As my projects continue to grow in size and impulse, it was becoming apparent I needed an alternative to a wire based launch control system. The stand off distance required of large, high impulse rockets is getting beyond the reach of my present launch control system. With my hybrids requiring 3 pyro circuits, a power line for the fill valve and a data line; at a distance of 500' that's 2,500' of wire! While you certainly could use Cat. 5 cable and solenoids, there's still the issue of getting data back. Now, with a "Q" class engine in the works and stand off distances of up to 2,000' required, a wireless system seemed not only practical, but almost a requirement.
Keep in mind I'm no electronics wizard, or a programmer. So my attempts at this are very much a learning experience. Early in the design process I decided to use MaxStream radio modems, they are more expensive, but since I've used them in rocket and balloon telemetry, I was familiar with their use. They also feature a three layer protection system against interference from other radio signals. An inadvertent command to the launch control system could be catastrophic, so the MaxStream's security levels sold me on the system.
Click Here for the BS2 code.
Click Here
for a zip file with the PC application installation software.
The above graphic is more a sketch than a schematic. All the wiring was done point to point with 16 and 22 gauge copper wire. Since most of the circuits carry fairly heavy current and since this would have been a multi-layered board, I opted for a perf board and point to point wiring.
I'll break down the design in sections. First is the power supply. I'm using a Basic Stamp 2 OEM board that has a two built in regulators, one on Vin that can power the board from a DC power supply of 9 to 15 volts. I'm using one of my 12 volt AGM batteries to power the entire launch box. The second BS2 regulator vdd, puts out 5 volts, so I can use this power supply for any circuit requiring regulated 5 volt power.
On the BS2 I'm using pins P0, P1 and P2 to gather pressure transducer data. I'm using a pressure transducer that outputs 0 to 5 volts, this works great since the output can go directly into my 12 bit analog to digital convertor chip, the 3201-C. I'm not using any filtering on the transducer input, and I'm getting very little drift. In some cases you may need filter capacitors to reduce noise. The BS2 can't do floating point math, and I had some problems coming up with a solution to calculating pressure in the BS2. To convert the ADC values to pressure, I needed to multiply the ADC value by .2441. The pressure transducer has a range of 0 to 1,000 psi, since the ADC has a range of 0 to 4095 (4096 numbers) divide the max pressure (1000) by the number of ADC values (4096) to get a conversion number of .2441.
An easy way to convert floating point math to integer math, is just to move the decimal point over to make a decimal an integer, then later on multiply to convert it back. But alas, the BS2 has another limitation, no integer greater than about 65,000. So I had to round off the decimal to only 2 places, this has the effect of reducing the accuracy, I could just have well used an 8 bit ADC, but I had the 12 bit so it wasn't a big deal dropping the resolution. In testing, the actual pressure readings I'm getting have never been off more than 7 or 8 psi, which is more than accurate enough for what I need.
The 3201-C requires regulated 5 volts, so I drew the power from the BS2 vdd pin. The code for the signal processing in the ADC by the BS2 was a snippet of code I modified from my homemade altimeter project.
Next up is the pyro outputs, BS2 pins P5, P6 and P7. Notice on the board I have 390 ohm resistors going from the output pins on the BS2 to each of the transistors. I'm not sure why those would be needed, but I found some documentation from Parrallax that used them, so I included them but don't think they are really needed. The transistors are all IRF 510's, capable of some 15 amps momentary and about 7 amps continuous. The transistors all have 10 K ohm pull down resistors, the pull down resistors make sure the gate stays closed on the transistor until the BS2 sends out a 5 volt pulse to close the gate.
I'm sending the transistor output to 30 amp relays. This is probably overkill, but I wanted the system to be robust and capable of lighting any igniter I could throw at it. The three pyro relays are an automotive style single pole single throw, Tyco part number 0-1432782-1. These relays require a socket, and I ordered PCB mount sockets to go with them. You can get these relays as an inline style, but I liked the idea of sockets so I could quickly replace a relay if needed.
The valve control circuit was probably the toughest, BS2 pins P3 and P4. I had to use two single pole double throw relays to reverse the current to the valve actuator motor. So that sort of made for a mess of wires to get that job done. Once I had the circuit working, I varied the time in the BS2 code to just allow the valve to fully open and close. That turned out to be .65 seconds, if you look at the BS2 code you'll notice a "High 650", that's 650 milli seconds or .65 seconds. The single pole double throw relay is also a Tyco, and uses the same socket as the other relays with a fifth pin added. Tyco part number 1-1432785-1.
To get valve position data, I used BS2 pins P8 and P9. These are input pins, so the 20K ohm resistors are required to keep the BS2 pins pulled low. I again tapped the BS2 vdd to send 5 volts out to the momentary switches that indicated the valve is open or closed. When the valve position reaches a switch and closes it, 5 volts is sent to the BS2 pin and it is read as a high input. That information can then be accessed by the PC application.
BS2 pins P10, P11 and P12 are used to test continuity on the pyro circuits. When a shunt or igniter is installed, 12 volts positive current is detected on the input pin on the BS2. I used 27K ohm resistors to drop the voltage under 6 volts, I wasn't sure if the BS2 can be damaged by 12 volt input on a pin, but I thought better safe than sorry. I also have 10K ohm pull down resistors on all three of these pins as well. Again, making sure the pin indicates a low state when not shunted.
As you may have noticed, I have three pins still open on the BS2. I could do another data line of some sort, or another pyro circuit.
Here's a screen cap of the PC application.
One problem I'm having is getting the BS2 to communicate properly with my application. It seems no matter how I try to format the data to send to the application, the application won't read it properly. That is, I can't use information such as a number in an equation. There is a way to do it I'm sure, but my limited programming skills have hit a wall. I can send formatted data to the BS2, and utilize it, but the data coming from the stamp can only be displayed as text. That's the main reason I wanted to convert the pressure data in the BS2, that way I could display actual psi, rather than just the ADC values.
Another little quirk is any time you click on a button to display data, the data displayed is whatever was left in the buffer from the last transmission. So you have to double click to clear old data and display the new data. Not a big deal as long as you know what to do... In time, hopefully I'll pick up more information or get better at programming, and I'll be able to clean up these problems.
Here's the system at least partially mounted in a tool box.
The tool box worked out quite well. You can see the relay board on the left, the 3x2 outlets mounted on the back side of the box, the box with the radio modem is in the lower left. Power comes in on the far right side side of the box through a 4 pin XLR connector. The 4 pin XLR is my standard power connection for all my 12 volt equipment. I'm also tapping power for the radio modem from the XLR connector. The tool box top tray still fits inside, so I used it to hold the transducer, power supply and wire harness. To set it up, it's as simple as plugging a battery in to turn it on, then plugging in any circuits you need.
I did a little more work on the controller, since I bought a new Canon mini DV camcorder, I thought I'd throw a few video stills in here...
Here's the box closed up, the port on the right side is for 12 volt power in.
Here's the other side of the box, on this side is the port for the N2O pressure transducer line in.
Inside the top tray is the pressure transducer and data line, as well as all the igniter whips. I made up new whips with soldered on alligator clips with new plugs on the other end. The pressure transducer power supply and the radio modem fit inside the toolbox, everything needed for the launch controller at the rocket end of things fits in the box.
I mounted the BS2/relay board to the side wall, not shown inside is the radio modem and the transducer power supply.
I picked up a couple of new pressure transducers so I'd have backups for the system. Here's a photo of the connector, since a mating connector would cost more than the transducer cost me, I cut off the connector and solder the wires instead. Notice the pin diagram. Pin 1 is the center pin, pin 2 is at top right, then pins 3 through 7 go clockwise around the perimeter.
Here's the other side of the connector, I had to get inside to determine the color code of the wires in relation to the pin out. Not as easy as it sounds since the connector has several layers of protection including a silicone potting material to dig out. Oddly enough I have a number of these transducers, and they don't use the same color wires...
Here's the pin out and color code for the transducer. All these transducers use the same pin out, that is:
Pin 1 : 28 Volt dc Positive
Pin 2: Signal Out Positive
Pin 3: Signal Out Negative
Pin 4: 28 Volt dc Negative
Pin 5: 75% Full Scale Calibration
Pin 6: 75% Full Scale Calibration
Here's the backup transducer wired up. The black phone plug is signal out to the BS2, the red RCA phono plug is 28 volt dc power in.
I ran another series of tests on the unit, leaving it powered up for several hours. Everything continues to work fine. The only thing I may add to this system is another data channel for a load cell. With a larger hybrid it might be nice to know the actual N2O load in pounds prior to launch. Of course, that information can be calculated from tank volume since I know the pressure and can calculate density from pressure.