home :: gallery :: rigs :: kites :: techniques

Kite Aerial Photography
Video Switch

Details of the design of a automatic power switch to remove power
to the video transmitter, and save battery life, when the video is not active:

Whilst designing my second KAP rig, I was interested in saving weight through using smaller batteries, as a result of lower power consumption.

A common cry in our home to my son is to switch things off that aren't in use, lights, TVs etc. So I thought I'd take this to heart in my KAP rig. The video downlink on the rig is really only needed when you're framing a shot, so lets switch it off when not in use!

You can do this, using an extra channel on your RC control, but I describe here an automatic way to do this if you are using a digital camera. When the digital camera goes into sleep mode, the Video transmitter automatically switches off as well.

The circuit is shown here (open in a new window), and breaks into 4 main section. First there is a video amplifier. This single transistor circuit is tapped into the video between camera and transmitter. Note that the signal ground comes from the camera, and the signal going back to the transmitter is NOT GROUNDED. The amplifier increases the 1Volt peak to peak signal to a higher level, in fact it clips the signal, especially if there is a lot of white in the picture. The two 10nF capacitors are selected so that the amplifier works whilst not loading the video signal to the transmitter, again this affects the quality of the signal on the collector (it has it's high frequency component removed) but this is not important in our application and the following circuitry. With no video present the collector of the transistor will be at about 1.8volts (power supply dependant), but with video applied the sync pulses on the video will make the collector rise to the supply rail. The type of transistor is not critical, any low current NPN (BC107/8/9) should work. Note however that the 1K resistor in the emitter of the 2N3904 should be dropped to 910ohms if using 4 (rather than 5) NiMH cells. This makes sure that the circuit remains operational over the extremes of voltage that are experienced as the batteries discharge.

The second part of the circuit, the detector, takes these pulses and they are used to charge the 1uF capacitor. 3 diodes are used (where one would do) to offset the voltage on the 1uF to drive the FET. The capacitor is <0.8V when there is no video, and >3V when on. This is used to gate the third part of the circuit the MOSFET switch. The diodes could be 1N914, 1N4148, or 1N4001. Any FET will work as long as it will carry the load current and it's characteristic (off to on points) are better than 0.8V to 2.5V.

The switch connects power to the MAX761 boost regulator which in this circuit is set for 9 volt operation. The IC is configured in a non-bootstrapped configuration, as per its application notes. I will not discuss this circuit in detail, many similar circuits are published, and you can use any IC or circuit you wish, in fact you don't need any circuitry if the supply you want to switch uses 4.8 to 6 volts. As with all similar boost switching regulators, it is important to use low ESR capacitors, tantalum are used for the 47uF input and output capacitors, and the 33uH coil has to have a low series resistance and be able to handle current pulses much higher than the input current.

Click for full size layout diagramThe circuit can be built on a small piece of 0.1" matrix board, 8 by 14 holes, as shown in the layout diagram on the right. In order to reduce weight and size I tend to build the circuit crudely as a "rat's nest" then once working I re-construct it as per the diagram. I do not use strip board as this tends to limit your ability to squeeze the circuit down to minimum size. I have used "spot" board, with copper pads on a 0.1" matrix, or you could use matrix board with no pads at all. Use the folded over leads from the components, or small pieces of wire, to form the tracks on the circuit side of the board, shown in blue. Occasionally, to make room for the fixing hole or reduce the size, I have not run the wires on the circuit side on the 0.1" grid, these wires (shown in dark green) should be insulated to stop them shorting to the wires on the 0.1" grid. Where the component is surface mount (the FET and the 47uF capacitors) it is connected via wires that come up from the circuit side, the opposite of the through hole components :-)

Although the layout is specific to my MAX761 design, I hope it gives you the seed for your design.

Flight Test

Further experiments show some picture interference when using the video-switch. These are not visible using a small hand-held TV, but can be seen on a larger monitor. The interference is because the video transmitter takes current pulses out of the supply every field sync (20mS). This was solved by placing a ferrite between the circuit shown and the video transmitter, and adding a 1000uF 16V electrolytic as close to the transmitter as possible. This means that the 47uF acted as the reservoir for the MAX761, whilst the 1000uF was the reservoir for the transmitter. 

All images on this site (unless explicitly stated) are the property of James Gentles under the UK Copyright, Designs & Patents Act 1988.
If you are interested in using any of this material privately or commercially, or seeing more images, please .