Can Commodity RC Controllers Stay Relevant?

Visualize some radio controlled airplane fanatic of yesteryear, with the requisite giant controller hanging from a strap, neck craned to see the buzzing dot silhouetted against the sky. It’s kind of a stereotype, isn’t it? Those big transmitters were heavy, expensive, and hard to modify, but that was just part of the challenge. Additionally, the form factor has to a degree remained rigid: the box with gimbals — or for the 3-channel controller, the pistol-grip with the big pot that looks like a cheesy race car wheel.

With so much changing in RC capabilities, and the rise of custom electronics across so many different applications, can commodity RC controllers stay relevant? We’re facing an age where the people who invest most heavily in RC equipment are also the ones most likely to want, and know how to work with customization for their rapidly evolving gear. It only makes sense that someone will rise up to satisfy that need.

Clunky Junky

The RC hobby is rife with bad hardware and software. It hails from an era that valued cheapness over openness, with little in the way of standards. Every manufacturer has their own way of doing things.

Like a lot of expensive hobbies, manufacturers have placed no value in openness or customization in RC — maybe complexity was an asset, giving participants a sense of eliteness when they conquer the interface. Basically everyone who couldn’t program their VCR without a manual was excluded from participating in RC.

Let me illustrate the suckiness of commodity RC gear in a nutshell: beep menus. These are multi-level menu systems that use sound as the only feedback mechanism and are almost universally loathed.

A lot of RC gear is difficult to work with because it hails from an era when people really didn’t use computers (let alone mobile devices) to interface with and modify their RC components. Lower-end RC — the cheap stuff — presupposes no computer at all. “Beep” menus are the reality of working with “headless” components like Electronic Speed Controllers (ESCs).

When you go into configuration mode, it starts listing off the menu options, one at a time, with different tones or quantity of beeps signifying different options. You use your control sticks to navigate the menus. It’s imperative that you pay close attention so as not to jump into the wrong menu and start reconfiguring something you’ve already finely tuned.

Beep menus can be navigated through special hardware programming rigs, where LEDs signify the various menu options. There are also software packages that do the same thing, though some on the lower end have really sketchy PC-only applications. Who wants to run an .EXE they downloaded off some mystery website? Not I. And you’re not solving the problem, just enlisting a machine to deal with it on your behalf without any hint of added functionality.

Don’t Hack Me, Bro

Most RC manufacturers do not want you to hack or modify their equipment in any way, aside from navigating menus. In all fairness, most commodity electronics work the same way. Do you demand to hack the firmware of your toaster? But the key difference between drones and toasters is that there is a huge base of hackers looking to play with RC hardware. Can anything be bigger than drones right now?

It seems like for commodity RC equipment, the sole purpose of software is to rope people into ecosystems. Permitted modifications are along the lines of reconfiguring sticks based on which one you want for throttle, ailerons, rudder, and elevator. To make matters worse, commodity controllers are designed for specific formats — the quad, the rover, the helicopter, the plane. If you change your format, you have to make modifications to your hardware as well. For instance, the throttle stick doesn’t auto-center the way the rudder/aileron controller does, and some have ratchets to help keep them in place. If you wanted to use that stick for something else you have to disassemble the case and hope the stick can be modified.

All of this is predicated on a plane-centric configuration, but what if you have some sort of project that doesn’t resemble that sort of thing? Forget quadcopters and fixed-wing planes. You should be able to control anything with a RC controller.

Just hack it.

Replace or Reflash

Needless to say, hackers have embraced the challenge of taking these controllers and making them work better. At its simplest level, a transmitter is just some control sticks, switches, and buttons hooked up to a microcontroller and transmitter. Maybe there’s an LCD. When dealing with mystery chips, often it makes more sense to replace just the motherboard and not the hardware. We’re not talking about robust processing power for the most part. Even something as basic as an Arduino UNO has a lot of capability.

There are a lot of RC-control Arduino projects floating around, for instance.  [Ryan Boland] used an oscilloscope to figure out how his transmitter and receiver worked, basing his project on the RC Arduino folks writeup on using an UNO to interpret RC receiver signals. If your microcontroller can listen to analog and digital pins and control ESCs, it can run a drone.

The Hard Work is Already Done

That said, the more commonplace models of transmitter are the ones more likely to be hacked. Having an existing knowledge base of pin diagrams and forum discussions makes everything easier. And there are some fun projects out there.

[Cyberman] radically modified his HK-T6A with fine-tuned controls and OLED supported. On the Turnigy side, their 9X controller is one of the most popular, featuring an LCD display and 9 channels. [dave1993] is building his own version of the 9X in a project he calls core9x.

Simply adding an ESP32 board to a controller’s guts could change the way you fly right away. Another useful project is the DIY multiprotocol TX module, which aims to allow transmitters to interact with receivers of different brands, rather than being stuck in the transmitter manufacturer’s ecosystem.

Firmware Projects that Unlock Latent Features

However, not everyone wants to go to all that trouble. It turns out there’s an even easier way: reflash the firmware! A lot of limitations placed on radios are software driven. For instance, a 3-channel radio could have the same processor and RF module as a 9-channel radio.

A lot of medium-plus quality radios have fairly robust chips, meaning that it makes more sense to reflash the microcontroller’s firmware while it’s still in the case. OpenTX is such a project, with firmware created for ARM-based transmitters, including the popular Turnigy 9x. It even has been chosen by FrSky to be the new operating system of their $200 Taranis 9x radio, proving that open source can work with industry.

OpenLRS is another open source firmware project, focusing on OrangeRX and Hawkeye transmitters.  Meanwhile, DeviationTX focuses on Walkera Devo radios. Another cool project seeks to reflash the FlySky’s M0+ chip — my favorite part of these projects is where the participants laboriously map all the microcontroller’s pins — 64 on the M0+’s Freescale MKL16Z64VLH4.

Reflashing makes it possible to switch RF modules on the radio, change the number and purpose of channels, and all those other things that help regular folks control one configuration of drone but that hackers find terribly limiting.

Meanwhile, Commodity Radios Keep Evolving

Let’s get back to the central question. Can RC controllers remain relevant?

It’s possible that old school transmitters are just a little stodgy. Skill with the sticks has taken a backseat to flight-controller-driven flight (drone racing excluded) and as components continue to shrink, some of the larger boxes begin to resemble an old school remote control where the quality is based on how many buttons it has, not the utility of the interface.

Laptops and mobile devices have expanded their role in drone control, highlighting the fact that video is playing an increasing role in the business. In response, Turnigy created the game controller-style radio like a Turnigy Evolution (seen to the right). It has a big screen, it’s light and handy, and it’s perfect for FPV. It has few of the buttons and switches associated with landing gear and lights and so forth, relying on a touch screen for those features. It has a five-model memory, allowing you to switch between models. It has a USB port so it can control flight simulators and charge its batteries. It’s only $40 on Hobby King, making it a tempting purchase.

That said, the Evolution packs the usual factory firmware and remains closed. As nifty as the Evolution might be for that price point, it still won’t be able to reach its full potential as long as it’s locked up. If only these radio manufacturers take a hint from so many companies that have to come to embrace open source hardware and software! Lend your voice to this, let us know the important ways open source controller firmware could change the way we fly by leaving a comment below.

[hacked transmitter photo by syvwlch, Creative Commons.]

42 thoughts on “Can Commodity RC Controllers Stay Relevant?

  1. I’m not quite sure what the point of this article is… “Can commodity RC controllers stay relevant?” is the leading question, relevant to who? Why are the becoming irrelevant? Who says they are a commodity? I don’t get it.

    FrSky is the most popular transmitter brand in the FPV community, and quickly eating Spektrum’s and Futaba’s lunch in the general aviation community. It runs OpenTX out of the box, and the Q X7 is less than $100. Kinda seems like the boat has already sailed on this conversation, and FrSky+OpenTX is the winner. FrSky+OpenTX+TBS Crossfire if you need longer range or lower latency.

  2. The other popular open-source TX firmware you should know about is er9x.
    Regarding TX’s, the 9x is based on an 8-bit Atmega, but STM 32-bit chips are where all the interest is now.
    The main problem with the Atmega isn’t so much its 8-bittedness, but rather the lack of flash space.

      1. getting deep is not that interesting, unless you drive a ROV-type sub.
        for scale models it is more of a challenge to keep periscope level (we use PID controllers for that).

        I might be wrong, but I go by
        – 3 meters in chlorinated pools (this is often bottom depth anyway)
        – 5 meters in clear sweetwater
        – 0 meters in seawater.
        It seems dependant on the ions/particles in the water. We notice because loss of signal causes the sub’s
        tanks to empty (IF you design it properly and include this feature of course..).

        I just cherish my old transmitter, Not sure if new ones can still be obtained.
        I’ve noticed efforts to build open R/C using the 433 MHz (and the 8xx MHz) free bands.

    1. I don’t see any reason an open source drop-in module, be JR or Futaba style could not be made for 40MHz, given the possibilities that are available today.
      btw just how far can 40MHz penetrate water? I imagine it’s not that much…

  3. “All of this is predicated on a plane-centric configuration, but what if you have some sort of project that doesn’t resemble that sort of thing? Forget quadcopters and fixed-wing planes. You should be able to control anything with a RC controller.”

    Meh!!!

    Some years ago I picked up a big bulky RC plane controller and a receiver that may or may not actually be matched to it at a garage sale or maybe it was a hamfest. I don’t remember. It was cheap and I remember these things being expensive. I wasn’t planning on any RC aircraft projects in the immediate future but I thought the same… I should be able to control anything I want to with it. I thought it was a great find!

    Then… somebody explained to me that the frequencies used by it were actually assigned by the FCC to be used for RC aircraft. RC other things are supposed to use a different band. Really! I just thought it would all be free and clear part 15, no strings attached!

    Of course I could have just used it anyway. What are the odds I would end up interfering with anyone or that anyone would even notice or care? Still, that pretty well killed my enthusiasm and down to the bottom of the junk pile it went where it still is today awaiting the day I have an airborne project for it or maybe just have a sale of my own.

    1. The old AM RC systems are banned from use. Dunno about analog FM. Both of those had separate frequency ranges for aircraft and land/water vehicles. PCM supplanted both, followed by digital PCM obsoleting that.

      DPCM allows multiple remote controlled vehicles to be in proximity without the transmitters interfering with one another. Gone are the days of only being able to fly one plane per channel close together, no more causing another plane to go out of control simply by turning on another transmitter on the same channel. Don’t have to check for other people active on your channel and get clearance to do a ground test.

      Unique (hopefully) identifiers and encoded signals so the intended recipient only pays attention to commands from the controller matched to it have done away with that stuff.

    2. For a “long time”, you needed a ham license in the US to do radio control, since there no frequencies assigned to RC. That eventually changed, a no-test license for radio control and some frequencies for it, at 27MHz (CB) and 72MHz.

      Toy radio control came later, making use of 100mW unlicensed allocations at 27MHz and probably 49MHz. But they are low power, and probably generally cheap, and not frequency flexible, so a hazard of others are using them.

      Circa 1976 a friend had a license free RC boat, used up lot of batteries. Weld take it to the park, where there was a decent sized “lake”. In t middle, contact was lost, so we had to wait for the wind to blow it closer to shore, which put it back in range, though sometimes that meant going around to the other side.

      Michael

  4. You are so right! I have a Spektrum DX8 which was advertised with “free adressable switches”. You can change some functions in a sub-sub menu, but this far away from free adressable.
    As you mentioned there a many open sorce projects for different applications. OpenTX, Cleanflight, Betaflight, BLHeli….
    Maybe a little bit off-topic, but what I still don’t understand why the developers/ community sometimes invent new protocols: Why DShot (protocol to talk to ESCs) and its successors and not SPI, CAN or another industry standard?

    1. People have been asking for CAN forever, but it’s never become popular. The standard answers are:

      CAN – requires extra hardware on both FC and ESC, concerns over limited bandwith (hype is all about 32khz update rate, regardless of how pointless that may be, requiring something like 8mbit bandwith, right near the upper limit of CAN FD)

      SPI – requires more pins on the MCU. unique CS per endpoint plus shared clock and shared data. DShot (and all it’s analog predecessors) require only a single pin per endpoint.

    2. CAN is fairly complicated and expensive to implement in such cheap devices, keep in mind that DShot was designed with multirotors in mind, where you typically need 4 EScs, not 1 – so cost is important. Also size, dedicated comm chips increase the physical size, which is also important in multirotors.

      The main reason DShot is made how it’s made is the ability to use it with existing HW with minimal to no modifications, just a reflash of the MCU.

  5. I’ve got er9x installed on a FS-TH9x, and I’ve been quite happy with it. Certainly enough capability for my rather basic usage, anyway. Considering the TX hardware I was used to in RC vehicles (old 72 MHz Futaba and Traxxas gear) it is definitely a huge improvement.

    Can’t say I’m using it enough to make an upgrade worth it, but I have to admit the new generation of TXs that look more like gaming controllers than flat slabs are certainly interesting.

    1. Though 1963 was all 2013 before 2013 was…. because you could have the 21st century Flappy Bird control method right then on your model aircraft…

      Seriously, you just hammered that button…. the advanced models would let you get up elevator every 3rd push and hold.

      1. Maybe of little interest: After thewar 46-47, a freind of mine ordered surplus micro radio tubes from the us to build his own rx. The plane only had one channel on the rudder so slow highwing model Tx as a box with a button. Pushing it made the reciever toggle a relay in the plane that made a rubber band wound clock mechanism advance one tooth. Control wa neutral-left-neutral-right for each button push. Talk about ingenuety back then.

        1. Yes, those escapements were something, sequential actions. They’d make a good retrotechtacular article, give people ideas about how to rig something similar in principle when they’ve only got a 1 bit output left on their uP or something and 4 things they want to actuate, though electronic decoding possible. But cams could be 3D printed easily these days. Maybe it would be suitable for such situations where mechanical actions such as shifting gears were needed though.

  6. There is no such thing as a ‘Taranis 9X’. You’re mixing/confusing the Turnigy 9x and the Taranis X9D, the latter being a much better and more powerful radio. The X9D is the old version, the newer and more popular version is the X9D+.

  7. “We’re facing an age where the people who invest most heavily in RC equipment are also the ones most likely to want, and know how to work with customization for their rapidly evolving gear.”

    Is it April fools already? Surely others remember actually making their own electronics? I don’t think you get more customized than making it yourself from scratch and surely you can’t argue there is more of that than earlier in the history of the hobby.

    1. When I was about 12ish the realisation of how expensive RC was back then, had sunk in, just a 4 channel sport model fully kitted out, would probably have set you back the equivalent of $2000 today and the RC gear would have been about a third of that. And that would be with building it, ARTF or RTF was $$$$ for the very few that were around then.

      Anyway, I’d decided I’d better start from basics and build myself gear, learning electronics as I went. In terms of what I could get hold of I was still something like a decade and a half behind state of the art. However with transport of models being an issue also, I’d settled on small models, park flyers before park flyers were a thing, and attempting to build subminiature gear. Inspired in this by Howard G McEntee’s articles, for example this kind of thing, though this one in particular is a bit of a stubby winged hotrod… https://aerofred.com/details.php?image_id=94502&mode=search&

      So, I found a likely receiver design and using more modern smaller components, managed to cram it onto a square inch of PCB. This went into a 15″ span slow and floaty biplane design adapted from a free flight plan. Used an “antique” even by that era escapement for an actuator, made a flight battery out of tiny salvaged nicads. Got it all together with a Cox .020 for oomph. Now I’d had to do my range testing and tuning by myself which was a bit difficult, so probably could have been better tuned and didn’t really have great interference rejection, I’d see it glitch every few mins. Did my glide testing, then trimmed it for huge circles just in case, gassed up and flew! Yay, it flew…

      However, at this point control authority, rudder only, was kind of slight, I’d set it up without much throw, didn’t know how twitchy it would be, so could modify the flight pattern to full left rudder = wider huge circles and full right = tighter huge circles… Okay, so next trip out after recharging batts, I’d set rudder control to full throw, this time I could make it turn lazily left and a tighter turn to the right, which if I held long enough did seem to sideslip it a bit and lose a little height… woohoo, I could make it go in whatever approximate 50ft square patch of sky I wanted now. Had several more flights with tabs stuck on the rudder trying to get more control out of it, but I guess it was just a hella stable FF design. I think it was fuselage side area, tall and flat, that was working against me, plus some generous dihedral. Kinda got shelved and started figuring out my next one…

      1. Not so sure that is entriely true.
        You just have to find the PPM stream and re route it to a new transmitter module like the FRSKY one which is tiny and shodl fit inside most old radios…

  8. Only useful and trustworthy resource for selecting R/C equipment is the horse’s mouth. Go out to the flying field or track and find out which units are popular with entry level folks, what drool stained ads the experienced enthusiast has folded up in his back pocket, and which the top dog$ own for their one-upmanship bragging antics. Then take a peek in your wallet and double-check your info with the clerk at the hobby shop. What? You mailorder? Wow. You won’t be flying much. Need stupid $2 part you broke yesterday in order to be able to fly today. Keep your Local Hobby Shop in Business! You need them to avoid the 7x times a summer critical part need that mailorder will help you to lose half a summer!

    If at the field you saw XYZ brand was in the hands of half the enthusiasts then buy that… every one of those users has the spare parts on hand you are likely to break and need right there at the field. You went out to the field to fly, not pack up and head home to place a mailorder for a $2 part which will arrive in 10 days.

  9. Have a SUPERB Futaba full digital proprtional 5 channel RC system I bought back in the 80s…. Can’t use it! Frequency since outlawed. Not retro-fixable. Oh I could get away using it around my rural property. Not much out here to interfere with. (Not even certain my closest neighbor has Intetnet). But officially outlawed. Never felt inclined to buy another.

  10. This discussion clearly presupposes a particular kind of operator, one with few manipulative skills. If you want your plane/car/robot to operate itself and you merely tell it where to go, such as when flying a drone, then yes, the traditional style r/c set is really not suitable. On the other hand, if you are looking to precision control requiring high levels of skill and years of practice then modern high-end r/c sets with ball raced gimbals, 16 bit hall effect rotary position sensors, pc interfaces, etc. exist, and they exist for a good reason – the configuration works and though updated and refined to meet new technology expectations, it is still basically the original good idea. I fly precision aerobatics and helicopter competitively and this could not be done without modern well designed radios. Sure, cheap Asian radios have lots of features but are simply not reliable enough to manage a model costing in excess of $10,000, weighing over 20kg, and traveling at 200kph. This is a dangerous device in untrained hands. High reliability electronics, triple redundancy, failsafe systems, and quad diversity radios are mandatory in my aircraft – along with good solid German engineering. Ask the cheap radio manufacturers and the hackers what software reliability standard they comply with in their firmware. What reliability certified compiler they are using. How about an audit trail on their components. So when someone describes equipment as clunky and old fashioned they need to be very precise about just what they are talking about and what its application is.

    1. Well put. Too much “toy” attitude in beginning enthusiasts nowdays, and a lack of realization that even with typical hobbyist aircraft you’ve got a 3-4 lb model plane flying typically at 90 mph, cause they won’t think to cut the throttle, and electric or glow engine on the front makes an effective armor piercing warhead. Cars get hit at flying fields from time to time, punching a hole in the metal body. People don’t get hit too often cause they can dodge. But you’ll still hear them proclaim “But it’s just basla and foam! And they should just get out of the way!” Nor can they handle control reversal for coming and going and are too used to quads that really are flying themselves on full autopilot and the owner is merely nothing more than the navigator. But hackers are the most fun to get out to the field and teach… they do learn… and make excellent enthusiasts of good quality. Planting seeds you are… must water and shine apon them.

      Hey! You should look around here! They got one R/C experienced enthusiast that built it big enough he got it in and flew. Almost perfect landing then pulled up into PIO and banked then PIO all the way to the nearest mass of obstacles, PIO still then struck one. Huge open area to fly with just one mass of obstacles…. only beeline he flew. Survived. Had a parachute, altitude of about 40 feet max. rofl… These are hackers. Wing and a prayer, but it’s us on the ground doing the praying.

  11. Relevant in the RC hobbies world, possibly, not in the drones industry and anywhere those same RC vehicles do useful things. Current commercial drone transmitters (grounds stations really) still have the gimbals often, but they’re being dropped. There’s research from a few years back already showing that a cheap tablet with the built-in accelerometers lets newbies control RC stuff much better than a traditional Tx and it’s still minimally better after those newbies have many hours of flying/drivnig experience (don’t have the link)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s