camera is a Sony ZV-E10 on a 3-axis brushless gimbal, nothing fancy but its been solid until recently. frame is a 5-inch, running 5045 triblade props. i dont think ive changed anything hardware wise but i did flash new firmware on the FC a couple months ago which is around when i first noticed it maybe? hard to say.

the gimbal dampening balls look fine visually but ive never actually replaced them. wondering if its a resonance thing at certain throttle levels that happen to correspond to that altitude band. or could it be the electronic shutter interacting with something. i shoot at 1/100 usually which should be fine for 50fps but maybe thats related somehow

anyway im looking at retrofitting a Ronin-MX onto this thing. the weight budget is tight, around 400g of margin left after everything else, and the MX is pushing right up against that. the 3-axis stabilization is obviously what im after but im wondering if anyone has actually done this on a non-DJI frame and what the integration looked like. specifically the CAN bus stuff and whether the stabilization modes actually work properly without DJI's own FC in the loop.

alternative is just biting the bullet and getting something like the Gremsy H3 which seems more agnostic about what its mounted on. less brand lock-in basically. has anyone compared footage quality between these two at similar bitrates? like is there actually a visible difference in the smoothness or is it more about the camera you hang underneath

the X7 raw files are incredible dont get me wrong, but im wondering if for 80% of commercial work the X5S or even a properly tuned Mavic 3 Pro setup would get me there with way better flexibility. the thing i keep going back and forth on is the gimbal integration ? the X7 system is so dialed in, the stabilization is stupid good. i dont know if i could get the same smoothness out of a third party gimbal solution on a smaller airframe.

curious if anyone has actually made this switch or run comparison shots between these systems in real conditions, not like a controlled youtube test

ive tried recalibrating multiple times, different locations, away from metal structures, all that stuff. balanced the camera as carefully as i can. the weird thing is it only seems to do it when im doing longer sweeping movements, if im just hovering it looks fine. im starting to wonder if its a motor issue or if i have the payload balanced wrong somehow. camera is a Sony a7III with the 24-70 GM lens which i know is on the heavier end but its within spec.

anyone run into this? is there some setting im missing or should i be looking at getting it serviced

the thing is i know theres a significant jump in complexity when you go from a self contained drone to something where you're managing a separate gimbal and camera. more things to go wrong, way more setup time, more weight which means shorter flight times. and for real estate im usually doing quick shoots, 20-30 minute window at a site, so efficiency matters a lot.

is anyone doing high volume real estate work on a heavier platform and actually finding it worth the tradeoff? or is the Mavic 3 Pro sensor honestly good enough that im just chasing diminishing returns at this point. genuinely not sure what to do here

my suspicion is the motors on the hex are causing some kind of vibration that's confusing the IMU but honestly i don't really know. the camera is a Sony A7C which is about 500g with the lens so its within the payload rating. running a 24mm f2.8 so nothing too heavy on the front. anyone had something similar or know what i should actually be looking at here

the weight jump is pretty significant and i know it's gonna hurt my flight time noticeably. currently getting around 40 minutes in calm conditions and i'm expecting to drop to maybe 25-28 with the heavier rig. that's manageable with battery swaps but im more worried about the gimbal handling the extra mass during acceleration changes and whether stabilization is going to be comparable to what im used to with an integrated system. integrated systems always feel like they have an edge because everything is tuned together from the factory.

anyone actually running a mirrorless body on a hex or octo class machine for professional work? curious if the image quality jump actually justifies the operational headache or if im just chasing specs

ive already recalibrated the IMU twice and done a full gimbal calibration through the app. motors spin fine when i manually move it, no grinding or anything weird. the footage is basically unusable because you can see this slow drift every few seconds and the correction jerk is even worse than the drift itself.

not sure if something physically got knocked out of alignment or if its a firmware/sensor thing. anyone dealt with this before? the gimbal itself doesnt look visibly damaged but obviously something is off internally

ive recalibrated the IMU twice following the DJI instructions exactly and it seems fine on the ground when im doing the calibration but once its in the air under load the drift comes back. motors all seem to be spinning, no obvious physical damage i can see, no bent roll axis or anything. payload is a Sony a6300 which is within spec.

not sure if i should just send it in or if theres something obvious im missing. anyone dealt with this kind of thing before

my current setup is nothing fancy ? betaflight on an F7, emax motors, the usual stuff. gimbal is out of the question on a freestyle rig obviously but im just talking pure image quality here. the rolling shutter on the Phoenix 2 is noticeable in fast rolls and i've kind of accepted it but watching some O3 footage last night made me wonder if i've just been tolerating something i dont have to.

anyone made this switch? was it actually worth it or is it one of those things where you spend the money and then realize the footage difference only matters if you're doing cinematic stuff not just bashing around

been looking at what other people run on the inspire 2 for situations where you dont need the full cinema setup. the x5s is the obvious answer i guess but im curious if anyone has tried third party options or lighter mirrorless rigs on a custom mount. i know theres people who run m43 cameras on inspire platforms with aftermarket gimbals but ive never seen anyone talk about how that actually performs in real shooting conditions, like does the stabilization hold up to the x7 standard or is it noticeably worse.

not really looking to cheap out, more interested in a smart payload swap that keeps image quality high while giving me maybe 3-4 extra minutes in the air. even that would make a real difference on some of the tighter jobs.

the problem is on the roll axis mostly. after about 3-4 minutes of flight it starts drifting clockwise, like maybe 2-3 degrees, doesnt sound like much but its enough to ruin a shot. ive rebalanced probably five times now, tried adjusting the PID on the roll motor, bumped up the power a bit. it gets better for a flight or two and then starts doing it again. temperature maybe? its been pretty warm here lately.

anyone dealt with this kind of slow drift on a brushless gimbal? wondering if its the IMU placement or if the motor is just getting too warm and losing torque. open to any ideas at this point honestly.

im looking at the whole ecosystem of small cinema-ish cameras and gimbals and honestly its a mess. the sony ZV-E10 with a pancake lens and a Gremsy or similar 3-axis gimbal puts me right at the edge of my weight budget but the image quality would be insane for what this thing is. on the other hand the Insta360 SMO 4K on a lightweight gimbal is way less total weight and the stabilization is surprisingly decent, plus no lens to worry about.

i dont do paid work, this is just for personal projects and maybe some landscape stuff at my cabin. so the ROI calculation isnt really about money, more about whether the extra complexity and weight of a proper mirrorless payload is actually worth it vs just getting really good output from a smaller fixed-lens camera. curious what people who have actually flown both types think, not looking for spec sheet comparisons i can google those

ive already done the calibration dance like three times, reset the IMU on the gimbal, checked that the camera isnt sitting crooked in the mount. the mount itself is a printed TPU thing i grabbed off printables which i guess could be part of the problem but it seemed solid when i was fiddling with it on the bench. vibration dampening is two layers of o-rings between the stack and the mount plate.

anyone dealt with this kind of slow drift before? im wondering if its a motor fatigue thing or if the gimbal just doesnt play nice with the vibration profile of a 5inch. or maybe theres a pid setting on the gimbal app im missing. open to ideas before i just throw money at a different gimbal.

the alternative ive been looking at is a Sony A7C II or even the FX3 on a Gremsy or maybe a Freefly gimbal on a larger platform. the FX3 in particular is tempting because of the active cooling and the fact that i already own a bunch of Sony glass. payload would be on the heavier side obviously but im building out a heavy lift hex anyway so weight isnt really the limiting factor here.

the thing i keep coming back to is stabilization quality. like the X7 integration with the Inspire's flight controller for horizon leveling and all that ? does a third party gimbal actually get close to that when you factor in signal lag and how the two systems talk to each other? or is that mostly marketing and in practice the difference is pretty minimal on a well tuned rig?

ive already tried recalibrating the IMU like four times, checked all the motor connections, and even swapped the camera plate thinking maybe something was loose. roll motor itself seems fine when i manually check the resistance. honestly at this point im wondering if the firmware update just broke something in the PID tuning because i dont remember changing anything else around that time.

anyone seen this or know if rolling back firmware is even possible on the T3 series? their support has been kind of slow to respond.

the Boson with a proper gimbal mount is like 90-120g depending on the lens, which doesnt sound crazy, but i know from experience that every 100g on this platform costs me like 2-3 minutes of flight and its already hovering around 18 minutes on my 6S 5200 packs. also not sure how thermal image quality holds up in daytime heat shimmer, ive seen some footage that looks really washed out. anyone actually running thermal on a medium class hex? curious what your real world experience is vs the spec sheet stuff

ive recalibrated the IMU like four times, redone the gimbal balance, even swapped the ribbon cable thinking that might be it. the gimbal itself is a fairly cheap two-axis unit so maybe thats just the answer and i need to accept it lol. but thought id ask before i drop money on something better. anyone seen this kind of slow drift before or is it just a trash gimbal situation

also wondering how people are handling the gimbal side of things for thermal, most of the mounts i see for FLIR units are basically just shock isolation plates rather than actual stabilization. wondering if thats good enough or if anybody has figured out a proper 3-axis solution for the vue series. the radiometric data is useless if the image is blurry from vibration i would think.

i dont think its a PID issue because i havent touched any of that yet, just wondering if anyone has run into this where going bigger actually made things worse before they got better. the new gimbal is rated for the weight so its not overloaded or anything. payload is right in the middle of the range actually.

the drone is a custom hex if that matters, hobbywing X6 motors, 30x10 props roughly. pretty stable platform generally speaking.

the H20T is obviously the cleaner solution since its all integrated and the sensor fusion between the thermal and zoom is actually really useful in practice, ive used it a few times. but the price is brutal. ive also seen people cobbling together third party FLIR cameras on a separate gimbal on the aux port and getting decent results, supposedly. the integration is obviously worse but if the actual thermal image quality is comparable or better it might be worth it depending on how the numbers work out.

anyone gone deep on this comparison or have real world experience with both? mainly care about detection range and image quality in low-contrast environments like treelines at night, not really worried about radiometric precision for this use case

i dont think its the IMU because the drone itself is holding position fine, its definitely the gimbal. ive tried re-seating all the ribbon cables, checked motor temps after a flight and they seem normal. payload is right at the edge of the Gremsy's rated limit but ive flown heavier with no issues before on a different rig. anyone dealt with this or have any ideas before i start swapping parts

the math on paper looks like building my own payload rig would be cheaper and i'd have more flexibility to swap cameras but every time i talk to people who've actually done production work they seem to lean toward the integrated systems. i get that the zenmuse x9 on the inspire is incredible but the whole ecosystem lock-in thing bothers me. and i already have the airframe so there's something to be said for not spending another 15-16k on a platform

has anyone actually run a bmpcc or similar on a gremsy under a heavy lift for real production stuff? curious how the image quality and stabilization held up compared to something like the x7 or x9

i checked the balance about a dozen times, tried adjusting PID gains up and down, reflashed firmware. the drift only shows up when the motors are warm which is making me think maybe its thermal expansion messing with something mechanical. or maybe the IMU on this thing is just garbage. has anyone run into this kind of thing with budget gimbals? im not opposed to just buying something better but id like to understand whats actually wrong first before i throw money at the problem

the problem is going that route means building out a custom frame, sourcing a gimbal that can handle the weight properly, recalculating flight times because obviously payload goes up, and just generally a whole project. i keep going back and forth on whether the image quality difference actually justifies all of that for the work i'm doing. like is anyone actually running something in that weight class for commercial work and finding it worth the hassle vs just using an integrated system