Quadcopter details:
-Custom CF frame built from 12mm tubes and 1m~3mm CF plate.
- RCTimer 17x5.5 CF Props
- Turnigy Plush 12A ESC's running BLHeli firmware
- RCTimer 5010 - 360Kv motors
- Panasonic NCR18650 13,600A battery pack
- Bareduino Arduino FC with MPU6050
- OrangeRX DSM satellite RX
- AUW 1296 grams with battery
- Frame AUW is 499 g
Everything is COTS (standard, off the shelf) equipment.
More details to be posted on the massive thread and amazing duration quadcopter's built and posted by EoD here.
RCGroups current all-time duration multicopter thread
- RCTimer motors had original 18awg wires replaced with 24awg bringing the total weight to 79.9g
- Props were sanded and balanced, motors balanced.
- Motor-prop screws are 3mm aluminum.
- Motor screws are 4 x per motor and nylon. All 16 screws are less than a couple grams.
- Super light frame came in at 66 grams
- Barduino Arduino board from seedstudio along with an MPU6050 breakout board was used to build a tiny Multiwiicopter flight controller.
- A Spektrum satellite RX was used for the radio link. A larger MWC board was used to tune the flight parameters first then the bareduino was installed. You have to load the firmware, edit the PID"s using the MWC software, then again reflash the board to enable Spektrum support on Arduino's with one shared serial port.
- The Plush ESC's are running BLHeli firmware, and were stripped of one voltage regulator each. Heatsinks were added.
- Most "data/esc" wires are ~36awg.
- The battery pack is the awesome Panasonic 18650 cells as posted by EoD on his duration quadcopter flights.
The flight:
- It started at about 11:30am. Battery pack was charged at 2A for about 7 hours. Rested for 30min, then flown.
- I installed a small battery voltage monitor so I could land before the pack got to low. 5grams, cost me a minute.
- About 30 minutes in I was at about .400mv down, I had about 4.2V of power. Looking good!
- 1 hour in. Still lots of power, however the voltage started to drop a bit faster. Every 0.01v ticked by and I could tell it didn't last as long as when the flight started.
- 1.15 minutes in. Still looking good, starting to realize maybe we won't make 120 minutes.
- 1.25 minutes in. Wow we are dropping faster and fast, I still need 11 minutes. After flying for almost 1.5hours I hoped I would not be seconds short of the record. Going for it!
- 1.30 minutes in. I can't remember where it was, 3.15 per cell I think. But I was able to quickly calculate how long a 100mv was lasting and what I had left to 2.7V (my personal cut-off point) and it I'm not sure if I will make another 6! minutes.
1.35.45 . The longest 5 minutes, 45 seconds ever.
- 1.36 minutes 2.80v per cell, some at 2.79, one at 2.75. I knew here I beat the current "hi-score" record! Pushing ahead.
- 1.37 minutes Started hitting 2.7 on some cells, decided to land. According to EoD he had a "couple" minutes at this point. Not wanting to damage my battery I landed it here.
Could this design fly longer? 100 minutes? maybe. I could remove the heatsinks on the ESC's, they didn't add any minutes and 'cost' me 5 grams of time, copper enameled wires, direct solder the battery instead of connectors, and not use the battery voltage monitor. This would save at least 20~30grams and I think you could safely fly to over 100 minutes. Maybe 101!
The 97 minute video!
https://www.youtube.com/watch?v=2ScZ8zDsVvk&feature=youtu.be
Some pictures inserted into the video. Any questions just ask. Sorry for the "Alien Autopsy" quality video it was hard to capture this indoors in poor lighting and it has been to windy outdoors.
contact: cptfrazz(at)gmail
Comments
I made carbon motor plates and glued them into the tubes.
hi Lan
how did you attach the motors to the carbon rod?
2 seconds max when soldering to the batteries. I use a large tip, and a high quality soldering station, an OKI MFR. If you use something that is underpowered you will spend a long time trying to heat up the battery tip to take solder and could damage the battery.
Please post your copter when it's built! are you building a long duration design?
@Ian
Copper braid. I remember having read that now that you mention it. The spot welding is likely better for a production line and for commercial batteries since it would be more rugged. Maybe? I was worried about the heat from soldering but, I'll look for the posted method. If properly prepared and executed there shouldn't be too much heat involved. AND I can start building NOW! :-D
@u4eake, thanks for sorting out the motor construction questions, I was going to post tonight but you beat me to it and with some great details.
For the battery pack construction, if you play the video, one of the slide in the beginning shows a close up of the battery pack assembly. I did not spot weld the cells. I used pure copper braided wire. Has flex, low resistance, light weight. Spot welded would maybe be better but I didn't go through the trouble or test that. I might have a closer look later and see if this is somewhere I could drop weight and increase efficiency.
Weather permitting, I'll be going for a long flight this weekend, hoping for a better time. Have made several small (hopefully) improvements over the past week.
Plastic revolutions per minute (rpm) - put on stress to a 3D print which would be thin walled. If the print is not good enough or the person printing does not post process the print with acetone to assure a good fusion of all layers then the bell would come apart. And since we print with ABS and PLA generally then the first crash and you are likely looking for magnets strewn across the crass field.
--- Iron pressed inside the bell? Good info.
Now, back to long flights. ;-)
The bell shouldn't be just any metal, you need an iron bell to close the magnetic circuit and keep the magnetic flux inside the motor, where it can do work. A motor with a complete plastic bell would probably work, but would be very inefficiënt. the weight penalty of the iron bell is easily gained by the increased motor efficiency.
Because the iron bell conducts the magnetic flux back into the motor, the magnets attract STRONG inside the motor, but very weak outside the motor, and stronger magnet force inside the motor means better efficiency. Take an outrunner apart and test the magnet attraction with a screwdriver or piece of iron. You'll easily notice that the magnets are multiple times stronger inside the motor then on the outside of the bell.
In fact, if your outrunner stongly attracts iron on the outside of the bell, you can make it more efficient by putting an iron cylinder over the bell, thus making it thicker. I have done this.
The aluminium motors only have a aluminium top of the bell. The actual cylinder where the magnets are glued on is always iron (pressed on the aluminium top). Take a closer look next time you take that aluminium motor in your hands ;-)
https://www.youtube.com/watch?gl=BE&v=EmVcsEmcIm4
i would worry if he rmp was not so low, giant props and low kv means low rmp props and little to worry over
I think my motors are aluminum.
With a 3D printed plastic bell I'd worry about imbalance and it flying apart at the fast rpms. Also, a wreck would leave a mess of plastic bits everywhere and lost magnets.
-
5
-
6
-
7
-
8
-
9
of 10 Next