If it’s not the best place to start this thread, sorry in advance…
My goal is to build a drone that can carry a camera and a simple gimbal for photography.
I have already built a 5inch drone to gain experience in drone building, configuring, flying (crashing ) - but that drone was destined to be underpowered and it is indeed underpowered.
Now it’s time to start building the final drone. Below setup is based on eCalc.ch:
Flight time: about 30minutes
Maximum TOW: about 4000g with 6s4p pack
Thrust-to-weight ratio: 3.5
Frame: H680 foldable carbon frame
Battery: 6s4p EVE 40PL (maybe 6s3p)
FC: MicoAir743
ESC: 4-in-1, 60A minimum
Motors: Emax ECO2 3115 500kv (eCalc.ch estimates 85C motor temp at full throttle, but I think it’s not an issue)
Propellers: 15x5inch, bi-blade, foldable
Gimbal control with simple servo
Current sensor: Mateksys I2C-INA-BM (max battery current is expected to be 191A, this sensor board can handle it)
GPS
Later I may add optical flow sensor and down-facing LiDAR for better position holding capability (after testing on a smaller drone)
VTX: analog or digital, not decided yet
Before spending a ton-of money, I would like to ask the experts’ opinion about this setup.
Is there significant improvement opportunity in the “design”?
Are there huge mistakes in the setup?
Where can I expect difficulties during assembly and tuning?
by this you mean a gimbal using servo as its motors?
if yes, I won’t recommend it if you are doing some videography work (you have mentioned photography but servo gimbal isn’t great for anything other than pointing in a general direction.
I would recommend you to buy a brushless gimbal (3axis preferably, 2 axis also works). Even cheap boards like Storm32 work fine, and they are supported by ardupilot.
If you still decide to use servo gimbal, you should go with motors based on payload; for light payload like analog cam, sg90 or mg90s are sufficent, but for a proper camera use mg995 atleast, preferably digital SBUS servos as they are far better quality and faster response (but check datasheet once).
Trust to weight ratio of 3.5 is very high, you’d need to have a payload fitted at all times.
You can easily reduce this by using smaller props for the times when you dont have a significant payload. Just put the bigger props on for tough jobs.
With any frame there will always be a certain amount of modification required, so be prepared for that.
Innloi have something similar and I’ve purchased frames from there before. I’ve found their carbon fibre is good quality and the frames go together well and are nice and stiff. Choose an option with the 2.0mm thick CF
@Ishitbh
Videography is not in my plans. I’m more like a photo-guy. On the other hand, videography would need a 3axis gimbal - that’s way more heavier.
Now I’m using a 20kg servo.
In the future I may switch to Storm32, I already have a board at home - I know it’s far better, sophisticated.
@xfacta I’m prepared for drilling, curring, soldering stuff. That’s the fun part
I already tried to “limit” thrust-to-weight ratio to a more considerate value, to limit the peak battery current. Both to be able to measure with Mateksys sensor both for keeping my connectors small.
I can decrease even more, if needed - be reducing either prop pich either motor KV (400KV is a lower from EMAX ECO2 3115 series)
I’m planning to use a similar frame what you shown:
I would recommend lowering the prop size, if possible.
it allows you to increase thrust for cheaper in the future, when you would add other things like 3axis gimbal etc.
changing props is far easier than putting new motors, go with the current ones unless you plan to add more stuff (like gimbals, telem, action cams, etc)
Interested to know why this may be an issue? I’ve had copters with a T:W ratio of 4-5 before and where I only fly them in LOITER, AUTO or GUIDED (no FPV or ACRO stuff) and they were completely fine. It was nice because I had weight budget for a payload if I wanted one attached.
It took quite a time to biuld the quadcopter due to many circumstances in life…
But it’s complete, it works
The frame was pretty weak so I added a 3mm thick CF plate to the bottom of the craft. I screwed it to the original frame at the landing gears. It’s much more sturdier now.
After performing all the necessary steps with AMC I have mounted 900gramms of canned food (cheaper to crash than a MILC camera ) to the place of the camera and did a few test flights.
A hover flight log is here: https://drive.google.com/file/d/1OSMnmCOrCPaM2MPWNXXQIdp0u9lp8Upg/view?usp=sharing
I think it’s acceptable except one thing: the motors are running at different RPM (~3500 vs. ~4000). I checked the motor mounts on the arms: they are not aligned on the rounded arms - I think this causes this inbalance. I need to fix it.
Other than the easily fixed yaw bias it’s looking pretty good. Out of curiosity how did you arrive at configuring the RPM parameters for ESC Telemetry?
Yes. Not sure why those would be set for standard ESC Telemetry. They are for RPM Sensors. This is one of the issues with Rosser’s guides also. All you need is the SERVO parameters.
@amilcarlucas what’s the story with that? Why configure in 2 parameter groups?
That is not a recommendation of AMC!
AMC detects that you used those parameters (their value != default) and presents them to you, so that you see them and have the chance to correct/disable them.
If AMC did not show you the RPM parameter you would have no possibility to disable them, would you?
@dkemxr So no, is not encouraging you to configure 2 parameter groups, it is informing you that you did that in the past (before AMC) and giving you the possibility to correct it.
AMC will never change a parameter behind you back, not even one that is clearly wrong, like this RPM stuff.
If the user hadn’t changed the RPM values before using AMC, then AMC would not even present those parameters in the GUI.
Bottom line, on a correctly configured vehicle those RPM parameters do not get displayed. But if the user misconfigures them before using AMC, AMC will start tracking them.
Unless you have very big arms, for that size I would stay away from 15 inch props which will be a challenge to tune. That is almost guaranteed. The 15 versus smaller prop would give an additional saving of maybe 10 to 15 %, but if you want only 30 minutes of flight time, you realkly do not need the 15 inch props. I think that you can build the frame and all electronics and motors included for about 1200 grams, and with a 24 cell battery which also weighs about 1200 grams you would be looking to about 50..60 flight minutes. Now, if you go to 4000 grams, that is, you want to put 1500 grams of payload, your flight time will probably be under 30 minutes, but I can not think of anything useful and heavier than 300 grams payload… Unless of course you plan to build a long range kamikaze drone…
I do not think that this is a problem. All of my long range copters have an excessive thrust to weight ratio, and the auto throttle learning goes down to 0.15 .. 0.18. It gives a reserve for high altitude flights and does not produce any efficiency penalty.
The motors could in theory burn at full power (in one case I have 300 W motor, and at maximum throttle the power would be 500 W), but I use sanity when flying.
) - but that drone was destined to be underpowered and it is indeed underpowered.
