I am a graduate student building a drone for a field project. The project requires at least 8kg of payload, so I decided to build a customized one. It is my first time building a drone, but I try my best to follow all instructions and other people’s topics. Any comments and follow-ups would be appreciated.
Here are the specs:
Propellers&ESC: DJI E5000 PRO
Frame type: Hex-x
Controller: Cube Orange
Firmware: ArduCopter 4.1.5
Frame weight excluding battery: 11.25 kg
Battery: Tattu 10000mAh 25C 6S 22.2v (2 to make 44.4)
After the basic configuration and testing. I conducted a autotune with a weight of 13.5 kg (frame + batteries). The drone can fly very steadily after the tuning.
Then I conducted a payload test with an additional 15kg weight using dumbbell weights shown below. I also adjusted the ATC_ACCEL_MAX parameters according to the ArduCopter documentation(Setting the input shaping parameters — Copter documentation).
However, two problems were observed during the payload test:
- The drone has some low-frequency vibration while hovering.
- The batteries dropped under 3.5V per cell after 1 min of hovering and triggered the battery failsafe (It can fly for 12 mins with only frame + batteries). Then they came back to 3.9V per cell after landing.
My questions are:
How should I reduce the vibration when a heavy load presents? Is it because my center of mass moved much lower so the tuned controller no longer matches?
The batteries seem to discharge very fast. How should I check if I may damage the batteries? Do I just keep the voltage under load always above 3.4 V while disregarding the discharge rate?
Since the batteries came back to 3.9 V per cell after landing, is it a good idea to go lower (i.e. 3.1 V) when the batteries are under load?
Here is the log.
Thank you in advance for any kind of help!
Did you secure the weights in any way or are they loose around the legs like in the picture? If so, thats more than likely the cause of your vibrations. Looking at your log, VibeX is really high, hitting peaks of 60 m/s/s which is double the allowable limits for good control.
Also, have you got your current and voltage monitoring set up correctly? The log is showing 6S voltage and <5A current draw which doesnt seem right, I’d expect the power draw to be around 80A (~3.5kW) based on the specs of the powertrain.
Regarding the battery, was it cold? Or do you have a way to measure internal resistance of the battery? That could tell you whether the cells are any good. Is this an old battery? Id exect it to be able to keep up with the power draw, even if we take a 50% error margin on the discharge rate.
Thank you very much for your comments.
About the weights, I will secure them and do another test in the following days to see if it helps.
I am sorry about forgetting to mention that the power monitor was only connected to one of the batteries since the power monitor (ProfiCNC Hex Pixhawk 2.1 Power Brick Mini XT60 Module 8S / 30A Continuous NEW) only supports up to 8S. I will also double-check the current monitoring setting.
The temperature on the test day was about -2 Celsius with a 10 mph gust. So I would say it was kind of cold? I have an internal resistance monitor and all cells are about 4 mOhm.
That definitely sounds like your batteries were too cold. LiPos don’t like being too cold, especially below 0degC (more info here). Try and keep your batteries warm in an insulated bag/box and only take them out and connect them up when you’re ready to fly.
Thanks for the advice and information. I will find a warmer day to do my next test flight for sure.
I really appreciate your timely response and informative description!
I did another test flight on 4/3 with secured weights. The vibration issue has gone, but I have some following questions regarding the power module choice.
About the vibration:
I found the vibration was not from the weights since the vibration persisted after I zip-tied my weights to the landing gear. The vibration is actually from the vibration of the landing gear. I used additional zip-ties to build a simple triangle structure, as shown below, and then the vibration disappeared. The drone flew stably! (This would be a temporary solution since I will re-design the landing gear)
Battery and power module:
For the battery issue, I think my power module (proficnc hex pixhawk 2.1) was not correctly connected, so the readings are wrong. The power module was not connected as instructed in the manual (i.e. the power module was not between the batteries and ESC. Right now, it is only used as a voltage converter.). However, since my batteries are 25C (250A), my power module (peak current 100A) cannot handle it if I connect correctly. I did some research but couldn’t find a power module for 44.4V and 250 - 400A. I also saw some posts regarding the issue, but most of them don’t have the voltage requirement, so they can have additional batteries in parallel to lower the current. Any thoughts?
Any comments and suggestions would be appreciated.
Here is the log in case it helps.