Add CNC To Your…Propane Tank??!?

It’s starting to be that time of year again; the Halloween-themed hacks are rolling in.

[John Lauer] needed a propane-powered flame effect for his backyard ICBM “crash site”. Rather than pony up for an expensive, electronically-controlled propane
valve, he made a custom bracket to connect a stepper motor to the propane burner’s existing valve.

With the stepper motor connected up, a TinyG stepper motor controller and [John’s] own graphical interface, ChiliPeppr, take care of the rest.

The hack is almost certainly a case of “everything looks like a nail when you have a hammer” but you have to admit that it works well and probably didn’t take [John] all that much time to whip up. Maybe everyone should have a couple spare stepper motors with driver circuitry just lying around ready to go? You know, just in case.

All the details of the build are in the video. If you’re done watching the flames, skip to around 2:50 where we see the adapter in action and then [John] steps us through its construction.

You may have seen coverage of the TinyG motor controller here before.

Additional thanks to [Alden Hart] for the tip.

14 thoughts on “Add CNC To Your…Propane Tank??!?

  1. Rather than pony up for an electronically-controlled, properly engineered and reasonably likely to correctly function propane valve, the decision was made to cobble something together. Newsworthy! ?

      1. I kind of agree though, the standoffs are plastic, the screws are open to the environment, zinc plated steel that are fairly coarse and it uses an off the shelf servo that lacks any kind of positional control or even awareness. Further, it has a limited torque range and a complete and utter lack of repeatability testing. All of these things would have been addressed in a commercially produced, properly engineered, mass produced gas valve. And the thing is, they are probably cheaper, in addition to being more reliable. When it comes to something like a gas valve, I personally think erring on the side of “make sure significant effort and testing has gone into ensuring this will work when called upon” is reasonable and appropriate.

        1. Hey guys, fun feedback all around. Yes, this was a quickie Saturday hack, but there’s some things you might be missing.

          1) This needs variable control. You could use an electric solenoid valves, but then you have no pilot light because solenoids are either on or off. So, you end up needing a more complex setup to keep the flame lit. Solenoid valves are around $40, so not awful, but nowhere near as cheap as a $9 stepper. By trying to work around the on/off of a solenoid you’ll spend another $30 on parts to get a constant pilot light. You can find variable gas valves instead but those are $250.

          2) This is a stepper motor. It’s highly repeatable. In the video it shows the Gcode repeating the steps consistently. This NEMA 17 stepper has really good torque. I estimate it has about 10x more torque than it needs to rotate this valve.

          3) This is a one-time disposable project. It just needs to work for about 6 hours and then tossed. It doesn’t need to run for 10 years like a furnace and a $200 Honeywell electric gas switch.

          4) This needed animation. One of the easiest ways to animate on the cheap was to run Gcode. If you do a more complex setup like a solenoid you’re having to write code to try to animate and you still don’t get variable flames. You could try to toggle on/off really quick to mimic variable flames, but you end up spending more time trying to animate than get on to the other projects for the party. You’ll also burn out the solenoid switching it non-stop.

          Of course working with propane is scary stuff, but I can’t see where this valve could fail and cause problems. If anything is going to fail it’s the acrylic pieces shaking themselves apart which will not cause any runaway flames. The valve will either get stuck closed or open. There’s a 15PSI regulator ahead of the valve for safety. If it’s stuck open for a few minutes, I go and turn off the propane tank and the show is over.

          1. You can do variable flow control using solenoid valves. If the response time of the valve is fast enough, you can use PWM (similar to how you control the brightness of an led). If that will wear out the valve too fast, you can use multiple valves connected in parallel, each one being restricted by a different size orifice or partially closed valve. Then you open and close the different solenoid valves to get different flow rates. You don’t have as fine of control over the flow rate, but with three valves, you can create 8 different flow settings which should be enough for this application.

  2. “It’s starting to be that time of year again…” My brain automatically filled in “…time to set trick-or-treaters on fire and keep all the candy for ourselves!” Hehe… In all seriousness though, very nice quick and dirty computer-controlled pyrotechnics.

  3. Nice solution, but I would suggest adding a bit of physical isolation between the tank and your valve so if it goes wrong you can isolate it more safely especially as the valve won’t automatically fail closed.
    I built a similar variable flame a few years back but went for a solenoid valve with a separate pilot light as we were drawing off about 1.5kg of liquid propane a minute and wanted something more robust.

Leave a Reply to QuirkCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.