Please Note: there are now two versions of my TV transmitter!  This page describes my original version.  Another page has been added to show the new version.  However, many of the topics on this page still apply to both versions since they both use the same camera, GPS and on-screen overlay system.  Ideally I would recommend reading this page first and then moving onto the new page.  At any rate, with that out of the way, here are the details on my first on-board TV transmitter system.

During the summer of 2000 I became very interested in designing and flying a TV transmitter to get live video from a rocket while it was in flight.  Furthermore, I was much more interested in building the equipment myself than in simply buying a commercial unit.  I wanted to learn more about high frequency transmitters and receivers.  After looking at all sorts of different options, I eventually settled on a kit from North Country Radio.  They have a relatively inexpensive kit for transmitting TV at 1.3 GHz in the amateur (HAM) band.  I liked the idea of working at 1.3 GHz because the antenna would be small enough to fit within the nose cone and yet I would get a longer transmission range than I would at 2.4 GHz.  Besides that, 2.4 GHz was just starting to become popular due to the availability of some consumer transmitters.  I also figured I would have less problems with interference with other people at large rocket launches if I was at 1.3 GHz and everyone else was at 2.4 GHz.

Videos obtained by this system can be viewed on the Wildfire video page.  Look for the screens on that page that are labeled "on-board video".

This video transmitter system is built as a module that gets deployed on its own parachute at apogee.  The camera is looking out the side of the rocket at launch and is looking down at the ground during descent on the parachute.  Optionally an external mirror can also be added to allow the camera to look down during lift off.  Details of the "look down" mirror design can be seen right here.

The transmitter antenna is in the nose cone of the rocket and the electronics are packaged in a unit that is just below and is attached to the nose cone.  Gravity is all that is needed to deploy this system at apogee.  It slides out of the front payload bay of the rocket once the rocket itself is suspended on its own parachute.

This is a view of the front of the unit once the outer covering as been removed. (The outer covering is simply a piece of "coupler" tubing. The coupler tubing makes the whole unit fit perfectly into the payload bay of the rocket.)

In this view we have rotated on around to the back side of the unit.  The two red battery packs are still in place.

The transmitter kit I selected is the ATV12-1300 from North Country Radio.  It cost me $143.  This is a 1-watt transmitter that has provisions for three channels in the 1.24-1.30 GHz band.  It uses AM modulation for the video and also has an input for the sound subcarrier.  The kit includes a big bag of parts, a set of schematics, some pc board loading diagrams, and some very well written instructions.  However, some familiarity with electronics is needed.  I would not recommend this kit to a beginner.  You will even spend a fair amount of time winding your own coils in order to build this kit!  You will also likely need something like a Bird Model 43 wattmeter (or equivalent) to help get it tuned up and running properly.  (They are available second hand on eBay and are also very useful for tuning the transmitter antenna to the optimum length.)  This kit took me about 3 weekends to build.  The transmitter pc board is 2.5 x 4.0 inches.  I mounted it inside an aluminum box to shield it from the antenna.  The box is a Hammond Manufacturing part number 1590BB available through Digi-Key as part number HM152-ND for about $11.

In order to receive the 1.3 GHz TV signal, I selected the DCNV-1300 down converter also from North Country Radio. This was also a kit and the cost was $45.  It converts the 1.3 GHz TV signal down to an intermediate frequency for chan-3 or chan-4 to enable viewing on a standard VHF TV receiver.  This kit was much simpler to build than the transmitter since it has far fewer parts. The down converter pc board is also 2.5 x 4.0 inches. I used the same type of box for mounting the down converter as I did for the transmitter.

For a receiver antenna I selected the model 23CM22EZ from M² Antenna Systems Inc. It is a yagi style antenna that is specifically made for the 1.25-1.30 GHz band and offers a gain of 16 dBd.  It cost me $125.  I mounted it to a boom that has two handles so that a person can easily hold it and aim it at the rocket during flight. This is a high gain and very directional antenna so it must be aimed at the rocket during flight.  However, while the rocket is still on the ground the signal is so strong that it overloads the receiver and the antenna needs to be pointed away from the rocket to maintain a good quality picture!

For the transmitter antenna I used an inverted-V configuration and tuned it by trimming the length to minimize the reflected power as measured by my Bird 43 wattmeter.  The dipoles on the antenna ended up 2.17 inches long and since they formed a "V" they easily fit inside the plastic nose cone of a 5.5-inch diameter rocket.

The color TV camera I selected was the CCD305 from Ramsey Electronics.  It was physically small enough and relatively inexpensive for the resolution it provides.  It is rated at 380 lines of resolution and it uses the SONY 1/3" Super HAD CCD sensor. It cost me $134.  The specifications for it are right here.

As for sound, I was not very happy with the results from the microphone that is built into the camera.  It tended to deliver some buzz in the audio and also seemed to distort and generate more buzz due to the loud roar from the rocket motor and wind noise.  Consequently I ended up adding a different microphone and a small amplifier card of my own design.   Perhaps someday I'll find  the time to post the design here.  It seems to have really helped the sound quality in the more recent flights.

In addition to the TV transmitter I also added a global positioning system (GPS) receiver to this design.  This allows the TV image to also display the altitude and the location of the rocket.  For the GPS receiver I selected the Garmin GPS25-HVS unit.   This is a small unit that is designed to be built into other equipment.  The specifications for it are right here.  The complete documentation manual (in pdf format) is available here.  This particular GPS unit is nice because it can be powered from the same 12V as the transmitter and the camera. 

An antenna is also needed with the GPS receiver. I used the GA27A  but I think it has been replaced with the GA27C also available from Garmin.  This antenna contains a preamp so it can be mounted remotely from the GPS receiver.   The GPS antenna, some cables and some software are all available as part of an evaluation kit (part number 010-10197-00) from Garmin.  I bought the GPS receiver for $140 and the evaluation kit for $110 from a distributor called GPS City.  By the way, if it has anything to do with GPS, you'll find it at GPS City.  Besides selling GPS units, they also provide a discussion forum and a whole lot of good informational links to GPS information.

While on the topic of GPS, here are two links I also found very useful.  They both have a great deal of information about every aspect of GPS.  One is Peter Bennett's GPS and NMEA Site.  The other is called Global Positioning System (GPS) Resources.

To get the GPS information to show up on the video image, I used an OSD-GPS on-screen overlay board from Intuitive circuits.   This board also provided an easy way to display my HAM call sign (KD7KYL) on the video image.  The board can be powered from the same 12V supply as everything else.  It connects between the camera and the transmitter and overlays the information from the GPS on the image.  The GPS connects to the board through a two wire serial interface.  The OSD-GPS board costs about $120.  One draw back is that the speed it displays is "ground speed" and not "vertical speed".  None the less, it also displays altitude and position. The position information makes it extremely easy to locate and retrieve the transmitter once it has landed after a flight.  Just plug the coordinates into a hand held GPS unit and have it direct you to the landing site

To power all this on the rocket, I use two battery packs that are designed for remote controlled electric toy cars.  These battery packs are high capacity NICAD cells. They are mass produced so they are relatively inexpensive.  Each battery pack produces a nominal 7.2V.  I wire them in series to get 14.4V nominal.  However, this is usually as high as 16V when the batteries are fully charged.  And 16V is too high for the transmitter and other electronics.  Consequently I designed and built a special low-dropout power regulator card that maintains a constant 12V to the electronics at all times that the battery voltage is above 12V.  If the battery voltage drops below 12V then the regulator basically becomes a "wire" and connects the batteries directly to the electronics. 

The complete electronics package draws a total of about 0.8 Amps. (The bulk of that is of course the transmitter.)  These battery packs will normally give 90 minutes of service at that rate before they are dead.  This is plenty of time for a normal launch sequence, even if some unexpected delays occur.  (And they often do!) The power regulator card also contains three status LEDs that indicate the battery charge.  Green is good, yellow indicates there is only about 10-15 minutes left, and red means the output is now below 12V and failure is likely within a couple of minutes or so.

To view the video transmission at the launch site, I use a small portable TV that runs on 12V from a car battery.  This TV is a combo unit that also includes a VCR.  I record the transmission on VHS tape and then later (at home) I play it back, digitize it and convert it to mpeg for display on the computer.   Unfortunately this whole process still needs some improvements.  I lose image quality when I record it on my relatively cheap, battery powered VCR.  I then lose a little more when it is played back later to be digitized.  The digitizing hardware is also not very sophisticated  so I lose more there too.   Someday I hope to improve this whole process so that what I see on the computer is as good a quality as what I see on the TV screen when it's "live".   The live video image is usually quite good.

One more future refinement is to replace my inverted-V antenna with something else.  Even with all the power I am transmitting, I sometimes loose image quality when the transmitter and receiver antenna are not in alignment.  This is a good project for 2003.