Go Down

Topic: How to power drone motor without lipo (Read 2171 times) previous topic - next topic

HermannSW

As last year I lost an RC airplane in a tree (this time >20m height at border of forest):
https://www.raspberrypi.org/forums/viewtopic.php?f=43&t=190407&p=1331388#p1331388

This time I bought a 29$ camera drone to help locate the yellow airplane, which I had restricted to a few trees acoustically before (by turning airplane motors on/off):
https://www.amazon.de/gp/product/B06ZYSFZL3

(small part from 1280x720 camera drone frame showing yellow RC airplane)

Besides helping on search part of search&rescue, I have more plans with the drone (inclusive Raspberry Pi doing 3D position determination of drone from two other PIs with a camera each and controlling the drone via RTC). My initial plan was to allow the drone unlimited flight time inhouse by replacing the Lipo with two very thin (0.2mm) cables to power on ground. And that is where I have a question here.

The drone lipo is 450mAh 25C. From what I read that means the lipo can deliver 0.45*25=11.25A for 1V.

I measured fully loaded lipo as 4.25V, and then while measuring voltage with one hand I turned drone motors to maximal speed with the other. The measured voltage dropped to 3.80V that way:



I use a 5V 3A mains adapter as input for a DC-DC step down converter:
https://www.aliexpress.com/item/LM2596-DC-DC-adjustable-step-down-module-with-digital-voltmeter-display-LM2596S-regulated-power-supply-module/32800211016.html

That converter can deliver 1.25V-37V at 3A on output side, I use it to generate 4V.


By above lipo calculation and the measured 3.8V under full speed, the lipo should deliver  0.45*25/3.8=2.96A.

I don't understand why the 3A mains adapter with step down converter is not able to power the drone the same as lipo (the electronic starts, but when trying to turn the drone motors, only two of the propellers move single digit degree and then stop). What am I missing? How do I have to power the drone via cable from ground in order to allow endless inhouse flight?
αβ, xy & L₁/L₂ positioning systems with two 28BYJ48 stepper motors:
https://forum.arduino.cc/index.php?topic=649769.0
stepper PT camera system:
https://forum.arduino.cc/index.php?topic=647703.0
https://stamm-wilbrandt.de/en/Raspberry_camera.html

MorganS

Try adding a big capacitor in parallel with the power supply. Maybe 1000uF to 10,000uF. Put it as close as possible to the drone, like directly on the battery leads.
"The problem is in the code you didn't post."

slipstick

That's not how you calculate the current coming out of a battery. 450mAh @ 25C means it can provide 11.25A maximum (at least in theory)...not at 1V but at it's normal voltage. To calculate the current based on the battery volt drop you would need to know the effective impedance of the battery which you don't. It's usually easier and a lot more useful to measure the current.

So basically the lipo can provide up to 11.25A. Your power supply can provide around 3A. It's not exactly surprising that it's not good enough.

Steve

HermannSW

> 450mAh @ 25C means it can provide 11.25A maximum (at least in theory)...not at 1V but at it's normal voltage.
>
Thank you for correcting, nominal voltage for that lipo is 3.7V, and lipo can provide maximum of 11.25A at 3.7V (41.625W).

I have a laptop mains adapter with 20V 8.5A, will try that.
Just ordered step down converter that can do 12A:
https://www.amazon.de/gp/product/B00HV4EPG8

Will test on Monday whether 8.5A are sufficient to let drone fly.

P.S:
Given that I burned normal breadboard cables with some amps for motors in the past, the idea of using very light 0.2mm cables seems unrealistic if pushing definitely more than 3A through. Best would be cables that are light weight and still can deal with say 10A ...
αβ, xy & L₁/L₂ positioning systems with two 28BYJ48 stepper motors:
https://forum.arduino.cc/index.php?topic=649769.0
stepper PT camera system:
https://forum.arduino.cc/index.php?topic=647703.0
https://stamm-wilbrandt.de/en/Raspberry_camera.html

MarkT

#4
Jun 25, 2018, 04:10 pm Last Edit: Jun 25, 2018, 04:12 pm by MarkT
Give up on the idea of a drone being able to lift the extra weight of a high current supply cable, its
not going to happen.

11A at 3.7V is about 0.3 ohm load.  thus the wires should be at most 0.03 ohms, which for 10m length
would mean perhaps 2kg of copper or aluminium in the cables...

Low voltage power doesn't travel long distances without stupidly thick cables.  The only hope for tethered
flight is to send the power at a higher voltage and buck-convert on the rotorcraft - a high power buck converter
may weigh a lot more than the LiPo battery.

So the correct approach is to fly from battery, and tether using fishing line, or spectra/dyneema kite line, which is much much lighter.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

HermannSW

#5
Jun 25, 2018, 06:49 pm Last Edit: Jun 25, 2018, 06:54 pm by HermannSW
Give up on the idea of a drone being able to lift the extra weight of a high current supply cable, its
not going to happen.
That sounds reasonable.

Today the DC-DC 10A stepdown converter arrived.
I connected 20V 8.5A mains adapter to drone with it, 4.19V without load, that dropped to 3.73V with drone connected.
This is better what I tried before, this time all 4 propellers turn, and they turn longer than last time, but after a second they stop.
So even 8.5A is not enough current to power the drone, as can be seen in this youtube video:
https://www.youtube.com/watch?v=jcu_DHSwb20


P.S:
Quote
The only hope for tethered
flight is to send the power at a higher voltage and buck-convert on the rotorcraft
Not the only hope -- this drone does fly without any battery, gets powered wirelessly!
https://www.youtube.com/watch?v=1_10oKH6z4A
αβ, xy & L₁/L₂ positioning systems with two 28BYJ48 stepper motors:
https://forum.arduino.cc/index.php?topic=649769.0
stepper PT camera system:
https://forum.arduino.cc/index.php?topic=647703.0
https://stamm-wilbrandt.de/en/Raspberry_camera.html

MorganS

Remind us how you are converting the 20V down to the LiPo voltage? That behaviour sounds more like you are overheating the voltage converter.
"The problem is in the code you didn't post."

HermannSW

#7
Jun 26, 2018, 12:54 am Last Edit: Jun 26, 2018, 12:55 am by HermannSW
Remind us how you are converting the 20V down to the LiPo voltage?
As can be seen in youtube video and screenshot above I use the step-down converter I pointed to (German text, google translate will help):
https://www.amazon.de/gp/product/B00HV4EPG8


I connected the 20V 8.5A from mains adapter to input of converter, and connected its output to drone directly. I used the screw of the step-down converter to make output 4.19V from 20V input.


Quote
That behaviour sounds more like you are overheating the voltage converter.
I checked that, both heat sinks do not even get wam.
αβ, xy & L₁/L₂ positioning systems with two 28BYJ48 stepper motors:
https://forum.arduino.cc/index.php?topic=649769.0
stepper PT camera system:
https://forum.arduino.cc/index.php?topic=647703.0
https://stamm-wilbrandt.de/en/Raspberry_camera.html

Southpark

#8
Jun 26, 2018, 02:23 am Last Edit: Jun 26, 2018, 02:27 am by Southpark
with two very thin (0.2mm) cables to power on ground.
Did you take your multimeter and measure the total resistance of your supply and return wires? Eg......  how many metres of wire is it? And what's the measured resistance along those lengths.

If you use a big hefty bench power supply, and if your cables have low enough resistance, then the flying device should be able to at least do something good. The rest would be up to the capabilities of the step-down converters that you're using.

MarkT

How is that tiny 'copter supposed to lift that huge buck converter then?

As I said, its not feasible, use a LiPo and some fishing line as a tether if you need a tether, and
swap batteries as needed.

Just the drag and torque forces from lifting a long wire are likely to mess up the flight stability too - I really
don't think its going to work.

If you want to run from external power you need high power high voltage low current motors
on the 'copter to allow very fine wires - and no-one to my knowledge makes such things.

Alternatively you'll need to find some cutting edge GaNFET buck converter module that can
handle enough power in a tiny lightweight package.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

HermannSW

#10
Jun 26, 2018, 02:42 pm Last Edit: Jun 26, 2018, 02:52 pm by HermannSW
Did you take your multimeter and measure the total resistance of your supply and return wires? Eg......  how many metres of wire is it? And what's the measured resistance along those lengths.
Thank you that guided me into the right direction. The long cables I used to connect step-down converter to drone had several Ohms each. Replacing by shorter cables now all 4 have 0.3Ω each.

How is that tiny 'copter supposed to lift that huge buck converter then?
Of course the buck converter has to stay on ground.

Quote
If you want to run from external power you need high power high voltage low current motors
on the 'copter to allow very fine wires - and no-one to my knowledge makes such things.
You are right with high voltage but wrong with nobody does it. There are many hits on youtube for "tethered drone" and many mention unlimited flight time:
https://www.youtube.com/results?search_query=drone+thether

Quote
Just the drag and torque forces from lifting a long wire are likely to mess up the flight stability too - I really
don't think its going to work.
I was sceptical first, but with Southpark's and your previous posts regarding cable resistance I tried again, and succeeded!
https://www.youtube.com/watch?v=CLYXHWaOv8s



The cable is too short, but hey, this is first flight ...

The lipo now not on drone does weigh 12.8g, the small cables from buck converter to drone connector do weigh 0.5g.
At least 12.3g of additional cable should be fine for drone to lift tethered.
αβ, xy & L₁/L₂ positioning systems with two 28BYJ48 stepper motors:
https://forum.arduino.cc/index.php?topic=649769.0
stepper PT camera system:
https://forum.arduino.cc/index.php?topic=647703.0
https://stamm-wilbrandt.de/en/Raspberry_camera.html

MarkT

My point is it won't scale to anything but a very short cable, as the cable resistance has to be very low.

The longer the cable the thicker it has to be, such that the total mass of copper goes up as the square of
the cable length, rapidly becoming completely unworkable for these impedances.

There is one approach that can work, which is to provide a constant current supply and thin wires.  On board
the copter you simply dissipate unused current in what is effectively a high power zener diode, and the wires
run hot due to the heavy losses.  The supply voltage has to be larger for longer cables, but the cable
thickness is constant.

However that still means the cable mass goes up with length - the only real approach is to choose
higher voltages for longer cables, then you can use finer and finer wires.

And dissipating a dozen watts or so has its problems, but at least you get good air-cooling in a 'copter.

It might be working looking at what PoE modules are available these days, PoE has the same problem
set to overcome, transmit power efficiently over long thick wires and convert to usable voltage at the
far end.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

slipstick

At least 12.3g of additional cable should be fine for drone to lift tethered.
Fair enough...provided you add that extra cable without increasing the overall resistance. Which means you don't get very much extra length because the longer the cable is the thicker and so heaver it needs to be to keep the resistance down.

Steve

vinceherman

So, there is a whole lot of theoretical solutions with fairly severe constraints.
But every real-world drone like this runs on batteries.
So run on batteries.

Want more flight time for indoor use? (which I think is the original point)
Get more batteries.

HermannSW

#14
Jun 26, 2018, 07:31 pm Last Edit: Jun 26, 2018, 07:35 pm by HermannSW
Want more flight time for indoor use? (which I think is the original point)
Get more batteries.
Good point.

I tried 60cm 0.2mm cables, and while all 4 rotors rotated, not enough power to lift. And rotation became slower and slower until stopping after 8 seconds.

Next I tried 60cm 0.5mm cables, with that propellers did rotate with more power, but not enough to lift. Rotation became slower and slower again, but ended after more than 15 seconds.

Then did cut the cables and here drone to buck converter cables are of length 39cm. I measured the cable resistance as 0.4Ω each. This time the drone flies (I did not control it good). Again propellers got slower and slower, but end only after more than 30 seconds:
https://www.youtube.com/watch?v=ognVuX01AXc



The question I have is why does propellers get slower and stop? No smell, cables seem to be fine, nothing gets hot or even warm.

If I disconnect mains power, wait some seconds, power again I can repeat this experiment. I assume it is nothing in the drone that is responsible for slowing down and stop. So is the buck converter responsible, or the 39cm cables?

P.S:
Cable length of 39cm would be totally fine for flying inside my motor test station with two cameras for drone 3D locating as mentioned before.
αβ, xy & L₁/L₂ positioning systems with two 28BYJ48 stepper motors:
https://forum.arduino.cc/index.php?topic=649769.0
stepper PT camera system:
https://forum.arduino.cc/index.php?topic=647703.0
https://stamm-wilbrandt.de/en/Raspberry_camera.html

Go Up