Wireless Launchpad Part 4: Prototyping

The next stage in this project is to start experimenting with hardware. Here you can see the (almost) complete setup. The yellow switch is the safety interlock, the orange push-switch simulates the igniter. The only components not present yet are the audio alarms for the various state indications.

Breadboard prototyping of Launch Pad hardware

Range Testing

After some rudimentary testing of the Wemos D1 Mini Pro, it seemed like a viable platform on which to build the system. It has enough GPIO pins, has a nice form factor, uses only about 80mA when running (not just idle) and seems stable.

What would be critical to the projects success however, would be the range both it and my mobile phone WiFi hotspot could achieve.

I knocked up a quick mobile version of the hardware, utilising the onboard LED to indicate the state of the WiFi connection.

With the external antenna connected, I managed to achieve a comfortable 100m range between the Wemos and my beat up, old Samsung Galaxy S7 as the hotspot/AP. Realistically the connection was dropping at around 120m, but I didn’t have the means to accurately measure the distance so I’ll stick with the 100m claim – still perfectly good enough for my needs.

Wireless Launchpad Part 3: State Machine

The Pad will be the ultimate controller of state, managing exactly what mode it is in at any one time, and which states it is allowed to transition into.

The Remote will mirror this logic, but the system is not reliant on the Remote to properly maintain and permit state transitions. The Pad will be in constant communication with the Remote to tell it what state it is in. Should the Remote issue an invalid command the Pad will protect against this. Any unexpected command received by the Pad will cause the Pad to disarm itself and revert back to its base Disarmed state.

Functional Overview

The system can be in one of 4 states at any one time;

  • Disarmed
  • Armed
  • Continuity-Test-Passed
  • Firing

These stages must be entered into sequentially, as represented by the below state diagram.

Note that this diagram does not detail the logic governing the permissability of any state transition, it simply describes the set of possible states and transitions between them.

State Transitions and Logic

Arming

The system can only enter the Armed state if the physical Safety Interlock is engaged at the Pad. The system remains in an Armed state for 20s, after which it reverts to a Disarmed state should it not receive a valid Continuity Test command beforehand. When in an Armed state:

  • a physical indicator light at the pad is illuminated (Armed)
  • a warning siren sounds at the pad
  • a virtual indicator is shown on the Remote
  • the Remote is allowed to issue a Disarm and/or Continuity Test command

Performing a Continuity Test

A continuity test of the electrical ignition system (Ignitor) can only be performed when the system is in an Armed state. A continuity test can only be performed from the Remote. For saftey there is no physical continuity test mechanism at the Pad. If the continuity test fails, the system immedately reverts to a Disarmed state. If the continuity test passes, the system transitions to the Continuity-Test-Passed state.

Continuity-Test-Passed

The system remains in an Continuity-Test-Passed state for 20s, after which it reverts to a Disarmed state should it not receive a valid Fire command beforehand.

When in a Continuity-Test-Passed state:

  • a physical indicator light at the pad is illuminated (Ready)
  • a warning siren sounds at the pad
  • a virtual indicator is shown on the Remote
  • the Remote is allowed to issue a Disarm and/or Fire command

Firing

The system can only enter the Firing state from the Continuity-Test-Passed state. The system remains in the Firing state for at most 3s. Whilst in the Firing state, a continuity-test is performed to detect ignition. Which ever occurs first, Igniter continuity-failure or 3s elapsed, triggers the system to disarm itself into the Disarmed state. Whilst in the Firing state:

  • a physical indicator light at the Pad is illuminated (Firing)
  • a warning siren sounds at the Pad
  • a virtual indicator is shown on the Remote
  • the Remote is allowed to issue a Disarm command

Wireless Launchpad Part 2: Safety Codes

Saftey by Design

These are all the mentions of rocketry launch saftey codes with relation to ingition systems that I could find.

NAR Saftey Code Excerpt:

Ignition System. I will launch my rockets with an electrical launch system and electrical motor igniters. My launch system will have a safety interlock in series with the launch switch, and will use a launch switch that returns to the “off” position when released.

NFPA 1127 Code for High Power Rocketry

4.13.1 A high powered rocket shall be launched using an ignition system this is remotely controlled, is electrically operated, and contains a launching switch that returns to the “off” position when released.

4.13.2 The ignition system shall contain a removable saftey interlock device in series with the launch switch.

4.13.3 The launch system and ignitor combination shall be designed, installed and operated so that liftoff of the rocket occurs within 3 seconds of actuation of the launch system.

Tripoli Saftey Code Excerpt

Cites NFPA 1127, also including the below stipulations:

A-5 A rocket motor shall not be ignited by using: a. A switch that uses mercury. b. “Pressure roller” switches

N.B. I’m interested to know what event caused these Additional Rulings.

UKRA Saftey Code

2.3.1 Igniter Rules … Any igniter should ignite the rocket within three seconds of the power being applied to the igniter. Continuity tests on any motor ignition system should not be carried out whilst the igniter is fitted in the motor, unless the continuity test is an integral part of a count down sequence. … When an igniter test is carried out, a clear and audible warning and countdown should be given of the test, and the tester should not carry out the test until all people who are likely to see or hear the test are fully alerted and prepared for the test.

2.5 Launch Controllers An electrical ignition system must be used which allows for remote operation of the igniter firing. … The launch controller must include a safety key to immobilise the system when removed. This key should only be in place at the time of the launch and is to be removed immediately after an ignition attempt, especially in the event of a misfire. The safety key must not be capable of being removed leaving the controller in a live firing mode. The firing circuit must only be live for a brief period sufficient to fire the igniter and must then return to an open circuit. Where a firing button is used, it must return to the off position when released.

4.2 Launching An electrical ignition system must be used which allows for remote operation of the igniter firing. The device should be operated from at least the minimum safe distance as determined by the total impulse of the rocket motor(s) according to the Safe Distance Table given above. This distance can be shortened with the express permission of the RSO as per section 2.5. Any igniter should ignite the rocket within three seconds of the power being applied to the igniter. UKRA currently only recommend the use of hardwired launch systems. However, progress in stable digital control systems over recent years mean that clubs are can to use radio-controlled systems should they wish to do so. UKRA recommends that such systems are thoroughly checked and tested prior to use.

There are two main concerns/rule-interpretations with this wireless, app-driven launch system which are not necessarily trivially solved or expected-to-be-trivially solved by an RSO.

  1. Hackability (deliberate or through electrical interference) of the wireless medium
  2. “Removable” series launch-button interlock/key

(1) Can be easily solved via a sensible choice of communication medium, as discussed in part 1.

(2) Is probably more of a contentious point. My interpretation is that the wording of these safety codes is legacy in its use of “physical” words, but very relevant and pertinent in its implications for safety.

I think there are two points here:

There should be a cermeony (inserting the key) which makes it obvious and unabiguous that you are entering firing mode. Mentally this is an intentional and deliberate action which you are hopefully doing after obeying all of the various other, previous setup saftey steps.

The Firing mechanism should be “removable” or “disableable” so that it is difficult to accidentally trigger a Fire event.

I’m confident that a combination of difficult-to-accidentally-trigger user interface, finger print authentication and rapid inactivity timeouts not only meet the spirit of all of the above saftey criteria, but actually go much further than most personal and club launch systems.

In part 3 I’ll outline the design for the overall system and discuss the safety features in more detail.

Wireless Launchpad Part 1: Communication Medium

Communication medium

Having recently acquired a military tripod which I plan to turn into a rocket launch pad, I wanted to experiment with the possibility of creating a wireless launch controller.

For this project safety is paramount, however I’ll the discussions around this to a later post. For now I wanted to experiment with different communication mediums and components.

I really want to be able to control the launch pad from my Android phone, without any extra hardware. That means the wireless options available to this project are:

GSM

Either communicating by SMS, GSM Serial or TCP/IP; the launch pad would need a paid SIM as well as sufficient signal (as would my phone). Most of the launch sites I fly from have good signal coverage however this is not my primary choice. I deemed this medium reasonably secure, as you’d need to know the number of the pad to hijack it – which is unlikely to be guessed. Additional encryption atop this is probably unnecessary. However, this implementation has many more “moving parts” than the alternatives so for now, it’s sidelined.

Bluetooth

The Bluetooth protocol is very well suited to this application, providing a simple, secure Serial channel over which to communicate. However, I struggled to find any cost effective units which had sufficient range for launching rockets (despite quoted module and protocol ranges). Perhaps they exist, I didn’t look very hard before moving on.

Right: HM-10, Left: BT-09 knock off

I did some experimentation with the HM-10 BLE module which I found to be excellent and can highly recommend. Just beware the myriad knockoffs which looks almost identical. These come with ancient, undocumented firmware which did sort of work, but there’s plenty of people having issues with them across the web and I failed to flash modern firmware to it.

WiFi

This is my current preferred medium and the one I’m pursuing. The pad can be configured to connect to a hotspot of my choosing using WPA2, so its extremely secure. Line of sight in a field should yield at least a hundred metres of connectivity.

Left: ESP-201, Right: WEMOS D1 Mini Pro v1.0

I’ve been experimenting with the ESP-201 and the WEMOS D1 Mini Pro, both of which use the ESP-8266 chipset and feature IPX external antenna connections.

The ESP-201 is a bit more primitive than the D1 and I had trouble getting it set up and flashing it. Documentation is non existent except on a few other experimenters’ blogs, each of which give conflicting instructions as to what magic state to put each pin in for flashing.

So far I love the D1, its plug and play and hasn’t given me any issues. A range test is to follow with a variety of external antennae.

AN-M30 100lb Bomb

General-purpose (GP) bombs use a thick-walled metal casing with explosive filler (typically TNT, Composition B, or Tritonal in NATO or United States service) composing about 30% to 40% of the bomb’s total weight. The British term for a bomb of this type is “medium case” or “medium capacity” (MC). The GP bomb is a common weapon of fighter bomber and attack aircraft because it is useful for a variety of tactical applications and relatively cheap.

Rocket Kit

The standard kit is a 4″ diameter body with a 29mm motor mount tube.
If you want the kit at a different scale or with a different size motor mount please get in touch.
It flies brilliantly on a Cesaroni F56-WT motor with a 3s ejection delay.

Included in the kit are:

  • Assembly Instructions
  • 3D printed ABS nose-cone
  • Cardboard nose-cone shoulder/coupler
  • Cardboard body tube (4″ diameter)
  • 29mm motor mount tube
  • 3x 3mm ply centering rings
  • 4x 3mm ply fins arms
  • 4x 1.5mm ply fins
  • 2x 3D printed ABS Rail guides (for 6mm rail)
  • 3D printed ABS tail piece
  • 3D printed ABS screw-on motor retainer
  • Stencils

Not included in the kit are:

  • Recovery hardware – although we provide recommendations for shock cord length and parachute dimensions
  • Motors
  • Launch hardware
  • Glue
  • Paint

Source Reference Material

Our AN-M30 kit was inspired by https://www.storenvy.com/products/18109943-ww2-an-m30-100lbs-bomb-replica which may give you some further detail on paint schemes.

Malyutka 9M14 ATGM

The 9M14 Malyutka (Russian: Малютка; “Little one”, NATO reporting name: AT-3 Sagger) is a manual command to line of sight (MCLOS) wire-guided anti-tank guided missile (ATGM) system developed in the Soviet Union. It was the first man-portable anti-tank guided missile of the Soviet Union and is probably the most widely produced ATGM of all time—with Soviet production peaking at 25,000 missiles a year during the 1960s and 1970s. In addition, copies of the missile have been manufactured under various names by at least five countries.
Since supplemented by more advanced anti-tank guided missiles, the Sagger and its variants have seen widespread use in nearly every regional conflict since the 1960s.

Rocket Kit

The standard kit is a 3″/76mm diameter body with a 29mm motor mount tube.
If you want the kit at a different scale or with a different size motor mount please get in touch.
It flies brilliantly on a Cesaroni F56-WT motor with a 3s ejection delay.

Included in the kit are:

  • Assembly Instructions
  • 3D printed ABS nose-cone
  • Cardboard nose-cone shoulder/coupler
  • 3mm ply nose-cone bulkplate
  • Cardboard body tube (3″ diameter)
  • 29mm motor mount tube
  • 3x 3mm ply centering rings
  • 4x 3mm ply fins
  • 3x 3D printed ABS decorative thrusters (rail guide takes the place of the 4th)
  • 2x 3D printed ABS Rail guides (for 6mm rail)
  • 3D printed ABS tail piece
  • 3D printed ABS screw-on motor retainer
  • Stencils

Not included in the kit are:

  • Recovery hardware – although we provide recommendations for shock cord length and parachute dimensions
  • Motors
  • Launch hardware
  • Glue
  • Paint

Source Reference Material

Our Malyutka kit was designed from the below reference material. We’ve reproduced the details as accurately as possible whilst maintaining a stable and lightweight model rocket airframe. You may wish to use the below for alternative paint schemes

Pershing 1A

The MGM-31A Pershing was the missile used in the Pershing 1 and Pershing 1A field artillery missile systems. It was a solid-fueled two-stage ballistic missile designed and built by Martin Marietta to replace the PGM-11 Redstone missile as the primary nuclear-capable theater-level weapon of the United States Army and replaced the MGM-1 Matador cruise missiles operated by the German Air Force. Pershing later replaced the European-based MGM-13 Mace cruise missiles deployed by the United States Air Force and the German Air Force. Development began in 1958, with the first test missile fired in 1960, the Pershing 1 system deployed in 1963 and the improved Pershing 1A deployed in 1969.

Bullpup 12D

The Bullpup was the first mass-produced air-surface command guided missile, first deployed by the United States Navy in 1959 as the ASM-N-7, until it was redesignated the AGM-12B in 1962. It was developed as a result of experiences in the Korean War where US airpower had great difficulty in destroying targets which required precise aiming and were often heavily defended, such as bridges.
The Bullpup had a Manual Command Line Of Sight guidance system with roll-stabilization. In flight the pilot or weapons operator tracked the Bullpup by watching a flare on the back of the missile and used a control joystick to steer it toward the target using radio signals. It was initially powered by a solid fuel rocket motor, and carried a 250 lb (110 kg) warhead.

Included in this kit are:

  • Assembly Instructions
  • 3D printed ABS nose-cone
  • Cardboard nose-cone shoulder/coupler
  • Cardboard body tube (3″ diameter)
  • 29mm motor mount tube
  • 2x 3mm ply centering rings
  • 4x 3mm ply fins
  • 2x 3D printed ABS Rail guides (for 6mm rail)
  • 3D printed ABS tail piece
  • 3D printed ABS screw-on motor retainer
  • Stencils

Not included in the kit are:

  • Recovery hardware – although we provide recommendations for shock cord length and parachute dimensions
  • Motors
  • Launch hardware
  • Glue
  • Paint

Accessible Nose Cone Bulkplate

In designing a couple of rockets recently I wanted to be able to fly them on different motors (read – different weight motors).

Due to the short nature of the rockets the CG would be significantly affected by this and so the nose weight would need adjusting accordingly.

With a bulkplate fitted to the nose however, accessing it to adjust the weight would be impossible.

Introducing the screw-capped nose cone bulkplate:

The threads are 2mm pitch printed at 0.15mm layer height.

The two holes in the bulkplate are for a loop of kevlar to attach the nose to the shock cord. The slot in the cap makes screwing and unscrewing the cap slightly easier as it can get tight (don’t strip the threads!)

You may need to make the female part 1% larger in the x and y axes depending on your printer tolerances.