Code port on a new hardware (https://github.com/FinistRC/rx_esc) to control two motors WITH REVERSE.
RC_RX_CABELL_V3_FHSS is an open source receiver for remote controlled vehicles. Developed by Dennis Cabell (KE8FZX) The hardware for this receiver is an Arduino Pro Mini (using a 5V ATMEGA328P at 16 Mhz) and one or two NRF24L01+ modules. Both are inexpensively available on-line. Be sure to get the version of a Pro Mini that has pins A4, A5, A6 and A7 broken out.
The transmitter side of this RC protocol is in the Multi-protocol TX Module project at https://github.com/pascallanger/DIY-Multiprotocol-TX-Module.
The protocol used is named CABELL_V3 (the third version, but the first version publicly released). It is a FHSS protocol using the NRF24L01+ 2.4 GHz transceiver. 45 channels are used from 2.403 through 2.447 GHz. The reason for using 45 channels is to keep operation within the overlap area between the 2.4 GHz ISM band (governed in the USA by FCC part 15) and the HAM portion of the band (governed in the USA by FCC part 97). This allows part 15 compliant use of the protocol, while allowing licensed amateur radio operators to operate under the less restrictive part 97 rules if desired.
Each transmitter is assigned a random ID (this is handled by the Multi-protocol TX Module) based on this ID one of 362880 possible channel sequences is used for the frequency hopping. The hopping pattern algorithm ensures that each hop moves at least 9 channels from the previous channel. One packet is sent every 3 - 4 milliseconds (depending on options chosen), changing channels with each packet. All 45 channels are used equally.
The CABELL_V3 protocol can be configured to send between 4 and 16 RC channels, however this receiver software is currently only capable of outputting up to 8 channels via PWN or PPM. This is because only 8 channels were conveniently laid out on the Arduino Pro Mini, and more than 8 channels over PPM would be too slow. SBUS output allows all 16 channels to be used.
I recommend reducing the number of channels as much as possible based on what your model requires. Fewer channels will use a smaller packet size, which improves transmission reliability (fewer bytes sent means less opportunity for errors).
The protocol also assigns each model a different number so one model setting does not control the wrong model. The protocol can distinguish up to 255 different models, but be aware that the Multi-Protocol transmitter software only does 16.
A 5V 16Mhz Arduino Pro Mini and one or 2 NRF24L01+ modules are needed for this receiver. Be sure to get the version of a Pro Mini that has pins A4 and A5 broken out (and A6, A7 too for telemetry analog inputs). The hardware folder contains a schematic and a PCB layout using a single transceiver module for PWN or PPM output; however, this version of the PCB has not been tested. I am still using the previous version of the PCB and modifying it to implement the changes included in the new version. If anyone tries this PCB version, please open an issue let me know how it works.
There is also a schematic for using 2 NRF24L01 modules for diversity. If anyone designs a board for this, please contribute it. This schematic also has the signal inverter needed for SBUS.
The PA+LNA versions of the NRF24L01+ module provide more receiver range and reliability than the non PA+LNA versions, However the less expensive modules also work OK if the on-board antenna is replaced with a better antenna, although randomly some units seem to work well and others not so well. I recommend range testing each module. These modules are also typically unshielded, but work better when shielded. Here is an outline of the modifications (TODO: add pictures)
- For the PA_LNA module that has an SMA Connector, remove the connector. A small butane torch to heat the connector works well. When the solder melts, it pulls right off
- Cut the antenna trace. On the PA+LNA module cut the trace back far enough so that no portion of the trace extends beyond the ground plane on the back of the board. For the boards with an antenna, cut the trace leading to the antenna shortly after the last surface mount component (a capacitor, I think).
- Scrape or sand away the solder mask over the antenna trace. On the boards with an antenna you will also need to sand away the area next to the trace for a ground connection. For the PA+LNA module you can re-use the ground connections that were used for the connector.
- Tin the ground and antenna attachment points with solder
- Strip back the coax and separate the braid, twisting it into a wire that can be soldered to the ground point. Tin this.
- Strip the center conductor back to within 1mm of where the braided part now ends. Tin this.
- Solder the coax to the board. Use hot glue to fix the coax to the board as strain relief.
- On the other end of the coax, strip back and remove the braided ground conductor to leave the center wire exposed to form the antenna. Leave the inner wire insulation on.
I am no expert, but based on the best understand I have been able to achieve on the theory, using RG-178 or RG-316 (both have speed factor of 69.5) the calculated theoretical coax lengths to use are:
- For the shielded portion of the coax, ideally make the length a multiple of 42.5 mm. 170mm seems like a good length.
- The length of the antenna at the end should theoretically be 32.5mm (based on center frequency of 2.425 GHz. I don’t have the proper equipment to verify these lengths, experiment for what works best for you. My tests showed 32mm seems to work better than 32.5 for a monopole antenna.
The antennas that come with the PA/LNA modules on EBay are often not very good. Get a good antenna for your transmitter. For the transmitter I use a 5dbi antenna constructed as recommend by Andrew McNeil at https://www.youtube.com/watch?v=bs8hvXGJdhM. Good RX and TX antennas make a big difference to performance.
Shielding will require using 2 layers of heat shrink tubing around the whole module.
- First solder a bare wire onto a ground point on the board.
- Encase as much of the board as possible in heat shrink tubing, leaving the antenna and ground wire sticking out, and don’t cover the pins.
- Use copper tape to wrap the module as much as possible on the heat shrink tubing, while staying away from the edge of the heat shrink. (Aluminum tape can work too, but you cant solder to it.)
- Bend the ground wire back over the copper tape and solder it to the copper tape.
- Encase as much of the board as possible again in heat shrink tubing. Don’t cover the pins. Be sure to cover all the ground wire and copper tape shielding.
These shielding steps are based partially in the article at http://blog.blackoise.de/2016/02/fixing-your-cheap-nrf24l01-palna-module/. Using the heat shrink and soldering the copper tape is a more robust implementation of the same concept.
The output method is controlled via the option byte in the protocol header. See the Taranis Setup section for more details.
- Sum PPM on Arduino pin D2 (the Roll pin). The channel sequence is AETR, more specifically Roll, Pitch, Throttle, Yaw, AUX1, AUX2, AUX3, AUX4.
- Individual PWM channel output on pins D2 through D9 in the sequence Roll, Pitch, Throttle, Yaw, AUX1, AUX2, AUX3, AUX4.
- SBUS (Experimental) output on the TX pin through an inverter. The channel sequence is AETR, more specifically Roll, Pitch, Throttle, Yaw, AUX1, AUX2, AUX3, AUX4, AUX5, AUX6, AUX7, AUX8, AUX9, AUX10, AUX11, AUX12. SBUS packets are output every 7 milliseconds.
The receiver must be bound to the transmitter. There are several ways for the receiver to enter Bind Mode. When a receiver is in bind mode the LED will be on.
- A new Arduino will start in bind mode automatically. Only an Arduino that was flashed for the first time (not previously bound) does this. Re-flashing the software will retain the old binding unless the EEPROM has been erased.
- Erasing the EEPROM on the Arduino will make it start up in bind mode just like a new Arduino. The Arduino sketch here will erase the EEPROM.
- Connect the Bind Jumper, or press the Bind button while the receiver powers on.
- The protocol has a Un-bind command (it erases the EEPROM), after which a re-start will cause the receiver to enter bind mode just like a new Arduino. After an Unbind the LED will blink until the receiver is re-started.
Turn on the transmitter and have it send a Bind packet. The receiver LED changes from always on to a slow blink when the bind is successful. Re-start the receiver after the bind and take the transmitter out of Bind mode, then test the connection.
The receiver fail-safes after 1 second when no packets are received. If a connection is not restored within 3 seconds then the receiver will disarm.
- At fail-safe, the throttle is set to minimum, and the other channels are set to the failsafe value.
- After disarm, the throttle will stay at minimum until the receiver is re-armed. To re-arm the receiver move the throttle to the minimum position.
When a receiver is bound the failsafe values are reset to the default values, which are throttle minimum and all other channels at mid-point.
Do not set fail-safe values while in flight. Due to the length of time it takes to write the new fail-safe values to EEPROM, the receiver may go into fail-safe mode while saving the values, causing loss of control of the model. Before flying a model, always test the fail-safe values after they have been set.
Fail-safe set mode will set the fail-safe values. This can be done one of two ways:
- A set-Fail-Safe packet can be sent from the transmitter. The values from the first packet in a series for set-Fail-Safe packets are saved as the new fail-safe values. The LED is turned on when a set-Fail-Safe packet is received, and stays on as long as set-Fail-Safe packets continue to be received. The LED is turned off when set-Fail-Safe values stop being received
- After the receiver has initialized, the bind button (or bind jumper) can be held for one to 2 seconds until the LED is turned on. The values from the first packet received after the LED is turned on will be saved as the new fail-safe values. The LED will turn off when the button is released (or jumper removed).
When fail-safe set mode is entered, the LED is turned on and stays on until the failsafe set mode is exited. Only the values from the first packet received in fail-safe set mode are saved (this is to avoid accidentally using up all of the EEPROMs limited number of write operations)
Values for all channels can be set except for the throttle channel. The fail-safe for throttle is always the minimum throttle.
Always remove the props when working on a model!
The receiver has a dis-arm feature to help prevent the motor from unintentionally spinning, however no system is 100%, so take care whenever you power up a model. Keep all body parts, clothing, and other objects out of the path of the prop.
Many ESCs and flight controllers have an arm sequence where the throttle must be set to minimum or some other condition to be met in order to arm. This receiver also has the following arm/disarm behavior which complements ESC and flight controller arming behavior.
When powered on the receiver starts out in an armed state. However, if no signal is detected within 3 seconds the receiver dis-arms. The receiver also dis-arms if an RC signal is lost for 3 seconds. While dis-armed, only the minimum value will be sent on the throttle channel. The receiver will re-arm when it receives an RC signal with a minimum value for the throttle channel. If your model disarms in while in use, move the throttle to its minimum position to re-arm once the RC signal has been re-established.
Powering on the model before the transmitter will cause the receiver to dis-arm in 3 seconds as long as there is no RC signal. During this power on time there is no output from the receiver until an RC signal is first received from the transmitter, so the ESC or flight controller will not arm if it requires a signal to arm.
Powering the transmitter off before the model will cause the receiver to dis-arm after 3 seconds, forcing the throttle channel to minimum.
There is a 10k pull-up resistor on pin D2 that is essential for safety when using PPM. This resistor prevents random signals from being output from the receiver while it is initializing.
Stay Safe! Only do this with props removed!
The reason the receiver initializes in an armed state to to allow an ESC to be configured with stick commands, which may require the initial throttle setting sent to the ESC to be max throttle. If you power on the transmitter before the model, then when the model acquires the signal the values received will be immediately output. This means that a high throttle will be immediately output. Do not do this if your ESC starts in an armed state!
Please refer to your ESC documentation for details on configuring it with stick commands.
Diversity is achieved by using 2 NRF24L01 modules, which improves link reliability. Try to orient the antennas close to 90 degrees to each other so at least one antenna has a good orientation to the transmitter antenna. The use of a second NRF24L01 module for diversity is optional. The code automatically detects if one or 2 modules are connected. See the "with diversity" schematic in the hardware folder for how to wire the NRF24L01 modules.
Both modules listen for an incoming packet. If the primary receiver does not get a packet when expected, the secondary receiver is checked for the missed packet. After each packet, the primary/secondary receivers are swapped, except in the case where only the primary receiver had the packet, in which case this receiver will retain the primary role. Telemetry packets are transmitted using the same receiver that was used to read the packet. If both receivers got the incoming packet, then the packet in the secondary receiver is discarded. The net effect is that the receivers alternate with each packet, except when only one receiver is receiving, in which case that receiver continues to be used for both the primary receiver and telemetry transmit.
With a Multi Protocol module installed in the Taranis, this is how to configure a model to use this protocol. These instructions for for a Serial connection using OpenTX 2.2. These instructions assume some familiarity with using OpenTX on the Taranis.
- Press Menu to go to Model Selection and Select the model you with to set up.
- Press Page to get the the Model Setup page.
- Scroll down to the Internal RF section and change Mode to OFF.
- Scroll down to the External RF section and change Mode to MULT.
- In the protocol field to the right of MULT, select Custom.
- In the protocol field to the right of Custom, enter 34. This is the protocol number.
- This number is temporary. It likely will need to change when the TX half of this protocol is merged into the main Multi-Protocol project.
- The next number to the right is the sub-protocol. For initial setup select 0 or 1. Valid values are:
- 0 - Normal usage without telemetry
- 1 - Normal usage with telemetry (please see the section on telemetry)
- 6 - Set fail-safe values (see below and the Fail-safe section)
- 7 - Unbind receiver (see below)
- Receiver No. should be filled in with the model number.
- Move down to the Option value. This value must be calculated to configure the protocol. This is done by entering a number that is the sum of the options you wish to use. Select the values from below and add them together to get the option value.
- Channel Reduction reduces the number of channels transmitted. This also reduces the size of the packet, which improves reliability. (Fewer bytes sent equates to less opportunity for error.) For best reliability reduce the number of channels to the minimum number needed for the model. Note that at least 4 channels must always be sent. Choose one of the following to add into the Option value:
- 0 - Send 16 channels
- 1 - Send 15 channels
- 2 - Send 14 channels
- 3 - Send 13 channels
- 4 - Send 12 channels
- 5 - Send 11 channels
- 6 - Send 10 channels
- 7 - Send 9 channels
- 8 - Send 8 channels
- 9 - Send 7 channels
- 10 - Send 6 channels
- 11 - Send 5 channels
- 12 - Send 4 channels
- Output Mode indicates how the receiver should output the channels. Choose one of the following to add into the Option value:
- 0 - Output servo PWM signals on pins D2 through D9 for channels 1 to 8
- 16 - Output channels 1 to 8 using PPM on pin D2
- 32 - Output channels 1 to 16 using SBUS (Experimental)
- Transmitter Power Overrides the Multi-Protocol's normal high power setting. See comments on power setting below. Choose one of the following to add into the Option value:
- 0 - Use the NRF24L01+ HIGH power setting. This is the normal Multi-Protocol module behavior.
- 64 - Use the NRF24L01+ MAX power setting instead of the HIGH power setting. This over-rides the normal Multi-Protocol module behavior.
- Channel Reduction reduces the number of channels transmitted. This also reduces the size of the packet, which improves reliability. (Fewer bytes sent equates to less opportunity for error.) For best reliability reduce the number of channels to the minimum number needed for the model. Note that at least 4 channels must always be sent. Choose one of the following to add into the Option value:
- Turn on the receiver in Bind Mode. (See receiver setup above.)
- In the transmitter Navigate to the Model Setup page.
- In the External RF section, highlight BIND and press enter.
- The receiver LED will blink when the bind is successful.
- Restart the receiver.
In order to un-bind a receiver using the transmitter, a model bound to the receiver must be configured in the transmitter. With a model selected that is bound to the receiver:
- Navigate to the Model Setup page.
- Go to the External RF section.
- Change the sub-protocol (second number after "Custom") to 7.
- The receiver LED will fast blink when the un-bind is successful. When the receiver is restarted, it will start in Bind mode.
Do not set fail-safe values while in flight. Please see the Customizing Fail-safe Values section for more information.
- Navigate to the Model Setup page.
- Go to the External RF section.
- Place all switches in the desired fail-safe state.
- Move the sticks to the desired fail-safe state. Hold them in this position until the fail-safe settings are recorded by the receiver.
- While holding the sticks, change the sub-protocol (second number after "Custom") to 6. DO not go past 6. If you even briefly go to 7 the receiver will un-bind.
- When the LED is turned on, the Fail-safe settings are recorded
- Change the sub-protocol back to its original setting. The LED will turn off.
Always test the Fail-safe settings before flying. Turning off the transmitter should initiate a Fail-safe after one second.
When the sub-protocol is set to Normal with Telemetry, the receiver sends telemetry packets back to the transmitter. Three values are returned, a simulated RSSI, and the voltages on the Arduino pins A6 and A7. A receiver module with diversity is recommended when using telemetry to increase the reliability of the telemetry packets being received by the transmitter.
Because the NRF24L01 does not have an RSSI feature, the RSSI value is simulated based on the packet rate. The base of the RSSI calculation is the packet success rate from 0 to 100. This value is re-calculated approximately every 1/2 second (every 152 expected packets). This success percentage is then modified in real time based on packets being missed, so that if multiple packets in a row are missed the RSSI value drops without having to wait for the next re-calculation of the packet rate.
In practice, the packet rate seems to stay high for a long range, then drop off quickly as the receiver moves out of range. Typically, the telemetry lost warning happens before the RSSI low warning.
The RSSI class encapsulates the RSSI calculations. If you are so inclined, feel free play with the calculation. If anyone finds an algorithm that works better, please contribute it.
Analog values are read on Arduino pins A6 and A7. Running on a, Arduino with VCC of 5V, only values up to 5V can be read. A value on A6 or A7 that exceeds the Arduino VCC will cause damage, so care must be taken to ensure the voltage is in a safe range.
Values from pins A6 and A7 come into a Taranis transmitter as telemetry values A1 and A2. You can use either of these to read battery voltage or the output of current sensor. The following article explains how to input battery voltage to A2 on an Frsky receiver using a voltage divider. The same method can be used to read battery voltage on this receiver. http://olex.biz/tips/lipo-voltage-monitoring-with-frsky-d-receivers-without-sensors.
The values sent are 0 - 255 corresponding to 0V - 5V. This will need to be re-scaled to the actual voltage (or current, etc.) in the transmitter on the telemetry configuration screen.
The Arduino development environment is required to compile the code and write the firmware to the Arduino Pro Mini. If you are not familiar with this, here are some resources to help you.
- https://www.arduino.cc/en/Guide/HomePage
- https://www.arduino.cc/en/Guide/ArduinoProMini
- https://www.youtube.com/watch?v=78HCgaYsA70
- http://lab.dejaworks.com/programming-arduino-mini-pro-with-ftdi-usb-to-ttl-serial-converter/
The receiver code can be compiled as a test harness. What this means is that instead of outputting the normal output signals, statistics are displayed on a 16x2 LCD screen about the packet success rate. To enable Test Harness mode, uncomment the following line in TestHarness.h. No output will go to servos, flight controller, etc. when in test harness mode.
//#define TEST_HARNESS
The LCD displays 5 numbers.
- Line 1: Packet success rate (percent); Number of missed packets in a row; The number of time the secondary receiver recovers a packet the primary receiver missed
- Line 2: The number of good packets received; The number of missed or bad packets
The LCD pin connections are:
LCD D4_PIN = Arduino D9 LCD D5_PIN = Arduino D8 LCD D6_PIN = Arduino D7 LCD D7_PIN = Arduino D6 LCD RS_PIN = Arduino D5 LCD EN_PIN = Arduino D4
More information on connecting the LCD is at https://www.arduino.cc/en/Tutorial/LiquidCrystalDisplay. However, I suggest using a 330 or 470 ohm resistor instead of the 220 ohms listed in the article to avoid pulling over 20 mA from the Arduino pin connected to the LED back-light.
You can also connect an I2C LCD display. This may require additional code editing and the installation of a different LCD library (from https://bitbucket.org/fmalpartida/new-liquidcrystal/downloads/). Please see the comments in TestHarness.h for more details.
typedef struct {
enum RxMode_t : uint8_t { // Note bit 8 is used to indicate if the packet is the first of 2 on the channel.
// Mask out this bit before using the enum
normal = 0, // 250 kbps
bind = 1,
setFailSafe = 2,
normalWithTelemetry = 3,
telemetryResponse = 4,
unBind = 127
} RxMode;
uint8_t reserved = 0;
uint8_t option;
/* mask 0x0F : Channel reduction. The number of channels to not send (subtracted from the 16 max channels) at least 4 channels are always sent.
* mask 0x30>>4 : Receiver output mode
* 0 (00) = Single PPM on individual pins for each channel
* 1 (01) = SUM PPM on channel 1 pin
* 2 (10) = SBUS output (Experimental)
* 3 (11) = Unused
* mask 0x40>>6 Contains max power override flag for Multiprotocol TX module. Also sent to RX
* The RX uses MAX power when 1, HIGH power when 0
* mask 0x80>>7 Unused
*/
uint8_t modelNum;
uint8_t checkSum_LSB; // Checksum least significant byte
uint8_t checkSum_MSB; // Checksum most significant byte
uint8_t payloadValue [24] = {0}; //12 bits per channel value, unsigned
} CABELL_RxTxPacket_t;
Each 12 bits in payloadValue is the value of one channel. The channel order is EART (i.e. Pitch, Roll, Yaw, Throttle, AUX1, AUX2, AUX3, AUX4). Valid values are in the range 1000 to 2000. The values are stored big endian.
Using channel reduction reduces the number of bytes sent, thereby trimming off the end of the payloadValue array.
davidbuzz on github. Although none of his code is in this project, my very first foray into DIY rx/tx was based on his esp8266_wifi_tx project.
iforce2d on YouTube. Although none of his code is in this project, the work he did with the NRF24L01 was a big inspiration.
All the contributors to the pascallanger/DIY-Multiprotocol-TX-Module project on GitHub. This is a great contribution to the RC community and houses the TX side of this protocol.
All the contributors to the nRF24/RF24 project on GitHub. This library was essential to learning to use the NRF24L01.
All the contributors to Arduino Core and standard libraries. Especially the Servo library, which I modified to adapt to the needs of this project.
The author of the PPM algorithm at https://code.google.com/archive/p/generate-ppm-signal/ I believe this to be [email protected]
The channel sequence is generated using a permutation algorithm described at http://stackoverflow.com/questions/7918806/finding-n-th-permutation-without-computing-others
Nick Gammon for his tutorials on interrupts and async reading of ADC pins at https://www.gammon.com.au/adc
My wife, for tolerating the obsession I have had with this project.
Copyright 2017 by Dennis Cabell (KE8FZX)
To use this software, you must adhere to the license terms described below, and assume all responsibility for the use of the software. The user is responsible for all consequences or damage that may result from using this software. The user is responsible for ensuring that the hardware used to run this software complies with local regulations and that any radio signal generated from use of this software is legal for that user to generate. The author(s) of this software assume no liability whatsoever. The author(s) of this software is not responsible for legal or civil consequences of using this software, including, but not limited to, any damages cause by lost control of a vehicle using this software. If this software is copied or modified, this disclaimer must accompany all copies.
This project is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
RC_RX_CABELL_V3_FHSS is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with RC_RX_CABELL_V3_FHSS. If not, see http://www.gnu.org/licenses/.