Project 013 - RC Quad Copter
DISCLAIMER: This design is experimental, so if you decide to build one yourself then you are on your own, I can't be held responsible for any problems/issues/damage/injury that may occur if you decide to follow this build and make one yourself.
INTRO
After a few weeks of deliberating, contemplating & rattling they old grey matter I have finally decided on my next project........a Quad-copter! (major influence from the RCGroups forum on the subject here).
The design is obviously a 4 motor/rotor design, and with a sophisticated electronic control board & RC receiver in the centre and an array of sensors including thermal and gyro's used to stabilize the aircraft.
The advantages of the current generation of quad rotor designs, versus comparable scale helicopters, are as follows. First, Quad-copters do not require mechanical linkages to vary rotor angle of attack as they spin. This simplifies the design of the vehicle, and reduces maintenance time and cost. Second, the use of four rotors allows each individual rotor to have a smaller diameter than the equivalent helicopter rotor, for a given aircraft size, allowing them to store less kinetic energy during flight.
With a good & properly setup design they can virtually fly themselves, something I am looking to attain as I want to be able to concentrate on flying FPV, meaning I'll have a video camera attached and wearing video goggles!
Much of the following details & photos I have gleaned from RCGroups, there's is a veritable plethora of information & guys there will to help on your build.
ORDERING THE PARTS
This bit should be relatively easy, however, I don't want a beginner or easy-build setup..........I want a good, cutting edge design, and I want to be tasked during the build.
BASE PARTS LIST (a starting point anyway!):
Qty 1 - Dammar Brushless PWB - (Dammar/Spectrolutions)
Qty 4 - TP2410-09 Brushless Outrunner Motor with mount - (HobbyKing.com)
Qty 4 - 10/12 Amp ESC - (HobbyKing.com)
Qty 2 - 8mm Carbon Fiber Square Tubing - (Avicraft.co.uk)
Qty 2 - EPP1045 10x4.5 Props (Set of 2 counter rotating) - (NextGen-rc.co.uk)
Qty 4 - 3.00 mm Prop Saver - (HobbyKing.com)
Qty 1 - Receiver - (LRS 430mhz)
Qty 1 - Futaba T6EXAP -(Ebay)
1 Lot - Misc Aluminium sheet - (Ebay)
1 Lot - Misc Carbon Fibre sheet - (Ebay)
1 Lot - Misc Connectors/Cable
PWB CONTROLLER BOARD
I've settled on the Dammar Brushless PWB as pictured below and for the RC Radio control side of things it is designed to be used with a BERG RC Receiver module as pictured. However, I want to use one of my own Radio Rx modules and thus here is the problem, the BERG Rx outputs in a serial form (PPM) which is contrary to the normal multiple PWM output from a normal Rx.
Now, it just so happens that the Rx I want to use is also a custom design called the LRS System from a chap called Thomas from Denmark. It actually operates on 430MHz and is designed for long range operation (perfect for FPV). Now the good news, all Thomas's LRS Rx boards have PPM output unlike most other commercially available Rx.
The only other caveat is that the PPM order from the Tx must be as follows: Pitch, Roll, Throttle, Yaw, Landing Gear, Flaps. I'm looking good on this front.
Took delivery of my own PWB, here's some pics:-
MOTOR / ESC
TowerPro Brushless Outrunner 2410-09 13A / 104W
TURNIGY Plush 10amp 9gram Speed Controller
RADIO Rx MODULE
Here's a pic of Thomas's LRS Rx, the extra pin on the Ch.12 connector is RSSI out, Ch.12 itself is the all important PPM out to connect to the PWB.
This is the Tiny-Slim version.
RADIO Tx
I was planning on using my existing 430MHz converted Futaba T6EXP, however as Ch.6 on that Tx is a switch (digital) I had gone ahead and bought a 2nd hand T6EXAP which is more or less the same except for a nice analogue pot for Ch.6. This means I will be able to have full control over the TI (Thermal Intelligence) feature of the PWB - see Page 3 - Getting Technical.
Here's a photo of the T6EXAP. Note the pot at top right of the set.
Update: I figured it was more work to move the LRS conversion to the new Tx, so I have gone ahead moved the main board & Ch.6 Pot from the new T6EXAP to my original T6EXP. Job done!
Here's a couple of pics of the swap over. Pictured below is the new T6EXAP, the large green pcb at the bottom is the one I swapped over. Below that is the Ch.6 pot that I was after.
Note: Funny how the main board from both Pcb's look identical, right down to the model number of the pcb. I guess therefore that the code is the only difference.
GETTING TECHNICAL
Before I can start assembling the Quad-Copter I want to document much of the technical elements. Much of these details I have gleaned from RCGroups. This will no doubt be usefull to me when I first power up and start configuring the Quad.
RADIO Tx CHANNEL ASSIGNMENT
Channel 1 - Left/Right Stick, Rev Mode, 100% Epa, -50 Exp
Channel 2 - Up/Down Stick, Rev Mode, 100 Epa, -50 Exp
Channel 3 - Throttle, Rev Mode, 100 Epa
Channel 4 - Yaw, Norm Mode, 140 Epa, 0 Exp (or even +25)
Channel 5 - TI enable, Norm Mode (Thermal Intelligence on/off)
Channel 6 - TI sensitivity, Norm Mode (best to use an analogue channel, but digital will work using endpoints to set sensitivity)
Note: Set up Dual Rates on Ch.1 & 2 to 60%.
THERMAL INTELLIGENCE (TI SENSOR)
A range of 10% to 35% (Ch.6 of Tx) is considered reasonable. The higher the value the more self-levelling occurs.
When TI is on outdoors with two green LEDs, the self-leveling is very obvious.
You can't test TI by tilting the frame as your body's IR heat signature will mess up the TI, and the gyros themselves will respond to that as well.
To test TI, if you arm the quad with TI 'on' and set the sensitivity to where it is active, then, with the quad held tightly to the ground, put in a minimum throttle to get the motors spinning. Now hold your hand about 2 feet away from one sensor and slowly move it close to the sensor. You will notice the motors on that side speed up. This is a good way to individually test all the sensors.
If you get no response, check to see that your channel 5 gear switch is in the correct position to have TI on, and/or put in a bit more sensitivity with your channel 6 knob or flap lever and do the sensor hand check again.
TI checklist for outside flights:
1. TX on, throttle down, then PWB on *red LED will blink faster if throttle is moved up prior to arming indicating channel lock.
2. TI cal - Check for one or two green LEDs (preferably two green LEDs for good TI) when holding the Quad from behind, front pointed to the sky, bottom to ground. If one or two green LEDs are lit, TI flight is possible. Continue holding the Quad with front to sky and press the arming button for three seconds (insure that you are actually pressing it as there is no other indication other than the red light stops blinking and the tactile feel of it being depressed) *note 1: The green LEDs both will be on steady after this on the PWB V. The green LEDs both go off on the "experimental brushless PWB V TI"* **note 2: Do not calibrate TI over concrete/asphalt, cooler ground preferred**
3. Place the Quad on level ground, front pointing away from you.
4. Stand at least 15 feet to decrease the effect of your bodies IR image, then calibrate the TI level by moving the throttle stick down and to the right.
5. Stand behind the Quad slightly increase throttle to get props spinning, insure proper nick, roll and yaw stick response.
6. Bump throttle to 50% to get a low hover, note any trimming required to get a stable hover.
7. Land and hold the TX throttle stick full down and left, then press the arm button to set the loss of radio signal trims.
8. Fly! The battery low indication will flash the red and green LEDs simultaneously. The PWB will decrease the throttle response requiring more throttle to keep it airborne, so then land as soon as possible and recharge/replace the battery.
PRE-FLIGHT CHECKLIST
1. TX and Quad batteries fully charged.
2. TX on throttle down, then PWB on *red LED will blink faster if throttle is moved up prior to arming indicating channel lock.
3. TX throttle down and to the left (turns TI off) and hold, then press PWB arm switch. (The green LEDs both will be on steady after this on the PWB V. The green LEDs both go off on the "experimental brushless PWB V TI" )
4. Stand behind the Quad, slightly increase throttle to get props spinning, insure proper nick, roll and yaw stick response.
5. Bump throttle to 50% to get a low hover, note any trimming required to get a stable hover.
6. Fly! The battery low indication will flash the red and green LEDs simultaneously. The PWB will decrease the throttle response requiring more throttle to keep it airborne, so then land as soon as possible and recharge/replace the battery.
THROTTLE CALIBRATION
Throttle Calibration programs the ESCs for maximum performance with a particular system. The following procedure calibrates the ESCs for use with the Spectrolutions quad controller (PWB).
Note: It's debatable that throttle calibration is absolutely necessary, it's really down to what ESC's and motors that you use. If you decide not to do throttle calibration, then the only thing you may need to do is set your throttle max EPA up to say 120%.
Advantages of thottle calibration are:
- More power at maximum throttle
- Lower motor starting position on the throttle stick
- Larger range of joystick operation for smoother throttle control
- Better match of motor speeds among the four quad motors
Calibration procedure:-
1. Set Transmitter endpoints for Channel 3 throttle:
Upper Endpoint = 66%
Lower Endpoint = 95%
Also make sure that the trim and subtrim settings are at zero.
2. Remove Receiver from controller
3. Select one of the ESCs to calibrate and remove the servo connector plugged into the controller board. (Keep ESC power connector attached to the controller)
4. Plug the selected ESC servo connector into the receiver channel 3
5. Turn ON transmitter and set throttle to MAX postion
6. Connect battery to Quad system
7. Immediately after the tones from the ESCs stop, move throttle to MIN position. (You should hear another set of tones from the selected ESC.)
8. Disconnect battery from Quad system
Repeat steps 3 thru 8 for the three remaining ESCs
9. Set transmitter upper and lower endpoints back to 100%
10. (Optional) If you want the motors to start even quicker on the throttle adjust the subtrim to -50.
Note:-
If you use a couple of Y-cables you can hook up the 4off ESC's to Ch.3 of the receiver and do all 4off at the same time.
GYRO TESTING
A 'shake test' in each axis will enable the roll and pitch gyro potentiometers to be set correctly.
It is easiest to unplug two of the motors and do one axis at a time. Just turn the pot in the axis you are adjusting a tiny bit in one direction or another and then do a hard shake with TI off at 1/2 throttle (after throttle calibration has been done).
Keep doing this until there are no oscillations. It should only oscillate once or not at all after a very hard shake in that axis you are adjusting. If it gets worse move the pot in the other direction. There is a relatively small sweet spot that stops the oscillations that you will find.
Then do the same for the remaining axis. Leave the Yaw pot where the calibration mark is. You can mess with that one later if your motors and trim are proper and your heading is still not holding well.
Update:
For my own quad, I found I still never got any oscillations with all three pots set to maximum, so that's where I have left them.
The three pots are shown in this photo at the maximum setting (red dot). The original default seetings are thus that 1/2 setting for all three will be the desired setting for most people. Setting the pots below 1/2 setting will be good for small quad frames and higher than 1/2 setting will be good for larger frame quads.
Note: If you have something really extreme for the platform/motors, you can remove the three circled 51 K resistors and have an additional 2x gain range for the gyros.
ESC CONFIGURATION
For best results the Turnigy ESCs need a little tweaking in respect to the low-battery settings, i.e. it's best to disable the low-battery cut-off. Since there are 4off motors/ESCs then if the battery voltage does drop then it's more than likely that one of the ESCs will react before the others and the QuadCopter will flip. From experience, this can happen when coming down to land and the throttle is raised to level out prior to touchdown.
The default Turnigy settings are for a Lipo battery and a minimum voltage. However, if you set the battery type to Nimh then the minimum voltage can be set to zero.
WIRING
Various wiring diags, including motor/ESC arrangement.
CONSTRUCTION - MOTOR MOUNTS
MOTOR MOUNT
First step is to work out a suitable method of securing the motor to the arms, and I've decided to throw away the plastic mount that comes with the motors and have purchased 4off alloy, finned heatsink mounts.
Here's the first motor.
CONSTRUCTION - Framework/Pod
Busy day today, here's the basic framework completed.
I had bought 8mm square tube carbon fibre for the arms, but I didn't like drilling them and couldn't be bothered trying to work out a way of clamping them to the central pod, so I just bought some aluminium U-tube. Still pretty light, but of course the problem now is that if anything is going to break on impact it won't be the arms. However, I'll no doubt build another frame, so this will do fine as a prototype/test vehicle.
Distance between motor centres is about 840mm.
Below are a couple of close-ups of the central pod with the PWB installed on the top section and the protective aluminium loops fitted. The middle section will house the Rx and DakarOSD pcb's.
The PWB is raised on quite lengthy pillars in order to allow room for the ESC cables/connectors.
The protective loops are orientated in such a way so that the TI sensors have a clear & visible 'view' of the sky.
One thing I am not too happy with is the plastic pillars, I may change them to brass later on to increase strength.
CONSTRUCTION - Wiring
I mounted the PWB, ESC's, Rx & completed all the wiring. Heck!, this project was meant to last me a few months!
Total weight without battery is 790g.
This first pic shows the power input to the PWB. Unfortunately, the power connector isn't shipped with the PWB so I have hardwire it. Off to the left of the PWB you can see the 4off ESC interface connectors.
Shown here is the Rx mounted below the PWB. I'm using PPM direct from the Rx to the PWB, you can see the 3 wires in the foreground running from the PWB.
Again the Rx mounted below the PWB.
This is 1 of 4 ESC's mounted on the arms ofthe Quad.
UPDATED 12/5/10 - Current incarnation of the Quad along with home built RC Tx.
Flight systems testing
With the basic assembly done it's time for power up, testing and 1st flight!
All wiring checked, props removed (the cautious approach!) I attached a Lipo battery and success!, the PWB powered up. Then after pairing the Tx with the Rx board I managed to get the motors running with input from the stick. I'm glad to say that Thomas's LRS system (the Rx) works perfect with the PWB PPM input.
After making the necessary settings to the various channels on the Tx (see Page 3) I fitted the props and holding tightly to the bottom of the Quad I increased the throttle to the point where the Quad was just about supporting it's own weight and then gave the sticks some input and indeed felt the Quad attempt to bank left, right, forwards & backwards. Also, physically pulling the Quad down at one side I could feel the gyro's working and the Quad trying to level it off.
So, that done I tried another test. Here's a video of my first 'tethered' flight! As the Quad takes off I am only very lightly holding/guiding it as I test all the controls. Note: TI is turned off.
I've now also calibrated the Throttle/ESC's (see Page 3) and adjusted the wiring to the battery to make it easier, safer & quicker to connect/disconnect the battery and also in readiness of attaching the video camera & Tx.
Next was the 'shake test' where I tested the gyro's and all I did at this stage was verify that they work, which they do. I haven't been able to test TI as I need more uninterrupted views of the horizon than my back garden allows.
And here is one of my back garden flights:-
And here's one of my first real outdoor flight via the Quadcam. I had decided that back garden flights were over and it was time to try it for real in more open space and also get a chance to try out TI.
Have to say that TI works a charm, and I got through a few batteries before the inevitable bad landing/crash. Only broke a prop though. It seems to me that the low sun confused one of the TI sensors, you can see in the video that the last thing seen by the camera before it went haywire is the sun....directly centre of the cam.
Here's a further back garden test with my new home built Joystick RC Tx.
Night Flights
1/9/10:
Here comes some night flying!
Using some cheap 3W Led's and 3W driver circuits via Ebay I've attached them below each prop. And with some soloured gels (off an old PAR 56 can) I've managed to have green pointing forward and the rest red. This should help with orientation.
3W * 4 Led's = 12W total. This shouldn't drain the main battery by any noticeable margin.
Pic below is the underside of the Quad in daylight (inside my workshop), so I'm hoping they will light up the sky in the dark!
3W Led driver.
Input = 8 - 12v AC/DC
Output = 3 - 13.5v DC, 700mA
White Lumiled Luxeon Star 3W LED.
Here's my first (quick) night flight.
2nd night flight test with 8off 3W LED's:
