|
The aim of this project was to get a digital camera into a small electric radio controlled (RC) aeroplane and still have it fly. The aeroplane shown, a park flyer, weighs between 400 and 550 grams depending on the size of battery and wing configeration that I use. The larger picture shows my park flyer with the original wing (a borrowed design, source unknown, from Peter Fuller of Model House) while the smaller photo, above, shows the aircraft with a wing which has ailerons (my own design) to give me a feeling for full house control. (photos thanks to Gelinda & Gert Stange 2005). Both photos were taken when testing (sucessfully) an early model FMA IR optical flight stabilsation autopilot which can be seen mounted on a raised platform behind the wing.
I came across a "Precision Mini Digital Camera" for $30 Austalian. It can take 26 poor photos at a maximum resolution of 640x480 pixels. The lens seems to be the cause many aberations, particularly at the edges.
The memory is SDRAM, not FLASH as is more commonly used, so, for the pictures to be retained in memory, the single 1.5V battery must (should) be kept in place, drawing a standby current of 2mA. On startup the camera can draw up to 150mA and then settles down to about 80mA. The draw on the USB port (at 5V) is only 50mA settling to 20mA.
For connection to a standard RC receiver, my interface circuit is shown in the diagram below:
The VIP of the circuit is a PICAXE08. If you don't know about PICAXE then check them out at www.picaxe.co.uk and then go googling to find your local supplier. Last time I checked they were about $3 ( Aust.) each, so postage was the main concern. They are handy little critters, so buy a few. You can program them from any Windows PC with a couple of resistors and a 9 pin socket. A programming and simulation program can be downloaded.
On internal examination, it appears that the 1.5V from the battery is switched through a small inductor and then regulated up to 3.3V for internal use. The USB port can operate as an alternative 5V power source and draws about 20mA. Since the RC system is between 5 and 6V, I decided to elliminate the 1.5V battery. [ERRETA Jan06': I would now tend to think a seperate 1.5V AAA battery is a better option. See discussion in video mode page.] To do this, I have run the 5V RC +ve through a diode to downstream (positron flow) of the USB power diode [D20] on the camera (labelled USB diode below). If you put the power in upstream of the diode it goes into PC mode and won't always take pictures.
I also needed to have 0.5V to the centre of the voltage divider [R12 & R13] (a blurred label at the bottom on the photo above) so that the camera does not think that it's battery is going flat. This is done with my own voltage divider and another diode.
[ERRETA Jan06': The lens housing is only held on with a tiny bit of glue. I almost lost it, so I've added a bit more hot glue.]
The next problem is that the camera, thinking that it running off battery power, turns off after 20 seconds. When it does, the memory goes into standby mode, seemingly drawing it's 2mA of current directly from the 1.5V battery. Possibly one could use a 1K resistor and diode from the 5V RC supply. I haven't tried this yet as I think there might be more problems. Instead, what I found worked was to supply power to upstream of the USB diode, when not taking pictures. This puts the camera in PC mode, which does not allow you to take pictures (not without an actual program running) but does keep the memory alive. However, instead of drawing 2mA, it draws 20mA. The power is switched through a transitor controlled by a PICAXE08. It could possibly work directly without the transitor, since no current is actually drawn.
The camera shutter is +ve triggered, so to take a photo, an output of the PICAXE08 is programmed to go positive. After a short period, this input returns to a high impedance input state - saves having an extra diode. If you look at the program, the PICAXE measures the pulse length for an RC channel. If this pulse goes longer than 1.2 milliSeconds long, it first turns off the power to the USB port (retaining power to the regulator) to get the camera out of PC mode, then it waits a little while for the camera to settle. Then when the RC channel pulse goes less than 1.2 milliSeconds, then it triggers the shutter. It waits a further period for the camera to settle before returning the camera to PC mode.
You can download the PIC AXE 08 program but it's so short I'll just show it here too:
loop1:
high 4 ' leave PC mode
loop2:
pulsin 3,1,b0 'RC input
if b0 < 150 then loop2 ' failsafe & no action
low 4 ' turn PC mode off
pause 500 ' give camera time to settle
loop3:
pulsin 3,1,b0 'RC input
if b0 = 0 then loop1 ' repeat off on
if b0 > 130 then loop3 'wait for shot
high 1 ' trigger shutter
pause 500
input 1 ' return trigger to high impedance
pause 1500 ' give it a chance to trigger
goto loop1 ' PC mode to keep camera on
As noted, there are a few steps which could be simplified.
When in doubt, leave it out. For installation I just stick taped the camera to one of the wing mounts.
The camera had a final "weigh-in" at 6 grams without a case. The case + LCD module weighs another 6 grams.
And now for the results of the first flight 23rd April 2005:
When I get a chance I'm going to wire the picAXE to the mode switch (as already shown on the circuit diagram) so that I can do some video shots (easier to aim) or some lower resolution stills.
As you can probably also see, the lens is not very good.
[ERRETA Jan06': The lens seems to have been focused a little short. See discussion in video mode page.]
Have a quick look at my RCvideo page.
If this wasn't really what you were looking for in aerial camera's there is a good chance you will find it at one of the following sites: