A key part of this system is the simulated hose nozzle with a built in wireless transmitter. This simulated nozzle should have a look and feel similar to an actual firefighting nozzle, but block the flow of water in the hose. Remember that the goal of this system is to allow firefighters to take a fully charged (pressurized) hose line into a building but not allow them to flow water and damage the contents of the building. When they open the "bail" on this simulated nozzle, a valve will open to allow water to flow outside the building, thus exercising the pump and pump operator.
This proof of concept simulated nozzle is built mostly of 1.5" PVC pipe and fittings, along with a few other mechanical components. The electronics inside consist of a Wixel transceiver board, a cheap flashlight, a couple magnetic reed switches, and a holder for 2 AAA batteries. After assembly I spray painted it black just to make it look a little more like an actual nozzle. The picture below shows what parts were used for the exterior (the picture was taken after it took some abuse in a training exercise):
Most of these parts can be found at any local hardware or home improvement store, with the exception of the 1.5" NPT to NH threaded adapter. That piece adapts from tapered pipe thread (national pipe thread, or NPT) to fire thread (national hose thread, or NH) so that the nozzle can be attached to the end of a fire hose. NH threads use a gasket for sealing as opposed to pipe threads which must be sealed with pipe dope or teflon tape. This adapter can be purchased at any local firefighting supply company.
While several of the joints in this simulated nozzle are glued together, several others are left dry and fastened with drilled holes and bolts. This allows the nozzle to be partially disassembled for easy access to the interior parts. The pictures below show how the nozzle is taken apart to gain access to the battery holder and the other parts mounted inside. The nozzle "bail" handle is made out of a piece of scrap metal that I bent in a vise to the shape I wanted and then drilled to accept a carriage bolt. One of the round drill holes was filed square to accept the carriage bolt head, which forces the bolt to turn as the bail handle is moved.
The heart of this simulated nozzle is a Wixel programmable USB wireless module with an attached magnetic reed switch from sparkfun. The Wixel code is set up to simply transmit the status of the reed switch to a companion Wixel. The reed switch is activated by a magnet that is attached to the carriage bolt holding the scrap metal "bail" or activation handle of the nozzle. Here's a picture of the Wixel with a magnetic reed switch connected from one of the GPIO pins to ground:
The Wixel is simply mounted to the inside of the nozzle body using double sided tape in the vicinity of the magnet on the carriage bolt. When the bail is actuated, the magnet rotates with the bolt inside the nozzle body. By fine tuning the location of the magnet and the Wixel, it's easy to get the reed switch to activate when the nozzle is "open" and deactivate when it is "closed". Power is supplied to the Wixel by two AAA batteries in a holder that's mounted in the "pistol grip" of the nozzle (the PVC sanitary tee). A slide switch has been superglued to the AAA battery holder to allow power to be turned on and off without having to remove the batteries. This battery holder and switch are covered by a PVC cap when in use so that the switch can't be accidentally deactivated as the nozzle gets drug, banged, and beaten around by ambitious firefighters.
What about that flashlight I mentioned earlier? The intent is that when the simulated nozzle is "opened", it would produce a beam of light to simulate water spray, and also produce a spraying water sound. This gives students and instructors visual and audible cues that water is "being sprayed" even though it really isn't. In a long-term "elegant" solution, these visual and audible cues would be controlled by the same microcontroller that is transmitting the wireless data. In this "quick and dirty" proof of concept though, I did something much easier and "hackish" just to demo the idea. I took a cheap LED flashlight (the kind you find in the checkout aisle for $1.99) and connected another one of those magnetic reed switches across the switch contacts in the tail cap. I then stuffed the flashlight into the front of the nozzle using some water pipe insulation as a spacer. After adjusting the flashlight by moving it in and out away from the magnet on the carriage bolt, and by rotating it, I could get the light to turn on when the nozzle is "opened" and turn off when it is "closed". Like I said, kind of hackish, but it works as a demonstration.
So, how does it work? Quite well, actually! Transmitter range is similar to other low power 2.4 GHz devices. I would eventually like to migrate from using the Wixel to using a WiFi-based solution like the TI CC3000, but I had some Wixels on hand and they are easy to work with.
I'll post the code for the Wixels (along with all of my other code and design documentation) when I can get things organized a little better. If you have questions or suggestions for improvement, feel free to leave a comment!