Arducopter Wiring Diagram
Rosella Brain
Here is a diagram of the radio settings of a Turnigy 9XR for use with APM/Pixhawk. Could also be used with any radio with ER9X or OpenTX firmware such as the FlySky Taranis or Turnigy 9X or Flysky TH9X.
Also featured is a wiring diagram for PPM. PPM is great for reducing the amount of wiring. Rangelink, EzUHF, OpenLRS are all good PPM capable 433mhz 1W systems. If you are unsure if your receiver is PPM capable then use the DSM2 selection on menu 1 of the 9XR and add a wire from each channel of the receiver (1,2,3,4,5 and so on) to the APM inputs.
This blog will be written in successive posts as it is a fairly long subject and my time is quite limited. It will also allow others to interact and add complementary input between my posts; to be enriched progressively.
Cf. figure 1 above illustrates a general wiring diagram for powering a typical drone. It illustrates 8 motors and ESCs but would remain applicable for fewer motors (battery voltage would probably go down to 3S or 4S for quad motors).
-As illustrated in figure 1, a first objective is to power ESCs and associated motors with a high power source in terms of amps/volts, able to sustain the maximum throttle consumption and voltage required by the motors and propellers configuration (with a margin is even better). A good way to dimension this for your own setup is to either measure real values on a test bench (not so practical because it implies you would already know what to do and what pieces to buy), or more practically to use an online tool which is not so bad to give rough estimates here : =en
-A second objective is to power all other components that are not part of the high power gear, namely APM and all other bits and pieces around it: a receiver, a telemetry unit, a GPS/compass module, etc. Optional servos that are require to power for gimbals for example ARE NOT low powered items; they CANNOT be powered through APM but must be powered directly from the high power source (via UBECs). As far as electronics is concerned, the objective is to feed a very stable and reliable power source at 5 volts (there may be other specific FPV/OSD components that require other voltages such as 12v or 7v but I exclude these components from this blog for now). This second objective seems apparently easy to reach but it is actually a very sensitive and tricky one as we will see in more details later.
-There is better one main battery as the power source for everything (which avoids different ground references, avoids potential ground loops, makes your build lighter). So we have a constraint to split the power chains to different voltages/amps (main voltage for ESCs and motors, and 5V for APM and associated electronics). These LIPOs are power monsters people should be aware of: they are capable of delivering very high amperage (e.g.: more than 100 amps in my X8 drone application) at a set voltage between 3S (11.1V) and 6S (22.2V) for most drones. If you make the math, we speak here of order of magnitudes of between 1 to a few kilowatts of power!
-The main power source must feed high amps and high voltage to the ESCs+motors. We have a constraint to measure both amps and voltage fed to the whole drone as it is a critical piece of information you want to check continuously on your GCS (or goggles) while flying. It would be insane to fly without such real time information, as you do not want to let your copter fall out of the sky once your battery becomes empty.
An assembly constraint is thus to wire every bits and pieces in such a manner that your current and voltage sensors measure the TOTAL (SUM) of all amps consumption. If you use two batteries , do obviously not connect your sensors after one of the two batteries. Another classical mistake is to connect items pumping current and voltage from the sensors wires!
Avoid connecting items in such a way you create ground loops. Connect everything in a STAR topology, with the center of the star connected right after the current/voltage sensor on the high power leads/wires. As it is difficult to visualize your wiring in the practice, make a wiring diagram of your connections and check it out (like I did on figure 1 for example)!
As you will notice on figure 1, I kept a ground loop with the ground wire of the Attopilot module, but it is impossible to connect otherwise. It is one of the weak point of Attopilot sensor usage, unfortunately. Some posts have been done about this here : -volt-current-sensor-significant-parasitic-ground
-At the same time the main power source must be derived to feed a set of sensitive electronics at 5 volts. APM has a constraint to be fed at a voltage between 4.6V minimum and 5.25V maximum. If you are below the minimum voltage, your APM will brownout. If you are above you will fry APM. However it is not so easy to get a stable 5V source as the load varies. Take a look at the picture below that illustrates how unstable an APM 5V power source could be (and there is even worse):
The illustrated version is the 90 amps attopilot. There exists also a 180 amps version. In practice, the attopilot is made so that the current sensor pin (I pin on the attopilot board) outputs a voltage between 0 and 3.3V (3.3V would mean 90 amps are measured on the I pin). This I pin is connected to an APM A2 sensor pin (see figure 1). APM measures on a scale between 0 and 5V. This in fact gives a possibility to measure a maximum current of more than 90 amps, that is (5x90/3.3)=136 amps maximum.
As you see on Fig3 n1, I tried to keep the integration of the attopilot module between the main battery leads and the PDB as compact as possible. This means it was impossible to use shrink tube to wrap the Attopilot and connections, because the wires being so short do not allow you to solder and place at the same time a piece of shrink tube between Attopilot and PDB. How to solve this? Use Plastidip:
It is a liquid plastic/rubber material that you apply with a brush on your parts. You can also dip your whole circuit in it. After drying (it does dry fast), the coating really shrink tight on the covered element. It makes a really nice alternative to heat shrink.
Note that you can choose where to position the different elements around the PDB. I specifically positioned the two main battery leads with the attopilot circuit on a position corresponding to the back of the drone. This will avoid battery wiring to come in front of the cameras I will use.
It is a good practice to mark with a permanent marker every useful information on the different components. For example, indicate with an arrow where the front is, number all of your motors/ESCs, etc.
-Fig3 n2: the switching regulator to feed APM. You need a switching regulator that will be robust and reliable and that provides a stable 5V to APM, even under load. APM uses a few hundreds milliamps at most, so you do not need to feed APM with a 3A or 5A power source.
Initially, I tried these Polulu switching regulators which I DO NOT recommend because they produce a very unstable 5V voltage on APM. I tried two models, one with fixed 5V output and another one with adjustable output: polulu 2107, polulu 2103.
It is a pity because they would be extremely compact and have nice theoretical features : takes an input voltage between 7 V and 42 V and efficiently reduces it to 5 V while allowing for a maximum output current of 600 mA.
A 2A (5V output) can be achieved without additional cooling, the efficiency is up to 90%, the switching frequency at 150kHz. The module is protected against short-circuit (10 seconds), but not against reverse polarity.
This switching module is ready to use out of the box (which is not the case of the Polulu because you still need to solder a capacitor on it). You just need to solder the wires that will connect to the PDB on one side, and on the other side you solder the two wires that will go on APM.
You will notice in Fig3 n2 that I twisted the two wires that will connect on APM and used a ferrite ring. You will also notice that I did not twist the last two inches or so of these wires that will connect on APM in order to have minimum vibrations transmissions to APM through the wires (when you twist the wires they become more rigid and they transmit more vibrations).
Number of all your ESCs with a permanent marker. Place them around the power distribution board before assembly. The order of the ESCs must match the motor positioning as defined in the wiki. It depends on your setup: a quad, a X8, a hexa, etc.
To limit electromagnetic disturbances to the maximum, use as short as possible power leads between ESCs and PDB. Some people prefer to use bullet connectors between the ESCs and PDB . I prefer to solder them directly on PDB as this is a much more secured way to avoid bad connections or even disconnections in flight (vibrations do wear connectors)!
Define very carefully what the length of your ESCs power wire should be BEFORE cutting them. For example if you have a folding frame and you need to flip arms, maybe you would need a bit of wire slack.
In my case I decided to use the aluminum arms as heat radiators for the ESCs. Therefore I need to glue the flat face of the ESC (the face corresponding to the integrated aluminum plate) on the arms. That meant that I had to twist the power wires on 4 out of the 8 ESCs (see figure 3).
You should use only 4mm bullet connectors (or bigger diameter) to connect your motors to the ESCs. Nowadays, most ESCs and motors are delivered by default with smaller bullet connectors. They are very dangerous to use when too short and too small as they really could disconnect in flight due to vibrations or by accident if a cable is pulled a bit too much. With 4mm, they are perfectly fine.
93ddb68554