| Quick Links |
| Home |
| Propellants |
| Rocket Motors |
| Rockets |
| Rocket Electronics |
| Ground Support |
| Launch & Static Tests |
| Software |
| Links |
| Motor Class Table |
| E-Mail Scott |
|
| Don't Click Here! |
So I started this project by soldering up the components on the OEM board. The instructions are clear and easy to follow. It didn't take long and I had the OEM board ready to go. So I hooked up a battery and tested it on my PC, low and behold it worked perfectly! So now it was time to add the OEM board to a bare PC board and add the components to make a new altimeter. I perused my selection of PC boards, finally deciding on a larger board, just in case I wanted to add some goodies down the road.
I wired this up identical to the first altimeter I made, using two 10 amps transistors, one each for the apogee and main deployment charges. It also uses two batteries, one to power the Basic Stamp 2, and another for the deployment charges.
Here's the finished altimeter. It's 2 3/8" x 6 1/4".
Once everything was in place, I again hooked it up to my PC and ran a few tests. First checking output voltage at apogee and main events. Then, I checked the amp output at both events. Again, everything worked as expected. To finish up, I used hot glue at the bottom of the OEM board to help retain it. It's pretty strong with all those pins at the top, but for insurance I added the glue. I also added a drop of hot glue under each of the output transistors, again, probably not needed, but... To retain the pressure sensor I used two #4 nylon screws through the board, I can't see that going anywhere either, better safe than sorry.
The battery clips you see in the picture are just for bench testing, when the altimeter is installed on an electronics module or bay, it will be wired through a power switch and the batteries secured in holders. Next I'll run some tests in my vacuum chamber with some high current light bulbs for more thorough testing.
Here's a schematic for the altimeter.
Using the OEM Basic Stamp 2 kit brought the total cost down to under $60, another nice feature is the ability to change out parts easily. Even the Basic Stamp 2 interpreter chip can be replaced, it's $11 from Parallax.
I considered adding continuity checks to the outputs, but I always carry a meter and test the deployment initiators before and after installing anyway. A beeper or LED could be added too, there are plenty of input/output pins left on the stamp. I think I'll add a beeper in the future, just so I know the altimeter is running inside the rocket. I suppose I could have it beep out the max altitude data too, I'll have to work on the code for that.
Update 31 May, 2005:
While preparing the Cosmo 2 rocket, I decided to run a few more tests on the newly built altimeter. While it worked well in ground tests, I did notice it had more drift in the output than the first altimeter did. So I did a little research on the subject to see if I could clean up the noise in the data. The first altimeter wouldn't vary more than 3 or 4 data values, this new unit will sometimes vary as much as 12 data values just sitting stationary. That's not good, and could result in premature apogee deployment.
There are really two causes, white noise and flicker noise. The noise to a certain extent is inherent in the design of the sensor, and I'm using a 12 bit ADC rather than an 8 bit ADC, making the noise problem even worse. Using decoupling capacitors in the power supply and the sensor output should help. But I tried that and it only seemed to make matters worse. It's entirely possible my wiring pattern is the cause, wires crossing where they cause interference and so forth. In the future I may try a different layout on the board to help eliminate some of the noise.
For now though, I only have one other option. That's to use software to sample and average the output value. The more samples in the average, the better the filtering. There's a caveat though, each sample requires two bytes of ram and the Basic Stamp 2 only has 26 bytes of free ram in the first place. After the programs use of RAM, I had 16 bytes available, of which I used 12 to take six samples and average them and two more bytes to store the new averaged sample.
Ok, so what's all this mean? Basically, I re-wrote the software to control the altimeter. It's a lot different than the first variations. I also went ahead with a piezo speaker and added some tones to the altimeter. I plan on flying the Cosmo 2 rocket with a transmitter, and by varying the tones at different events in the flight I should be able to hear what's going on during the flight.
When powered on it beeps two one second beeps.
When sampling for liftoff (looping for liftoff detection) it makes a high pitched warble sound.
Once liftoff is detected it is silent during the mach delay.
Once out of mach delay the computer is sampling for apogee (again looping until apogee is detected) and makes a low pitched on/off sound.
Once the apogee charge fires, the computer goes to a high pitched .5 second on/off beeping while sampling for main chute (looping again) detection.
After the main chute deployment the critter shuts up, because at this point my sonic beacon should have deployed and it's really loud and would drown out any tones from the altimeter.
Another problem I ran into was with the output transistors to the deployment initiators. I've been using an IRL520, and I don't think it's quite the transistor I should be using. For some reason I've had several of them go out on me while ground testing on altimeters/timers. Here's another case where my lack of knowledge in electronics shines through, so I installed a transistor that's used in other commercial altimeters, while it's only rated at 1.1 amps, my initiators only draw about .5 amps, so they should work fine.
I've tested the new software 10 or 12 times in my vacuum chamber, and it seems to be working properly. I'll post the new software after a successful flight or two. I considered waiting to launch the Cosmo 2 rocket until after I tested the altimeter on a lower altitude flight, but you know what? If it doesn't work I don't think it will matter if the rocket comes in ballistic from 1,000' or 7,000'.
Update 6 June, 2005:
The new altimeter with the new code was successfully flown in launch test 97. I posted a link to the code in the altimeter home page.