Cinewhoop Style Quadcopter

ArduCopter Copter 4.6
1

Hi everyone,

I’m running a custom cinewhoop-style frame (similar to a DJI Avata drone) with the following setup:

  • Motors: EMAX ECO 1106 4500KV motor

  • Flight Controller & ESC: Custom

  • Props: 3-inch HQ tri-blade

  • VTX: Walksnail Avatar

  • Battery: 2S Li-Ion (Molicell 3000mAh)

My current AUW is around 320g, and I’m getting about 4–5 minutes of flight time.

I’ve also noticed a error Potential Thrust Loss once voltage falls below ~7.4V, which is affecting performance toward the end of the flight.

I will work on:

  • Fixing FC voltage/current inaccuracies

  • Reducing overall weight

However, I want to keep the following constraints:

  • 3-inch frame (fixed)

  • 2S battery only (fixed)

So my questions are:

:backhand_index_pointing_right: Is it realistically possible to achieve 12+ minutes of flight time under these constraints?
:backhand_index_pointing_right: What are the key parameters I should optimize ?
:backhand_index_pointing_right: Are there better 2S-compatible motors that could improve both efficiency and low-voltage performance?

Any insights would be really helpful.

this is the Log

2

Hi there!

Nice project. Can you share some pictures of your build?

The discharge rate is probably too high to use the full capacity of your batteries. I think you need to reduce the weight a lot.

For motor/prop calculations ecalc.ch works quite well.

1 Like
3

pic
pic1600×1559 216 KB

This is just a rough prototype build, and we still need to refine and optimize the design further.

Before moving ahead with improvements, I want to first confirm and understand the right choices for components like motors and other key parts.

I’d appreciate any suggestions for improvements in any aspect of the build.

4

@YupsUAV I’m looking for suggestions to increase flight endurance on my 3-inch cinewhoop.

If anyone has tips, recommendations, or improvements for efficiency, motor/prop combos, or setup optimizations, I’d greatly appreciate it.

Additionally, if someone has good tuning parameters for a 3-inch frame using motors with similar KV (latest stable ArduPilot version), please share them.

5

The Diatone Taycan AMC template, is a very similar build. Also 3 inch, also similar frame, also similar FW version.

6

Your build looks basically fine but I think there are a lot of aspects which could be improved.

For such a lightweight build you need to fight for each and every gramm in weight.

  • Short the motor cables to ESC
  • Short the battery cable (is this an XT60 plug? If yes, switch to XT30)
  • Use plastic screws and standoffs for the connection of both carbon layers
  • Put the motors between the layers, so you can remove the landing gear

Don’t be lazy, minor weight savings count:

→ Thrust efficiency rises

→ Usable battery capacity rises

This is additional to the linear flight time improvements.

You need to get at least below 300g

7

Ecalc is an underrated and underutilized tool when building!

8

Agree. And if you find that you are concerned about your custom FC/ESC and the motors/props, then ask yourself if it’s possible to yaw the board 45 degrees. Keep the pusher orientation by just inverting the frame. Those long wires…. you’re probably going to drop 10g just cutting them all to length. Is a cap really needed on a 2S… save a few grams there. Can you switch to the WS Avatar Lite setup? It takes 3.1-13V and will certainly save you some weight.

I have a custom designed and 3D printed 2.5” build that is 580g AUW and getting up to 15 minutes. It’s heavy. So yes, I think you could realistically get 12+ minutes of flight.

Your prop pitch will also impact flight time. And you may find going to a bi-blade will stretch your flight time… perhaps minimally but all those seconds gained and grams dropped start adding up.

9

For this I even replenished my ecalc subscription.

Ecalc says, assuming 3030 props (3” diameter, 3” pitch), that it needs about 85% (linear) of thrust for hover on a full battery, needs about 15A to hover, and cannot even hover on a medium-charged battery.

This seems in quite a remarkable agreement with the log, where the throttle is barely enough (and one motor is always saturated, e.g. rotates at the maximum output) to hover. As the battery voltage sensor is not set up at all, I cannot see whether it’s the batteries that are dying too fast, or it’s the required throttle that goes up to unflyable ranges. But the current sensor shows something between 15 and 20 amps. I would usually not trust analog current measurements on random poorly tuned copters, because factory defaults never work and one should calibrate each such sensor, but well, this time it shows something not unexpected.

I would guess that within the hard constraint of 2S batteries, the smallest component change would be to swap for 6000kV motors of the same series, instead of 4500kV. Also, not sure which props are available, but something in the 3020 ballpark would eat your battery less aggressively, and 2-blade props even more so.

According to eCalc, same EMAX but 6000kV motors, and 3020 2-blade props would get you about 11 minutes of hover time. Then you just have to polish your design further.

And don’t forget to actually configure and tune the machine properly. Default PID gains never work, and given that each quad is different, you need to tune them yourself, thankfully, this site has a lot of guidance, and something like @amilcarlucas‘s Methodic Configurator ( GitHub - ArduPilot/MethodicConfigurator: A clear ArduPilot configuration sequence · GitHub ) may help you a lot with that.

3 Likes
10

This is a good point (they all are). Sometimes a lower pitch prop can be advantageous.

11

Ecalc.ch - Do it. It will answer a lot of these questions.

I’m working on a 3” quad on 2S. So far I’ve tested it to 15 minutes and I expect when I’m done I should be able to fly for ~20 minutes.
A lot of hours on ecalc.ch playing with different motors and props and I ended up with 1303 6000kv motors.
Weight matters. Probably one of the most important factors. Everywhere you can drop a gram or two is worth it.

12

@Allister @dkemxr @MaxBuzz @amilcarlucas @YupsUAV
Thanks everyone for the suggestions! I’ll implement them on my drone and hope to see good results. I’ll share an update here once everything is working perfectly.

1 Like