Need help replacing existing controls.

Fletch

Member
I'm a bit of an electronics novice so I'm hoping someone can answer what is probably a dumb question. I want to wire a Picaxe into an old camcorder to allow it to be remote controlled. Since the camera doesnt have nice things like a LANC port this will have to be done by replacing various switches and pots with PIC controllable equivalents.

For the switches I had been thinking of using reed relays because that way the PIC and the camera circuits would be seperate. Is there a type of transistor that could be used instead? For low voltage signals I'm assuming a switching transistor is all I need? Would a simple transistor switch be enough to isolate the PIC from the camera electronics?

Does anyone have any experience using digital pots with the picaxe and know of one that works well?
 

BeanieBots

Moderator
Without a circuit diagram of your camcorder, nobody can answer that for certain. The problem is knowing WHAT the existing switches actually switch and and at what voltage. In many cases, it is likely to either pull a signal down to 0v or up to some power rail voltage.
With a DVM, you could probably work most of it out by measuring either side of the switch to a common reference such as the battery 0v.
Reed relays would be the safest option. Analog switches such as the CD4066 would be another option but even with those, it is not total isolation. The voltage that they switch must be less than their own power supply voltage.
As for PICAXE controlable pots, there many types available that can be either controlled with simple up/down pulses or by I2C commands. However, just like analog switches, they are NOT isolated and are bound by supply rail voltages.
I am not aware of any that offer true isolation but that does not mean they don't exist.

Opto isolators is another avenue worth investigating. You could even consider running a digital POT from the camcorder power supply and using an opto in turn to control it. That would offer true isolation. You would still need to know the signal magintude that you want to control in order to correctly specify which pot to use.
 
Last edited:

Fletch

Member
That's a good answer but while it helps it's not exactly what I'm looking for. I didn't mean to press "isolation" quite that hard. I don't know that true electrical isolation is absolutely nescessary for my application, I just want a solution where the pic and the camera dont interfere with each.

What I'm looking to do is build a system that allows for the remote control of cameras. Most basic use will be to fire off the shutter on a simple P&S digital camera most complex application would be to control all basic functions on a small digital video cam. The end objective is Kite/Balloon aerial photography, IE taking pictures/video from the air from a kite or balloon tethered a few hundred feet above the ground.

I decided to use a microcontroller for this so that I can build in some functions that would not be in a standard hobby remote control. Chief in these are to send a periodic "stay awake" signal to stop the cam from closing down if a picture is not taken for several minutes, and automating "rotate and shoot" functions for making stitched panaramas.

Anyway, I was going to do this with reed relays since my electronics knowledge is not really enough to think of a better way. In researching this more I found a post on a pic board (maybe even this one, it's become a blur) that suggested that for the small digital signals involved the PIC could switch the focus and shutter signals of a digicam itself. I dont think I would want to go that far but it did raise the question as to if I could do the switching with transistors/darlingtons/FETs so the final design would be smaller and lighter than using maybe a half dozen small relays.

My problem is that I have a lot of potential choices and not really enough information to know which ones are good and bad. The picaxe manual has darlingtons and FETS, the analogue switches you suggested are yet another route. I'm sure they all have advantages/disadvantages over the each other but it's not clear what they are.

If you wanted to "press a button" on another device with a picaxe, which kinds of switching would you consider, what kinds of conditions would make you prefer one solution over another?
 
Last edited:

Tom2000

Senior Member
If you wanted to "press a button" on another device with a picaxe, which kinds of switching would you consider, what kinds of conditions would make you prefer one solution over another?
As Beanie mentioned, you really need to know how your camcorder is handling each button before you can think about getting tricky.

Your reed relay idea is safest. And, if the buttons are scanned (and why wouldn't they be?) probably the only solution.

Tom
 

kevrus

New Member
Choosing a device to switch something without knowing the format of the signals bieng switched...i.e. current bieng switched (probably quite low in this case), voltage levels and polarity (e.g. negative or positive switching) is going to be a stab in the dark.
An electromechanical relay overcomes most of these basic issues if they are unknown but can be a little more power hungry and cumbersome in size in comparison with a small signal transistor for instancewithout further information. Another issue may be in the length of time a 'contact' needs to be 'activated' for example in the case of zooming in and out. With regards to the panoramic photos.
How will you know when to take the next shot, if taking stills, to ensure seamless photo stitching if the camera is airborne and remotely operated unless you had a transmitted video signal and a remote monitor to view the picture/scene?
 

Fletch

Member
Actually the pano part isn't that hard. The basic rig for KAP uses a 360 degree modified servo to pan the camera assembly. Modern software stitching software works best the more overlap you have between pictures, for shots on the ground I normally chose to overlap each picture by 25-30%.

Pan and shoot would work like this. When I push the auto-pano button on the remote the Pic will instruct the pan servo to rotate aproximately 30 degrees, stop and then fire the shutter. The pic would repeat this 13 times such that there will be considerable overlap on the first and last picture.

There are already small PIC controllers made for KAP that work completely independently, you program them in advance with rig movements and camera shooting instructions then send them up with no further control from the ground. It works because your average digicam can store hundreds of pictures on a storage card and so multiple duplicates or bad pictures can just be discarded. I have a small 2GHz video transmiter/receiver pair that I intend to repackage to send live video back from the DV cam but that's a more advanced project that needs a bigger rig/kite.

Size and weight are the main problems for KAP. If there is a way to do as much as possible with a small amount of hardware then you go with it. That is really why the PIC is so interesting. If I can get it to move the rig and work the camera then I have eliminated half of the hardware I need to fly. If 5 transisters take up the same space as 1 relay, that saving is useful too, which was why I was interested in potential solid state solutions.
 

BeanieBots

Moderator
As has already been repeated, you will need to find out more about the signals being switched before you can go solid state with the switching.
Get hold of a scope and have at look at the waveform between the two contacts of each switch you wish to replace.
If it is a flat DC voltage referenced to a common line, then a single transistor is likely to work but still can't be promised. If you 'see' a repetative waveform, then it is likely to be a scanned button. You will then need to know the voltage of the scanning signal.
Armed with that information you could power analog switches at the required voltage and use them to do the switching. (CD4066 as mentioned before).
If the voltage is above 5v, then you will also need interfacing between the PICAXE and the analog switch which will add to your component count.
Very small reeds are available (5mm X 3mm X 3mm are the smallest I've ever seen) but they cost a lot due to their specialised use. Regular 14pin DIP size are much cheaper and easier to source.
 

Michael 2727

Senior Member
As has already been repeated, you will need to find out more about the signals being switched before you can go solid state with the switching. ,,,,,,,,,,,,,,,,
If the voltage is above 5v, then you will also need interfacing between the PICAXE and the analog switch which will add to your component count.
You may get kucky with the 5v or it may even turm out to be 3v, in which case you will still be lucky.

Just remember, if there are any switches that carry some real current e.g. 50mA > in the
unit I wouldn't use a reed switch, as the contacts may stick/weld together.

As I recently pointed out to a local electrician who tried to use a magnetic door switch
(reed) to activate a 12v 200mA strobe light. He wondered why the thing was getting stuck ON, Dµh
 
Top