16 Rocket Semi-Simultaneous Launch Controller

[LydaRA]

So about to buy parts for a Cub Scout model rocket launch control system! Yay!!!

It is made of multiple boxes, each with a big 12V battery and an Arduino, networked together. At the launch pads, each Pad Box needs to power both the static 5V control logic AND 12V through relays to between one and eight model rocket igniters. Although the RSO has the Master Arm, _each_ Scout has their own Launch Button.

Any thoughts on what DC : DC power supply I might use? For the Pad Boxes, the big 12V motorcycle battery needs to supply:
1) 5V to Arduino
2) 5V to relay drivers
3) 12V to quantity 1-8 x 1-2Amp each igniters

But that means the Amps could swing a good bit:
1) If firing all 8 rocket igniters at once, then large cumulative load. Got to make sure still enough not to starve the Arduino.
2) If only firing one rocket (a lone scout goes early or late), too much current to one channel? Got to make sure not to weld relay or melt other components.

Thanks in advance!

Robert Lyda, Sr.
ASM, Troop 148
COPE & Climb Level II
WAC Rockets Lead
NRA RSO

 

[mikec]

I personally think that launching multiple rockets at once is not worth the safety compromises, but as long as you stay within the constraints of the safety code: "When conducting a simultaneous launch of more than ten rockets I will observe a safe distance of 1.5 times the maximum expected altitude of any launched rocket."

Having multiple launch buttons like this, while it could be done safely since you have a central lockout, seems like an accident waiting to happen.

That said, you are going to have to use a voltage regulator of some sort to power the Arduino from 12V, probably just the regulator already on the board. I think it's unlikely that the 12V voltage will slump enough to brown out the Arduino given the capacitance on the board.

 

[mikec]

It'll probably be trickier to arrange the relay drivers. You might consider having a single arming relay that would protect against relay welding, though welding is rare in the relay systems I've maintained. I see people using relay systems for very short cable runs when they aren't really necessary and introduce a lot of complexity. A simple arrangement of switches and a central battery would be sufficient if you launched fewer than 10 rockets and perhaps removed the requirement to have multiple launch buttons.

 

[sghioto]

If only firing one rocket (a lone scout goes early or late), too much current to one channel? Got to make sure not to weld relay or melt other components.

That won't happen. An igniter is only going to draw the current it's resistance demands regardless if the battery is rated 10 or 100 amp/hr.

Although the RSO has the Master Arm, _each_ Scout has their own Launch Button.

Does this mean there are as many launch buttons as pads?
I would be interested in seeing a schematic of this launch system if possible.

 

[LydaRA]

This is technically an "Engineering" station at Webelos Adventure Camp, for 4th & 5th graders. The requirements are to learn about a type of engineering, then to follow blueprints to build something. We chose rocketry, as who doesn't get excited by rockets? But that's only really fun if they each get to _FLY_ what they build! And since these are 4th & 5th graders, well they aren't the quickest builders....which then pushes our time to get flights in before they have to move on to another station. And since this is educational: we fly once, stand where they land as a "living scatterplot," then make aiming adjustments and fly again to see the results....which all takes even more time.

So we "drag race" by counting down together and then all push our buttons! Ten individual stock Estes launch controllers in the hands of ten children... I have to muster a LOT of Range Safety Officer discipline to keep everyone safe. Although we need to increase our throughput by going to 16 Scouts launching at once, there is no way I'm taking that on without adding some additional system controls.

The big difference between systems I have seen elsewhere versus our needs: it seems almost all of the other multiple-pad controllers I've seen--one-person launches for all. We want to continue to allow each Cub Scout (after the RSO arms) to control their own individual launch buttons. Most Scouts comply with the shared countdown & "drag race." However there are often Scouts who want to "confirm their own control" or "call attention to themselves" by launching early or late! A little early or late, no problem....these "control freaks" may become real rocketeers someday! (But we do want the RSO's master-arm to keep it within bounds).

So all of the individual launch buttons are merely gathered as inputs to the RSO's Arduino control. Only when the Pad Boxes have their local power turned on do the Arduinos (or anything else) come on. Only when this RSO pushes the Master Arm do the Pad Boxes' Arduinos get commanded to power on the relay boards. Similarly, only when the RSO's countdown clock enables the two second window, then do the collection of launch button signals get forwarded to the Pad Boxes. So remove any of FIVE keys or momentary buttons, or the control logic, then NOTHING happens. Add in extra distance, buzzers, strobe lights, "Apollo-style" countdown/message display, Scouts launching from under a tent, "peanut gallery" watching/waiting under another tent... Well you see we take safety _very_seriously_ indeed!

 

[LydaRA]

A simplistic overview of the Pad Boxes... Suggestions for battery and regulators? TiA.

 

[sghioto]

Very interesting the thought behind all this.
I would use a simple 7805 linear regulator for the Arduino 5 volt supply and drive 12 volt relays using transistors or FETs.
12 volt motorcycle battery should suffice.

 

[LydaRA]

Very interesting the thought behind all this.
I would use a simple 7805 linear regulator for the Arduino 5 volt supply and drive 12 volt relays using transistors or FETs.
12 volt motorcycle battery should suffice.
View attachment 461619

Yes, how I'd scratch-build it. Except zero connection between each channel. Each transister/relay/igniter set is driven independently by a separate Arduino GPIO pin. This is not a "RSO-launch-all" but an "empower each child to launch their own" system.

 

[LydaRA]

I personally think that launching multiple rockets at once is not worth the safety compromises, but as long as you stay within the constraints of the safety code: "When conducting a simultaneous launch of more than ten rockets I will observe a safe distance of 1.5 times the maximum expected altitude of any launched rocket."

Having multiple launch buttons like this, while it could be done safely since you have a central lockout, seems like an accident waiting to happen.

That said, you are going to have to use a voltage regulator of some sort to power the Arduino from 12V, probably just the regulator already on the board. I think it's unlikely that the 12V voltage will slump enough to brown out the Arduino given the capacitance on the board.

And thus another feature of this modular system: Pad Boxes can be commanded individually, even if scaled to a world-record number (like the NASA launch).

Like many NAR club events, launch pads can be clustered in different parts of the field....and wired up all at once (overlapping all that effort and time). Yet the RSO can then lead just "LP Bank A" to countdown & launch, then "LP Bank B," then "LP Bank C," then "MP Bank D," then "HP Bank E," etc...

Only instead of the RSO/LCO doing the unified single Launch button push, my system would have the RSO/LCO arm a bank of individual Launch buttons for the fliers to press for themselves.

 

[RocketRev]

Hello Robert,

OK, it looks like you are actually thinking of building a 2-bank launch system with 8-pads in each bank.

The simplest and easiest method would be for you to wire eight Estes Launchers, each with its own plug, to the same power source with one keyed heavy duty switch in your control while the ten scouts each hold their sub-controller. This method would cost the least, would require little in the way of maintenance over time, would allow for simple replacement of defective Estes launchers, still maintains the individual continuity check, as well as the individual ignition switch, while adding an adult overall removable key power control over the whole system.

The NASA style timer is an excellent addition to which you might add video in order to "time" which rocket(s) get closest to "lift-off" at ZERO, adding an incentive for all to launch as close to the same time as possible. And your keyed relay power simultaneous to all the Estes controllers would give you control to keep anybody from trying to launch too early or too late.

Sure the Arduino path might work, (though you have several hurdles that you only just suspect exist at this point), but the system you will end up with will unnecessarily complicate your launch system without giving you any assurance of long term usability. Its only eight single engine model rockets at the same time so why all the extreme effort to reinvent the wheel and complicate the process? I built a 10-pad launch system with both individual continuity checks and 10-pad drag race capability with one key switch to power it all, around 1967. Its not that complicated.

Please, before you go the complicated Arduino path, you need to add up what its going to cost you to build that Arduino launch system. And be prepared to have it cost a lot more than your first design. Things like this always cost more than you think they will when you're in the design phase. Half-again what you think it will cost is a good minimum place to start.

Then figure out what it would cost to build a central powered key switched controller with those 8 plug-ins for Estes controllers and compare that price to the Arduino system price.

Then think about what's already available commercially. There are any number of launch system manufacturers who would be glad to build you a system to fill your specific needs. And there are probably off the shelf systems ready to do close to the same thing. Find out what they cost and compare. You have nothing to lose but a bit of your time and possibly lots to gain by checking out all the alternatives. I know that there is at least one Scouting Council (Chicago Area - Pathways?) who bought a commercial 32-pads in four banks, launch system that they use every year as a recruiting tool. It would also work very well at a camp. It might even be available for a loan. Though I have no idea where you're at. Contact them and check it out.

And whatever way you decide to go, remember, its going to be getting used by scouts who in my personal experience are pretty hard on equipment. Whatever you get is going to need to be pretty much bullet proof or it will not last. Whimpey equipment will get destroyed in next to no time.

Brad, the "Rocket Rev.," Wilson

 

[mjennings]

Not to derail the camp plans but isn't requirement 2 on Engineer for the scout to create the blueprints and then build the project? There are Rocket related requirements in Adventures in Science and the Townes Super Nova award too.
It's a really neat idea. I'm sure I follow the need for the Andrinos, I think it could be done more simply.

 

[LydaRA]

Not to derail the camp plans but isn't requirement 2 on Engineer for the scout to create the blueprints and then build the project? There are Rocket related requirements in Adventures in Science and the Townes Super Nova award too.
It's a really neat idea. I'm sure I follow the need for the Andrinos, I think it could be done more simply.

The Arduinos would give us the ability to _embed_ more control of the process. With a small display on the RSO Box, we could even give speaking prompts (like PowerPoint speaker's notes). Looking towards that inevitable day....sometime I'll have to train my replacement. And if we lend it out to a district or unit....well, we can enforce some of the same range discipline.

Going "smart" electronics also means that we can increase the distance between boxes and people--without having to worry about wire resistance. It would also speed setup and teardown--and keep down damage from children trying to walk through the "snake pit" of individual cables from the stock Estes controllers. Using common STP CatX or DMX connections would also mean whatever length of control cables can be sourced as cheap commodities at Walmart, BestBuy, wherever.

I did consider "simpler" concepts, where the individual launch buttons and continuity lights wired directly and independently to the pads, using the RSO Box only as a relay interrupt on the master power supply. But that seemed to either need soooo many cables, or to require a Centronics or other less-than-common (today, impossible-to-find tomorrow) high-pair-count connector.Cables would then likely have to be custom made for each desired length.

Still haven't worked out the Arduino to Arduino synchronization over these distances yet. And yet it seems "simpler" to maintain long term. Most electronics that don't fail in the first 100 hours run "forever." Cables drug around by volunteers in the field....well, they are just trying to help...

 

[mjennings]

Fair points. I do hope you have plenty of testing time planned. I ran a camp launch for Apollo 11's 50th. Wasn't time to build so I pre assembled the scout themed Qubits off of Art Apple wrights site and gave a little rocket talk. Great fun was had by all. Ran 3 pads with Estes controllers and one safety key.

 

[LydaRA]

Another part I'm conflicted with: the Launch Panels. At each eight foot table under the edge of the tent, we have four Scouts kneel shoulder-to-sholder. There is to be a Continuity light and a Launch button in front of each of them, just like the individual stock controllers.

But to keep down the number of cables, to speed setup & teardown, thinking of joining these to a single "box" or "panel" per table. A single STP CatX jack allows connection back to the RSO Box.

I've seen an example of an open plexiglass panel with lights and buttons drilled through. Seems weak and floppy. Better not have any water on the table to short the exposed components and inadvertently fire! How to transport and store without bashing the components out?

So I'm WIDE OPEN to suggestions... But so far thinking a 6-8' stick of metal wire raceway moulding. Drill it, mount components, lay wires in place, spray foam, and put the pack on. Foam will help stiffen even more, keep wires from moving around, and help keep moisture out. Would have to be picked away at to make repairs though. For transport and storage, unplug the RSO link and slide the stick into a piece of PVC pipe as a "case." Simple, but then again more materials.

 

[LydaRA]

Fair points. I do hope you have plenty of testing time planned. I ran a camp launch for Apollo 11's 50th. Wasn't time to build so I pre assembled the scout themed Qubits off of Art Apple wrights site and gave a little rocket talk. Great fun was had by all. Ran 3 pads with Estes controllers and one safety key.

My plan is to build in May, test in June, then run in July. I doubt either my wife or my HOA will let me test in my back yard for more than a month. ;-p Then again, with asking for so much input before, and with unit testing as I build....better not get to the end with many problems. (Otherwise _I_ am the unsafe element, and should outsorce this....or just use the stock Estes controllers.)

Building electronics is not exactly my normal day job of networking, phone systems, and general IT though...

 

[sghioto]

There is to be a Continuity light and a Launch button in front of each of them,

As far as the logistics of the wiring how many conductor cable is going to be required from the 8 pad range box to the RSO box?
How many conductors from each 4 button launch panel per table?

 

[mikec]

My plan is to build in May, test in June, then run in July. I doubt niether my wife nor my HOA will let me test in my back yard for more than a month. ;-p

Obviously you should be testing for a good long while with 12V light bulbs or something, you can do that anywhere.

I build high-rel electronics for a living, and I've built and maintained launch systems for the local club. I'm probably being a conservative buzzkill here, but there is *no way* I would build something this complicated without both a whole lot of testing and a really good reason. If you're having fun with it, that's great, but if you have a problem the day of, I can pretty much guarantee that won't be fun.

 

[LydaRA]

The reason for the STP CatX or DMX from RSO Box to Launch Panels: shield for power + 4 x (1 wire for light + 1 wire for button). Four independent cables from RSO Box to four Launch Panels. Lets us scale up (to 16 rockets) or down (to 4 rockets) as needed....drag out only what is needed for the day. Also lets us ignore or replace a segment if there is some problem.

Similar reason for multiple Pad Boxes, scale and durability/maintenance.

Not quite sure yet what the Arduino synchronization bus will require. I'd like the simplicity of the built-in I2C or Serial. But distance may require more... Once I nail down the Arduino synchronization bus, thenI'll know the cabling required between those. Any suggestions???

(May try to squeeze in a visit to the local university's EE department (which teaches some Arduino labs) to ask their opinions....before they all bog down in end-of-year exams...)

 

[LydaRA]

Obviously you should be testing for a good long while with 12V light bulbs or something, you can do that anywhere.

I build high-rel electronics for a living, and I've built and maintained launch systems for the local club. I'm probably being a conservative buzzkill here, but there is *no way* I would build something this complicated without both a whole lot of testing and a really good reason. If you're having fun with it, that's great, but if you have a problem the day of, I can pretty much guarantee that won't be fun.

YES! Indeed, lots of simulated load....like in amateur radio. As a HAM, I was always impressed at that anecdote of the 100W bulb used as a dummy load--that inadvertently still carried a radio call half-way across the Pacific Ocean!

I have extra bulbs. Also have an extra box of Estes igniters. That way a lot of final testing can be done with the actual targetted igniters.

But of course the true test (and FUN) will be to FLY a few actual completed rockets!!! :-D

(And we have a full week of camp staff "shakedown" to setup, test, and train in situ. For fallback, I'm servicing all of the individual stock Estes launch controllers.)

 

[sghioto]

For testing we used a #1157 bulb, but a #1156 should be just about right with a 2 amp current draw.

 

[LydaRA]

For testing we used a #1157 bulb, but a #1156 should be just about right with a 2 amp current draw.
View attachment 461739

I LOVE that! A simple-to-reuse test load!!!

 

[sghioto]

I'd like the simplicity of the built-in I2C or Serial.

I was thinking that myself but was also wondering if the data would be corrupted if 8 scouts are pressing buttons at the same time.

 

[LydaRA]

I was thinking that myself but was also wondering if the data would be corrupted if 8 scouts are pressing buttons at the same time.

Actually, since it is just on/off, we can pack eight rockets' firing command statii in just a single byte. As the RSO Box's Arduino completes its' simplistic program loop, it merely polls the eight pins/buttons, packs the byte, then transmits it to the Pad Box's Arduino. As the program loops back around, it packs and sends whichever pins/buttons have been pressed or released at that moment. Make it a double-byte to pack in the destination Pad Box ID and a checksum for the firing byte.

Similarly, the Pad Box simply polls the Continuity status of the eight igniters, packs it into a byte, adds it's own ID and checksum to a second byte, then sends it to the RSO Box.

So _trivial_ amounts of data! A fully-envisioned 16 rockets only needs to sync two bytes in each direction per second. Seems _very_ similar to the two-byte Hex coding used between PC and microcontroller for many SOHO automation systems. Slowest I ever operated a network at was Token Ring at 4Mbps. Could even go cheap wireless (have seen other rocket launch controllers with such)--but I've no interest in opening _that_ can of worms (even though I have four FCC licenses).

 

[LydaRA]

So here is an updated (partial) schematic... I'm still learning KiCAD. Thoughts?

 

[sghioto]

On the Igniter_Sub-circuits I see the the igniter getting power through a buzzer, Is that correct?
Is there a separate buzzer for each pad?

 

[LydaRA]

Yes. The buzzer is an "oh ****, stand back" warning that the igniter has power before anyone walks up to the launch pads. Not really helpful during the expected launches. But if the relay contacts weld (or any other failure) that leaves the igniter cable energized, then you know not to even try wiring up the next rocket! No one wants to launch from their hand!

I want it to be as passive and foolproof as possible. But yes, I've thought about OR'ing all of the channels together with diodes or logic gates. Not sure. Thoughts? Examples? THANKS for continuing to bear with me.

 

[sghioto]

I understand the warning if say the contacts are welded shut but when the relay activates the current through the igniter is in series with the buzzer. Will that pass enough current to fire the igniter

 

[LydaRA]

I understand the warning if say the contacts are welded shut but when the relay activates the current through the igniter is in series with the buzzer. Will that pass enough current to fire the igniter

Yeah, thought about making it go in parallel. So only little current in side path (maybe still add massive resistor before the buzzer), with most unobstructed to the igniter. Still want it on the N.O. "Launch" side of the relay, so not in the N.C. continuity test. Haven't seen such a sample though.

 

[LydaRA]

Changed to simpler NPN instead of MOSFET. Moved buzzer.

 

[sghioto]

That won't work the Buzzer is shorted out. I would keep the mosfets.
Here are my suggestions.
Connect the LT3092 to 5 volts not 12 volts as shown. When the igniter is not connected or open there will be 12 volts present at the input pin.
Here I'm showing using only one Buzzer connected through diodes (OR-d) to the drain pin of all the other transistors in the pad box.
If any of the transistors are shorted the buzzer will sound. That would also mean a relay is energized and the red LED for that pad will illuminate. And the same LED will also illuminate if any one of the contacts is shorted.