RF remote slave flash?

[212]

I have used remote slave flashes with cameras before. The ones I have detect the cameras flash and trigger the slave when it sees it go off. I want to mount a slave where it can not see the camera though, or even cover the cameras flash so it can not be seen.

I think I can figure out how to do this, but before I waste my time... how fast do the little RF units and the Picaxe, that you guys have been using to send data, work? Think the flash can be made to work fast enough to be in close sync to the cameras flash???

 

[Dippy]

No, not for short exposures.

Think of the time involved. Receiving, decoding, code responding.
Even if you went the total-hardware-route you would be pushing your luck e.g. HT12E-RXtx--RFRx-HT12D. Maybe if you had an arrangement where your TX end sent the RF then controlled your shutter you may get some joy. i.e. delayed the shutter to compensate for RF/code/decode delay ??

 

[212]

That's too bad The camera would be in a case unattended, and it would take a lot to change the camera settings for a slower shutter at night. At night the flash freezes movement, but during the daytime, pictures are blurry with any subject movement... if it is too slow.

 

[Dippy]

If the pictures are blurry in daylight you must be using a pretty slow shutter ... unless the subjects are real fast.

I'm not saying it's impossible. There's a lot of playing you need to do e.g. find out the time delay between your shutter release command and the shutter firing, which may be a big problem esp if Autofocus as delay would be unpredicatable.

If Autofucus is disabled and the delay is exact every time then you're in the game.

Do you have a flash synch socket or hotshoe adaptor? Measuring synch pulse to (connected) flash may be a useful figure too.

You my be able to do it with a single RF blip with the right gear - and miles away from Dr-Acula with his 15MWatt Beijing transmitters which will be transmitting on every frequency inc 434.

The main problem as I can see using the normal RF modules with data is the 'delay to stable data' figure on Rx modules and then a serin response time or even with HT12E/D there is quite a delay.

I'd also consider single blip with microwave modules too. Signal/ratio will be much better for single blip and you can get Tx/Rx devices with no middle-man electronics so you can tap into the bare signal. I don't know whther it would work, I'm just throwing suggestions to you. (i.e. thinking out aloud).

Whilst delays can be compensated for, the prob is that the longer they are then the more likely you are to get out of synch. If there any inconsistencies then you're muggered.

Actually sounds like a fun project. I wish I was doing it.


Are there not devices on the market already? I haven't looked.


If you can be bothered, then good luck. Very interesting project.

PS. You'll need a good 'scope.

 

[moxhamj]

Eee ecky thump, Dippy!

Hey, Manuka and I need to get some sort of comms working across the Tasman. If we can't use the Beijing Bangers we will have to go to spark gaps.

Actually, Dippy might be on to something. Use an RF module and a picaxe, and flick a manual switch at the Tx end that says "I'm about to send a flash". The Tx then sets a pin high and this goes into a digital AND gate (74HC08) along with the flash signal. The Rx picaxe then receives this packet and sets a pin high. Then use a logic AND gate at the Rx end and take the signal from that picaxe pin and the signal from the radio Rx. The next thing that will happen is the flash goes off and the radio signal goes high and this all works at megahertz speeds and the high is detected at the other end and sets off the flash. Once the Rx unit gets this pulse, it could then trigger a 555 monostable to extend the pulse width and the picaxe could detect that pulse and go back into waiting mode. This ensures no spurious flashes go off.

How far do you need to send the signal, and does it have to be radio or could you use wires?

 

[Dippy]

Mugger Me. I didn't follow that.
How do you ensure the shutter goes at the right moment? There's always a delay (autofocus, exposure calcs, etc). And is any delay consistent? Lots to check on 212's camera I reckon.

I'll leave it to you guys.

 

[212]

Yes, there are commercial RF slave flash units, but they are expensive, and the ones I have seen all use house current for powering them. (I have not looked recently though)

I want this for use with my "trail cams" A bunch of us build our own camera units for taking wildlife pictures. Most are used for scouting deer before and during hunting season, but there are many other reasons to have them too....security...or just to see what comes in your yard after you go to bed at night. For me it is just a hobby, and a toy to play with...I like to see what comes around down by the creek.

We mostly use point and shoot digital cameras now, and the ones with a fixed focus are hard to come by anymore. Like you said, it would be almost an impossible task to predict when the flash will go off. But....ther are per-flashes that check exposure and set the white balance before the main flash. That might be a way to work with this???

Right now we have a slave controller that maintain the charge and triggers a slave, but it has to actually see the cameras flash. This works well enough I have to say, but if it were RF, we could place the flash anywhere...and cover the cameras flash up. Animal eyes light up brightly with a cameras flash. But...put the flash farther away from the camera lens, and it does not ruin the picture. Another reason to go wireless is, animals tend to like to eat the insulation covering wires... Oh yeah...these cameras do not have a hot shoe connection on them.

I know a chip can be fast enough, because I have a "learning" flash trigger that detects the proper flash timing. The question was really ...do the cheap RF units have the speed needed???

 

[Dippy]

"ther are per-flashes (pre-flashes?) that check exposure and set the white balance before the main flash. That might be a way to work with this???"
- I thought those were for anti-red-eye to close the eye-ris(haha). Oh well, I've learnt something.

Maybe.

But if your expecting to trigger RF then receive at t'other end using Data/Serin then you'll be lucky UNLESS you can pre-empt the camera shutter.
i.e. if you trigger the camera shutter and transmit RF triggered by hot-shoe or synch plug then it'll be too slow. By the time the 'other end' has decoded the RF and said "Oh, I'd better flash now" your shutter will have shuttered. Hence the suggestion about pre-empting the shutter or delaying the shutter, to give the RF a chance to get through to the slave end.

It was only a thought and maybe someone has a brilliant and instant solution.

 

[212]

I forgot about the 'anti red-eye" flash setting. Many cameras do have that in the menu, But using that delays the shutter event more time too. In many cases a half second will make a difference in getting a good shot of something moving fairly quickly, and getting a poor one. Here are two pictures taken that show what I mean about speed being important. If the first one had taken just a little longer to shutter, I would not have gotten it at all.

I think it can be done, but I'm not finished with my first 3 Picaxe projects yet haha...I will come back to this once I get the Picaxe working like I want it... with the RF video camera controller....and remote lights...and...and...

This is FUN stuff

 

[eclectic]

Possibility?

212.

It might be worth buying the USA-ebay version of this.

http://cgi.ebay.co.uk/Hot-Shoe-Flash-Wireless-Radio-Remote-Trigger-Studio_W0QQitemZ180223249854QQihZ008QQcategoryZ48515QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

Then, you could use it for testing whilst you're developing a Picaxe version.

e.

 

[212]

Hey thanks!...and Liu Hui Xia thanks you too...he just got another $25.89

 

[Dippy]

Abs. bargain.

 

[eclectic]

I wish I had a crystal ball.

Then, I could work out the tenor of Dippy's replies.

Gnomic.
Literal.
TtP.

e.

 

[212]

Hey....this RF flash trigger is cool...and it works too

I used a Sony P31 as a test camera. The camera has a pre-flash, so I connected the RF transmitter to a Wein digital peanut, which accounts for the pre-flash, and it works perfect! Then I tried the RF trigger with a homebrew flash trigger, one that has a learning mode to determine when the main flash goes off...that worked too! It is only connected temporarily so I did not range test it, but it works across the room just fine. Now I know it can be done, I'm going to have fun building my own custom unit with a charge controller and all

Yes, that flash puts out a lot of HEAT!

 

[212]

I put the thing in the closet for a while because I needed more knowledge to make one myself. Now I have learned how to use serial to talk back and forth with some cheap RF units...thanks to you guys. I tried to figure out a way to have Picaxe ignore the cameras pre-flash, but had no success I think I need to have the code written to look at the rising edge to make it work, but I could not find what I needed using search. The camera first fires a pre-flash, then the main flash. I need to trigger on the second, but just going to a sub after the first flash is seen does not help at all...they are too close together to do it that way.

Can some kind person please steer me to looking at a rising edge please, I think I can take it from there.

 

[Dippy]

I'm confused. In the previous post you said you'd got it working????

So you have one pe-flash followed by the biggun? Is that right? I don't know how far apart these flashes are time-wise. I'm surprised you can't do it in software.

Speed issues aside, when you say "rising edge" do you mean the rising edge of the second flash i.e. main flash?

Couldn't you use a shift register (or similar) and tap off the second pulse?

Or maybe I've got the wrong end of the stick as I haven't read the thread.

 

[212]

Yes, one pre-flash followed by the biggun. Yes, I mean the rising edge of the second flash...if that is the correct term for what I need...newbie ya know

Sorry...I bought an RF flash trigger off ebay, and it is meant to be connected to a camera that has a connection for an external flash. The cameras I use do not have this. I can connect the RF trigger to a digital camera flash trigger, one meant to fire an external flash when it "sees" the flash from the camera go off. That works with the RF trigger I purchased.

I want to build my own version of the RF flash trigger. I want to have mine optically detect the flashes from a digital camera. It needs to ignore the first flash and fire the slave on the second. What I tried was to use an LED to detect the flash, which works fine. I used something like:

if pin3 = 1 then goto sub

sub:
if pin3 = 1 then
high 2.....or something along those lines

Well...I guess pin 3 is still high when it goes there because it still fires on the first flash. I may have the wrong idea about using a rising edge. My perception is...when sitting in "sub" Picaxe will wait for the flash to get bright again...after dimming some from the first flash.

 

[Dippy]

Well, first of all get yourself a fast sensor like a photo-transistor with a base terminal. I don't know how you have it hanging off your camera flash, but I would have thought that with a bit of fidding you could set an effective light threshold to completely ignore the first flash (pre-flash). You can bias the base to sort of adjust senitivity.

Obviously before you can do ANY code you've got to make sure your sensor is going on/off at the right moments. No point wasting time on code if your input is up the creek. Wth a simple photo-transisor circuit you could hook in an LED to check. So LED flashes when your camera flashes. Note with Photo-transistors saturate quick so background light will have to be cut out.
The best solution would be a photodiode (fast and tolerant) but they can be fiddly.

Say, I've just thought, can you pick up the flash signal inductively?

Code for synching up. I would have thought a couple of repeat/untils could be used. I was using a similar arrangement on a non-PICAXE project to synch up to PWM and it worked a treat. But code not relevant to PICAXE so I won't list it.

Question: Is big flash always and consistently x milliseconds after pre-flash? If so....

 

[Tom2000]

You might try a switch debouncing algorithm on your flash-detection LED:


When LED is high

...wait for LED to go low

...continue


HTH,

Tom

 

[eclectic]

212.

Perhaps I've missed it in the thread.

Why not just switch off the pre/red-eye flash?

Then there's just one light to deal with.

The DSC-P31 manual, pages 31/97 appears to give all the details.

e.

 

[hippy]

I used something like:

if pin3 = 1 then goto sub

sub:
if pin3 = 1 then
high 2.....or something along those lines

Well...I guess pin 3 is still high when it goes there because it still fires on the first flash.

I wouldn't disagree with that, if pin 3 is high when "sub" is jumped to there's a pretty good chance that pin 3 is still high when it gets to the if.

So, with your thinking cap on ... what would you need to add before the if to ensure the previous flash has finished before looking for the second flash ?

 

[212]

eclectic, thank you very much for the interest, but I'm afraid turning off the red-eye reduction still leaves the pre-flash used to set the exposure and white balance. Almost all point and shoot digitals have this

Thinking cap on...I like that

I believe the resistor in the diagram here is used to ensure the previous flash has finished before looking for the second flash ?

http://www.fancon.cz/slave-flash-trigger/slave-flash-en.html

I have built this, but had to buy a pre-programmed chip from someone to make it work... I do not want to steal his design though; I want to make one of my own with my Picaxe. I won't be selling it, but would like to be able to post the code once I get it working. Instead of firing the slave, I will be sending the trigger signal to a receiver connected to another Picaxe to fire the slave.

 

[Dippy]

Is "the resistor" = R1?

The thing above it is a photodiode not an LED.

My guess is that the cap is acting as a baby integrator and the programme gets a general level on the ADC so that ambient light level can be calculated then any sudden and significant signal on the ADC triggers the ouput. Thus he can 'subtract' daylight from the photo-D I/V produced across R1. Neat.

So you want to 'replicate' this, then send it via coded RF to a PICAXE to decode and then trigger a flash?
What do you envisage the total delay from sense-flash output to be doing this method?

 

[212]

It is neat, and it works well too, but it does use about 500ua I believe. A member on our homebrew trail camera site wrote his own code and got the current way down on his. I don't need the learn feature on mine, everything I will use it with is a two flash camera.

Yes R1 is what I believe lets this see the second flash as another flash. I did try my own with a photo diode too and had the same results as with the LED, but I will use the photo diode if it makes a difference.

I intend to send just something like:

main:serout 1,N2400,("UUUUUUTW" ... or whatever else works.

I don't know if it will be fast enough or not, but maybe

Here is a schematic for a trigger that does not use a Pic, but I'm having fun trying to learn programming. Jon is the one I heard say kludge before lol..

 

[212]

I'm trying to get others into this stuff too Here is what I put over on the trail cam forum:

A bunch of members, me included, bought some Picaxe chips. Picaxe is basically just like any pic chip, but the people offering it have pre-loaded it with their own code to make it more user friendly for beginners. Having gotten my feet wet with it, and having seen other versions of compilers and code, I have to give them credit...it is much easier for the first timer. I'm not going to push the Picaxe system, but I like it personally. Most people who have been programming for a while are going to use something other than Picaxe, and everyone is surely welcome here.

 

[moxhamj]

"I don't know if it will be fast enough or not, but maybe"

You can work it out. 2400baud means the line goes up and down 2400 times a second. A byte is 8 bits and then there is a start and stop bit, so to get the number of bytes per second, divide by 10. So that is 240 bytes per second. If you sent "UUUUUABCHi" that is 10 bytes so would take 1/24th of a second.

 

[212]

hmmmm...I checked the properties on several pictures from different cameras. All of the night time pictures from my Sony cameras say the exposure was 1/40 of a second, when taken with flash. I know the RF flash trigger I bought off ebay works with the Sony, as long as I use a digital slave trigger with it. I just figured maybe I could use a Picaxe to do the first flash ignore, then tell the receiver end to fire the slave too.

OK...I need to do some more stuff before I get too excited I guess. I'll first write some code that will fire a flash, timed to the shutter press on the camera. The P31 has a fixed focus setting, so it should be consistent. After I get that part, then I can try using my cheap little RF units, and see if the flash is still effective.

It's probably gonna go back in the closet for a while though, I'm playing with several other projects too...Thanks for the help so far, I'll probably ask again later.

 

[moxhamj]

Or course, if you are sending raw pulses rather than through the picaxe, the timing could be in nanoseconds. Say you send some data to a picaxe with the serin/serout. This primes the picaxe which sets an output pin high. Run that pin through an AND gate (74HC08) and run the other input from the radio module. Then send a pulse - it will go through the radio module, through the AND gate and the propogation delay through the whole system will be only a few nanoseconds. Then to turn it all off, send another data packet to make that output pin low.

 

[212]

I was typing still and did not see your post. So there is still hope


Oh...here is the code written by one of the trail cam site members for an all in one controller he designed. In there somewhere is how he tackled the pre-flash issue. This is why I'm here learning Picaxe instead of that gobbledy gook lol...

http://members.aol.com/trailcamcontrol/samsung60sec.asm

 

[moxhamj]

Looks like a perfect picaxe project! One picaxe and one 74HC08. The code will be much less in picaxe basic. I agree the machine code is gobbledy gook but the explanation at the beginning shows the framework of the code;
; Skips camera refresh, charge slave every 10 minutes if night time, Slave charge time 10 seconds (15 seconds right after picture taken), Test mode time 60 seconds, PIR warm up 30 seconds,
; If valid PIR signal then turn on camera, take a picture and then 10 seconds later turn off camera (and slave charging).
; Wait additional 1 second to allow camera time to shut off after camera is told to turn off after taking a picture
; Wait 60 seconds between picture taking

 

[212]

I did not put it in the closet, I just moved it to the corner of my desk. I'll put that 74HC08 on my next parts order.

I was wondering what the Picaxe used for code...basic huh...sounds more my speed

 

[Dippy]

Can you draw/sketch that one out DrAc?
I really cannot see how you can get (as you imply) nS timing here??

Many RF modules with no signal will chuck out noise on the Data Out. And the settling time for RF-in/Data Out can be milliseconds. Som RF modules faster than others agreed. But not nanoseconds.

 

[212]

OK....I popped the transmitter open and here is what is in there. I have not found a data sheet on the chip yet, but did find info on the unit itself, from people that use them. They say it is good for shutter settings up to 1/200 of a second, but beyond that they will not sync properly. There are a bunch of manufactures selling simular units too, and offer as many as 12 channels...mine has 4.

 

[moxhamj]

"Many RF modules with no signal will chuck out noise on the Data Out". It depends on the module. Hmm - maybe this might be hard. Maybe it might be milliseconds if the module does processing on the signal. Might have to home brew it. Though thinking of some of the 1-5 watt 2.4 Ghz video senders - these are sending megahertz signals so the propogation delay is going to be pretty short for those.

 

[212]

Working on something to test the RF with here, with parts on hand Before I can even try the RF part I need to be able to send the command at the right time. If I'm confusing you...sorry...I'm even more confused myself. I started a new topic asking about clearing W1 and W2 in a trial code I was messing with, and it got to discussing this project too.

Anyway, I have the LED lighting up when the picture is taken, but I am probably doing it wrong. Here is a picture and a drawing of what I have for a sensor to detect the flash from the camera.

 

[moxhamj]

Can you send a data packet using the RF modules yet? If they can send data then that means the signal is clean enough to send a signal for a flash. Will assume for the moment that there isn't too much RF noise in your area.

 

[212]

Oh yes sir, about the only noise around here is the tap...tap...tap of the electric fence, and that does not seem to bother anything. I have been playing with other RF projects also, and you got the data thing going for me

 

[moxhamj]

Sounds like a lot of the pieces are working. RF data link. Detecting a flash. I'm a little worried about the electric fence - say you put it in "ready to flash" mode and then it is tripped by a pulse from the electric fence. I don't know what those pulses look like on a scope? An electric fence is a bit of wire with lots of volts going up and down - in other words a rather effective antenna. And a square wave pulse will have harmonics. Hopefully that won't be a problem though.

Where is the project up to and what do you need help with?

 

[212]

Well...hmmmm...the first flash ignore triggering seems doable, but I need to play with it more to get it just right. So I can move on to the RF, I will go back to the idea of timing the flash trigger to the shutter press on the camera controller. I was going to try that pretty soon anyway. Then I might be able to use a cheapie camera that has no flash too. Once I get a scope I can see what it going on, but for now I think that's what I'll do. The shutter delay time will also give me a few more milliseconds to play with too if I need it. Thanks for asking sir, I'll have to answer that after I get my ducks in a row.

The fence will be off now, till next spring. We had only about 4 inches of rain this year, and the corn crop just got cut for hay

 

[moxhamj]

4 inches of rain. That is pretty marginal farming.

A scope certainly helps a lot.

I'm thinking of a circuit at the moment. You send a packet from one picaxe to another that says "get ready for a flash RF pulse". Then the receiving picaxe sets a pin high and that goes into an AND gate with the RF signal. Then an RF pulse arrives and the flash goes off. Then you want the circuit to rest for a few seconds to charge up the flash capacitor, so you want to ignore any more RF pulses. So the receiving picaxe also needs to know that the flash just went off. Then it goes back into waiting for another flash pulse to arrive.

But I see a small problem. How do you turn the receiving picaxe off so it ignores future flashes? If the receiving picaxe goes into Serin waiting for a data packet it can't do housekeeping like the delay after a flash. So maybe this might need two 08s or 08Ms? I'll think about this some more. Or maybe (shock, horror) a 555 solution might work. Do you need the current drain as low as possible?