority, I actually made a fair amount of progress with the first two approaches. Here's a couple of photos of my Solo Gimbal hack.
The blue pad is Moon Gel. Notable differences with fpvsteve's approach are the Hero 4 connector, I just cut it off with a rotary tool, the blue Moon Gel pad to provide a small amount of offset and some vibration dampening, the micro HDMI cable, I cut it off with some scissors and replaced it with a 30cm micro HDMI ribbon extension cable, and the counter weight on the back. Regarding this counter weight, although it looks crude and ugly, it's actually the result of countless hours spent futzing with the placement and attachment of many different sizes, shapes, and masses of counter weights. The reason for all this futzing is both my hacked Solo gimbals (yes, I hacked two of them, one is now deceased, and the other is permanently brain damaged, but more on this later) proved to be extremely sensitive to counter-balancing issues. I could finesse them into smoothly rotating through 90-degrees of tilt on the bench, but then lose that smoothness once in the air.
Here's an early video of my first hacked Solo Gimbal in the air. Note, flying low and slow without any in-camera stabilization was done deliberately to better expose all the warts of gimbal performance. To be fair, some (but not all) of the pervasive judder is due to air turbulence on the free-hanging HDMI ribbon cable, but bottom line, try as I might, I simply failed to produce any commercial-grade stabilized airborne footage from my hacked Solo gimbals.
By contrast, here's an early video of the Tarot TL3T05 gimbal in the air, and the difference in gimbal performance is dramatic. This gimbal is virtually out of the box--no endless hours spent hacking, trouble-shooting, and futzing with counter weights. I simply T-spliced the TL3T05 red and black power leads into the red and black wires in the Solo gimbal control cable, mounted the TL3T05 gimbal to the Solo drone, and voila--near 100% commercial-grade stabilized footage--eazy peazy. Again, though, to be fair to the hacked Solo gimbal, this particular flight test had no HDMI ribbon cable attached, so no possibility of air-turbulence induced vibrations, and the TL3T05 tilt function was not connected.
Thus, these two videos are far from a scientific, apples-to-apples test. Nevertheless, they present enough evidence to embolden me to stand up and be a wet rag on the whole hacked Solo gimbal movement. Permit me to take a couple of steps back here: when I got my first hacked Solo gimbal working on the bench, I was wildly excited. So much so that I immediately began hacking my second Solo gimbal, and it too performed brilliantly on the bench. Unfortunately, once I got around to flight test, the illusion melted and the dream became a nightmare. Worse yet, even though I hacked both gimbals in an identical manner, they (mis)behaved radically differently in the air--so no consistency, and more importantly, no repeatability.
In an ever-increasing effort to discover the golden recipe, I somehow damaged the first gimbal and killed it dead (possibly while trying to desolder the Hero 4 port connector). When I moved the second hacked gimbal to my primary Solo drone, it inexplicably lost all linear movement capabilities--it would not, could not move in a straight line. Worse, when I moved this gimbal back to its mother drone, the wonk behavior moved with it. Two stick calibrations and one gimbal calibration (via Mission Planner) later, and it's still busted. I've given up on it.
I'm ok with being a wet rag here because the results of the TL3T05 gimbal are so dang much better. So to pivot to the
TL3T05 here are some detailed photos:
The T-spliced power connector for the TL3T05 is on the right.
The TL3T05 is mounted to the undercarriage via two 3D printed struts. I've published the "TL3T05 Gimbal Strut.stl" file on thingiverse.com. ABS 100% infill. Strut dimensions are 6x12x21mm. Four screw holes are 7mm OC. If memory serves, diameter is 1.25mm (tap drill diameter for M1.6 screws) by 6mm deep. The screws are the second largest/longest in this kit. At one point, I exerted considerable effort to mount the the gimbal up inside the drone, but discovered there simply isn't enough space to allow the dampening balls full range of movement. Even if the top motor is placed upwards by a few millimeters, it could potentially crash the side of the mounting plate.
This photo documents two important points. The first is the placement of the strut screw holes on the mounting plate. The mounting plate is 75mm wide (l to r) and the gimbal mounting holes are 54mm OC, thus the screw holes are on parallel lines 10.5mm from each edge. I used the four vacant corner holes to measure a horizontal line half way in between, and the screw holes are located 3.5mm above and below this line (7mm OC total). For better accuracy, I made all markings on the other (smooth) side of the plate. I used the smallest tip in this soldering kit to "drill" the holes, then used an Xacto knife to clean up the melted edges. Keep the diameter of these holes small (< 2mm).
The second important point of this photo is the pinout diagram on the top of the TL3T05. This is where I could use some help back from the community. Based on Googlefu alone, I'm about 97% certain the default mode of the TL3T05 is PWM (as opposed to SBUS). I've also gleaned from multiple sources that the Solo drone's tilt signal is located on pin 14. My plan is to enable tilt by running a wire from pin 14 to one of the two PWM connections on the TL3T05 header. Can anyone out there positively confirm or deny the soundness of this plan?
One final tip, I recommend the 30CM or even 40CM version of this HDMI ribbon cable. The extra length is needed to preserve the full range of motion of the gimbal while at the same time routing the cable with the least amount of air turbulence.
Good luck everyone, and please let us know if you have any information to add.
