Another update on my HAB project. I’ve been working on the radio system for the past month or so. Mainly I built two antennas that were capable of reaching the required 50 mile or so range. When I first started researching the required antenna specifications, I was extremely confused by all the different measurements and types of antenna. This is where talking to a real expert can be awesome.
David Patterson of Edge Research Lab kindly assisted me in figuring out all the fine details of making my antennas and design considerations I needed to take into account. After an exchange of emails, the feasibility of this radio link hinged on a line of sight connection and a high gain antenna setup. For starters the antennas have to be circular polarized so the spin of the payload does not cause any loss or noise from signal bounce. Additionally, the recieving antenna must be high gain to capture the weak signal, while the transmitting as omnidirectional (low gain) as possible to broadcast the signal so it can be picked up from all sides of the payload.
Specs of the receiving helical:
- 4.15 inches diameter drain pipe for the former
- 8.5 feet in length
- 30 turns of 1/4″ copper pipe
- ~30dbi gain
- HPBW of 20 degrees
- Ground plane reflector is 10 inches in diameter 3mm below the last turn of the helical
- Soldered on to the first quarter turn is Alex Greve’s (IBCrazy on RCGroups forums) waveguide impedance matching strip; the bare helical measures about 140 ohms, this should match it to the radio’s 50 ohms.
The cloverleaf is based off this design by Alex Greve. I had to enlarge it for 900mhz dimensions, which made the construction a bit more difficult than for the more common, higher frequencies used in RC aircraft. I used 12 gauge solid core copper wire for the lobes, carefully soldered to RG58 coax.
One of the snags I hit was the RG58 I had on hand was part of a SMA cable with too-large connectors. Thus i had to order a properly sized one from Digikey, but ended up destroying part of it in the process. Turns out the center conductor is not mechanically fastened, so when i tried to strip the cable it pulled out and rendered it useless. That night I scrapped a few of the stock dipoles that came with the EasyRadio modules, soldered the SMA connectors onto a fat length of the RG58, and built the cloverleaf on that. I had to build a jig to hold the wire lobes in place while I soldered, then covered the joint in hot glue. Both antennas are right-hand circular polarized.
For this launch, I’m going with my Canon SX210IS. I loaded and tested the CHDK build for it, works like a charm. I did some battery life tests and the intervalometer took 1400 images on a 10 second interval. It’s pretty large but has good specs and I don’t want to buy another camera. The only reason I’m not using the SX110IS from last year is the screen has broke and I want a few more megapixels and sharper images. I took apart the zi8 that failed last year, might include it if I can get it to record 30 minute clips with the pic32, but at the moment it seems like a lot to work with.
I’m using the same 1.5″ thick pink insulation board, in an 8x8x6 inch cube. The interior dimensions are a lot smaller and the payload a lot lighter. On the bottom will be a hole drilled in the center for the cloverleaf antenna to poke and rest on the bottom. Since the antenna is fragile, It will probably be covered in a plastic Tupperware for protection.
I bought another 1200 gram balloon from Kaymont, and the smaller payload (under 1.5 pounds I hope) should allow me to reach 100,000 feet and higher quite easily, while maintaining a high decent rate. The parachute is a 48″ from RocketChutes, same one I used last year with great results. Some 1/8″ nylon cord will finish off the flight string.
The ERA900TRS modules should provide two-way communication the whole way, but I have the SPOT II onboard as well. It tested well on the first launch. It only sends a GPS locatin every 10 minutes without altitude, and cuts out above 60,000 feet, but keeps transmitting when it gets below 60,000 feet again. Hopefully on my third launch I can omit this and lighten the payload further. Sensors onboard will include a Copernicus II GPS i verified on my first launch, two versions of the Ultimate GPS, BMP085 barometer, and maybe some MEMS devices if I have time to hook them up. Temperature is measured by the barometer and radio, as well as the MPU-6050 also onboard.
Testing and Launch Date
I plan to launch in eastern Washington on September 22nd. This past weekend I attempted to distance test my antennas, but the easyRadio software wouldn’t cooperate with my laptop although we tested it successfully last weekend. Troubleshooting isn’t fun :(
Low Power Radio Solutions has kindly donated lots of time and radios for my project. The EasyRadio modules are very nice to use and I got them up and running quickly with my uController.
Adafruit Industries also donated several Ultimate GPS breakouts, and have been helpful to my questions when I go on the weekly Show and Tell.
Many thanks go to David Patterson at Edge Research Lab for helping me with the whole RF system and sending me a Airview9 spectrum analyzer and Rick Winscot who helped me get the EasyRadio modules online and donating a few breakouts to LENSE.
Hopefully my next post can include some test results and pictures of my PCBs when they come in!
The coax cable goes Inside the former and out a hole in the ground plane
The coax feed line comes through a hold in the base of the former and is carefully soldered on to the groundplane and coil.