Yet another, another wireless launch controller ... this one with remote camera control

[curtisheisey]

There have been a bunch of threads lately on both TRF and Facebook on launch controllers and wireless controllers. Some time ago I designed the RocketTrig system for remote control of ground based cameras (https://www.rocketryforum.com/threa...-detection-project.152779/page-2#post-2469231 and Sport Rocketry Feb/March 2023. I had a eureka moment and realized that with basically no modifications, I could turn this into a wireless launch controller and retain the remote camera control. This thread is a discussion of the design that I came up with and main lessons learned.

I wanted a foot switch as launch trigger, so I could have my hands free for even more cameras. I wanted a launch logic that would automatically disable. I came up with an "arm" logic. Hit a button and system is armed. It times out after 60 seconds, or when you hit a "clear" switch, or when you launch. This is addition to a traditional toggle safety switch.

I put a traditional launch button on it, so I could use it without the foot switch.

It has a heart beat LED, so you know that your remote node is in range

The system consists of two nodes, local and remote. I use the same software on both nodes. The respective behavior is driven by which pins are hooked up. I use LoRa radio modules from Adafruit that have the micro controller .

I use 4S LiPo battery and an automotive relay.

Both local and remote nodes have two jacks each for camera remote control (although, ha, I don't have four cameras to use!!).

The system is not cheap ... on the order of $120 a node in parts. The main cost drivers are the Adafruit 3178 LoRa module ($34) and the pcb, which is on the order of $30.

I included one traditional wired channel on the local node.

Being an engineer and always wanting something better, I redesigned my RocketTrig pcb to better accommodate this system. It can also grow to control 8 channels.

I put all my designs and progam on github. Feel free to use or borrow for non-commercial use.
The hardware pcb files and KiCad are at schematic, pcb and fab files and program here .

The automotive relay is bulky and created somewhat of a wiring rats nest. I did design a relay board, which I am testing out that greatly simplifies wiring.

 

[curtisheisey]

Relay board.

I found a tiny automotive relay that can handle 30-amp, the Omron GN8-1U-AS DC12. It comes in 5V flavors as well. The challenge (for me at least) is designing a pcb that can handle 15-20Amp of current for short times. I used a very wide trace and had a double think layer of solder put on.

On the driver pcb, I used a UNL2803A Darlington transistor array. It has 8-channels, each of which can drive 500 mA.

Here is the schematic here . It supports 4 channels of output.

Note: the pcb image does not match the schematic, as I made some corrections but have not updated the layout yet.

Initial smoke testing, and it can withstand the current for a 2 second "on" time, but I need to do some more comprehensive as well as destructive testing.

One relay is devoted to switching between continuity and "live" for all channels. Each channel has a relay for firing. I have a SPST toggle as a safety as well.

 

[curtisheisey]

Other lessons learned:

The Adafruit 3178 module simplified the RF circuitry, as the LoRa radio module and microcontroller were on one board. I used header pins to connect this to the pcb board. However, the trade-off is cost. It would be cheaper to incorporate both mcu and radio module directly on the pcb board. I used the Arduino IDE for programming. The adafruit module is a supported board in the Adafruit board extensions. I used the radiohead library for doing the RF comms.

I mentioned the UNL2803A Darlington transistor array in the "relay" post above.

I used a MCP23017 i/o expander. This gave me 16-channels of i/o over two I2C pins on the mcu. This freed up mcu pins for other functions. There are a number of Arduino libraries that make using each i/o channel look like a regular Arduino i/o pin, so it was easy to program.

I came up with non-blocking timer classes.

I used a Tattu 850mAh 14.8V 4S Lipo Battery Pack with XT60 plug. I did use a 0.75 ohm wirewound 25Watt resistor to limit current to below 20 amps, in case of a direct short on the output.

I used KiCad for schematic entry and pcb layout. This produced a set of outputs that I sent to the pcb fab. I used JLCPCB for fab, and I used their assembly service for several of the ICs.

I considered marketing this, but wow, the $120-150 base cost for parts is a killer. If I used a model 1/3 for direct costs, 1/3 for indirect costs, and 1/3 for profit, that would put a 2-node system at $900. And that doesn't really include my design or assembly time. It's not really a model that works for me, and I would have to deal with all the marketing and customer support stuff. Not my bag, so here are the designs, info and lessons learned.

I am building a 5-channel version for one of the local clubs, but I'm doing it as a favor and providing at cost.

And I decided not to light the LED in the safety output circuit. See the discussion in this thread https://www.rocketryforum.com/threads/wireless-relay-diy-launch-controller-build-thread.170114/