|
impressive... i wish i had the skills to make these kind things.
Posted 46 months ago.
(permalink)
|
|
Its what I do for a living. This was a quick hack job. Much more fun than the projects that drag on for 6+ months. Now only if I can learn to take a decent portrait.
Posted 46 months ago.
(permalink)
|
|
Dude. You rock. Analog circuitry isn't my forte - so basically in your initial circuit the pulse causes the photo transistor to conduct, thus gating the BJT and that's what's shorting trigger(+) to ground, right? How does the added inductor "hold open" the circuit longer?
BTW, a lot of commercially available photo triggers are just simply potted inside polyester resin in ice cube trays.
Also, most photo transistors are sensitive in the IR range; you can stack rosco color gels to form a DIY IR bandpass filter, and therefore turn your flash into an IR only flash. This then allows you to take the pic without the triggering flash contribute to the picture.
Originally posted 46 months ago.
(permalink)
ttstam edited this topic 46 months ago.
|
|
Sweet!
Posted 46 months ago.
(permalink)
|
|
wow!
Posted 46 months ago.
(permalink)
|
|
|
the pulse causes the photo transistor to conduct, thus gating the BJT and that's what's shorting trigger(+) to ground, right?
How does the added inductor "hold open" the circuit longer?
Yes.
No the inductor does not hold the circuit on longer. Q3 and all the stuff above and to the right hold it on longer.
The photo transistor conducts depending on the amount of light. If the light does not change fast then the inductor just looks like a low value resistor. Only if there is a rapid increase in light will a significant voltage drop occur across the inductor. Once you have more than .6 volts on Q2 transistor's base the collector will be pulled to the negative side of the circuit.
Now once the collector gets pulled low this will cause the base of Q4 to go low and turn on. Because of this Q4 will feed the charge sitting on C1 to dump through Q4 through R3 and keep Q3 on until the charge on C1 drops.
So starting off with the light at a low level all the transistors are off. C1 gets charged up to the trigger voltage of the connected flash through R1. As long as the continuous light level is not excessive the positive terminal will not be dragged down by Q1.
When a flash goes off the sudden change causes a significant voltage across L1 which turns on Q2. Q2 then turns on Q4 which turns on Q3 turned on which keeps Q4 turned on which keeps Q3 turned on .... until there is no more charge on C1 to keep Q4 on.
Originally posted 46 months ago.
(permalink)
rudy__ edited this topic 46 months ago.
|
|
great stuff, looks like a fun project to build! :)
Posted 46 months ago.
(permalink)
|
|
|
BTW, a lot of commercially available photo triggers are just simply potted inside polyester resin in ice cube trays.
Yeah I thought of that but I didn't have any. Also I wanted to have control over the direction of the initiating light source.
Also, most photo transistors are sensitive in the IR range; you can stack rosco color gels to form a DIY IR bandpass filter, and therefore turn your flash into an IR only flash. This then allows you to take the pic without the triggering flash contribute to the picture.
I though of that and didn't bother since I wanted more sensitivity initially. I can add the gel later in front of the lens if I wanted to. This is one of the reasons I didn't want to go for a multi direction view.
Originally posted 46 months ago.
(permalink)
rudy__ edited this topic 46 months ago.
|
|
Nice one! :-D I've been looking for a batteryless optical trigger circuit for a while ... this one looks very neat and elegant ...
In your experimentation, how much light is "excessive" in terms of keeping Q1 turned on to the point that the flash either goes off or C1 never charges? I assume ambient room light is fine, does it also work outdoors (assuming direct line of sight between Q1 and the master flash)?
Posted 46 months ago.
(permalink)
|
|
|
great stuff, looks like a fun project to build! :)
It was simple enough and easy to just keep adding the parts wiring them point to point. No circuit board to make. No added wires, just the component leads. None of the parts values are critical. As long as they are in the ball park.
Originally posted 46 months ago.
(permalink)
rudy__ edited this topic 46 months ago.
|
|
|
how much light is "excessive" in terms of keeping Q1 turned on to the point that the flash either goes off or C1 never charges? I assume ambient room light is fine, does it also work outdoors (assuming direct line of sight between Q1 and the master flash)?
Well it depends on how sensitive you want it to be. The first two I built had a lower sensitivity photo-transistor. It worked well enough. But then I used a different type. Much more sensitive but also easier to saturate.
The more sensitive ones had a problem outside. I thought of some kind of light shield to cut back on the light intensity. Or I thought of a small slide switch to select a lower sensitivity photo-transistor. I have not used it much outside. The outdoor testing I did was in the middle of the day with a slightly overcast sky. Because the mosquitoes were biting I did not do extensive testing. I would like to do some more tests when I have time. I have access to a Lux meter and I would use some photographic measurements, F16 at 1/250s.
If Q1 saturates the trigger voltage gets shorted so it makes no difference if C1 never charges. The time constant is quite short.
My Canon 420EX has an open circuit trigger voltage of about 4 volts with an internal pullup resistor of about 50k. Works fine with that. Works with my 580EX. I also have a bunch of flashes that had high voltage triggers. I added an optocoupler circuit (MOC3023 I think, and C106D scr) powered by the flashes's internal battery. Source resistance is much lower than a typical flash. I think I used about 300 ohms in series with the optocoupler's LED when run off 6 volts. The optical slaves work fine with them. With the lower series resistance the photo-transistor should not saturate as easily as if it were connected to a higher internal resistance flash.
It works fine outdoors as long as it is pointed at the flash source or a reflection from the flash. Once again the more sensitive you make it the more likely it will saturate outdoors in bright light.
I plan to do dusk or dawn shots with multiple flashes. More architectural stuff. Colored gels. I have about a dozen flashes. Some I plan to use radio triggers. Any more I need I would use the optical slaves.
Originally posted 46 months ago.
(permalink)
rudy__ edited this topic 46 months ago.
|
|
Great post, nice circuit and great explanation. (Thanks for the schematic also). You rock.
Posted 46 months ago.
(permalink)
|
|
So your saying that these work with the EX series Canon flashes. That's pretty cool as most other optical slaves will lock up on those (the 580EXii is an exception).
Posted 46 months ago.
(permalink)
|
|
Sweet project!
Posted 46 months ago.
(permalink)
|
|
|
So your saying that these work with the EX series Canon flashes. That's pretty cool as most other optical slaves will lock up on those (the 580EXii is an exception).
I have only tried it on my 420EZ and my 580EX (not 580EX II). I don't see why this could not trigger other Canon flashes. If I can I will try it at a local camera shop on a 430EX.
I think the reason why some slaves lock up after the first firing is that these slaves probably use an SCR. With an SCR you can trigger higher tigger voltage flashes. What might be the problem is that the holding current could be keeping the SCR from turning off. This is a guess as I have not used one of these.
My circuit above has a latch like and SCR (Q3 and Q4) but to remain on it is dependent on the charge on C1. Once that disapears it shuts off.
Posted 46 months ago.
(permalink)
|
|
|
I once devised a way of getting an optical slave to work on a 430EX (I have this stuff already). I put a spst switch inline with the return path of the optical trigger to the flash. It worked, but the parts would have cost more than it was worth.
When the circuit is open, the voltage goes through to the open out side and powers the optical trigger. When the optical trigger gives the fire signal, the coil closes the curcuit and fires the flash, but it also interupts the circuit to the optical trigger essentially resetting it with every flash thats emitted. Since the power to the optical trigger gets cut via the spst then it releases the power to the coil, letting it rest back to open circiut status.
NOTE: I'm not that good with electronics and this may in fact be a spdt and not a spst. The spst that I got was from Radio Shack and would fry pretty quickly with a high trigger voltage. But I had it hacked together with wires everywhere and it worked, I just never finallized the project with a housing.
Sorry for the crappy pic, but its better than my hand drawings that I scanned.
Originally posted 46 months ago.
(permalink)
cabbiinc edited this topic 46 months ago.
|
|
Hi Rudy,
i'm trying out some optical slave trigger stuff as well.
I purchased some nice old METZ 45's. They have trigger voltages from 12 to 200 V, depending on the model.
So I tried some stuff with the parts I had flying around, being:
a Thyristor (SCR) and an IR-LED (in lack of a photo- diode or transistor).
[http://www.flickr.com/photos/dadita/3071519417/sizes/s/]
(the diode being an IR-LED)
[http://www.flickr.com/photos/dadita/3071568947/sizes/o/]
Well, I hooked them up and it kind of works... The flash is slave-triggered, when I power the master flash at 1/128 and hold it about 10cm away from the diode. That's not very sensitive, but hey, it's minimal.. It might work with a small cap between the diode and the Gate as well, to get rid of ambient light effects.
Question is:
Is it "allowed" to connect them this simple way? At least it shouldnt blow anything. And it might work better with a proper Photodiode.
Posted 43 months ago.
(permalink)
|
|
|
The problem with the IR-LED is that it is not designed to generate a voltage. You are supposed to power it and it is supposed to act as a light emitter. A LED can change light to a voltage but it is very inefficient as you found out.
SCRs are great to trigger higher voltages on older flashes. But when triggering an SCR gate you need a voltage. The easiest way to do this is to have some small solar cels to convert the light to voltage. You need two in series to get enough voltage to trigger the gate. I pulled some out of an old dead solar powered calculator I had.
So if you would put the solar cell in place of the LED you have listed. So now you have a light to voltage generator to trigger the SCR. The problem with this is that enough continuous light will also generate enough voltage to turn on and keep on the SCR. So what you also have to include is an inductor in parallel with the solar cell.
What this does is shorts out the solar cell for continuous light since the coil has relatively low resistance. But for a flash of light you get a sharp increase. This produces a sudden rise in the voltage. The inductor will look like a higher resistance to this sudden change and not cause the voltage to be shorted out. The short voltage pulse will cause the SCR to conduct triggering the flash.
I've done this and it works pretty good.
Where do you get an appropriate coil/inductor? For me, work. For you, it depends what you have available. What you could try first is a small relay coil of a small relay. If you have any small relays just solder the coil across and try it out. Not the best thing in the world but it allows you to experiment.
Some SCRs are better for this application than others. I just used some C106D since that is what I had.
Originally posted 43 months ago.
(permalink)
rudy__ edited this topic 43 months ago.
|
|
|
Hey, thanks for the fast reply! So I'll slaughter some old calcs. hehe. Well, i've been thinking about some way to power the trigger circuit from the flash. your circuit should do with high voltage strobes as well if you use a BF220 for the NPN and the according PNP, right? (The BF220 has a voltage rating of 250V)
Posted 43 months ago.
(permalink)
|
|
|
If you have high trigger voltage flashes I would stick with the solar cell and triac circuit. It is simpler.
While it is possible to adapt my circuit above it could get a little tricky with some of the component values that would have to change because of the higher voltage.
Posted 43 months ago.
(permalink)
|
|
"so basically in your initial circuit the pulse causes the photo transistor to conduct, thus gating the BJT and that's what's shorting trigger(+) to ground, right? How does the added inductor "hold open" the circuit longer?"
I'm going to remember that, as proof that I have no knowledge whatsoever in electrical matters.
Posted 43 months ago.
(permalink)
|
|
Hey Rudy, interesting way of doing it. This will surely be a great resource for DIYer's. Just to be flopping my mouth, I would love to have seen a Capacitor be used to differentiate the signal coming from Q1, that would take care of any ambient issues.
Gee, you almost made me start up my analyzing software :))
Posted 43 months ago.
(permalink)
|
|
|
Software?
I do it the old fashion way. When I went through school the software was real crude and cumbersome. I never did get into software simulation after that. The circuits I designed were not that complex that I couldn't handle with a good calculator. Most of the work I do is microprocessor based and not that complicated circuit wise.
gpphoto so basically in your initial circuit the pulse causes the photo transistor to conduct, thus gating the BJT and that's what's shorting trigger(+) to ground, right? How does the added inductor "hold open" the circuit longer?"
Right, on the first question. Wrong on the second. The inductor shorts out the base of the transistor so it does not respond to continuous light.
The inductor looks like a higher resistance at higher frequencies and fast changing waveforms. The inductor drops a higher voltage across itself and this causes Q2 to conduct. Now if the rise were long enough we would not need anymore than Q2. But the rise time for the rising edge of the flash pulse can be really short. While you would get a pulse from the + terminal it may not be long enough to trigger the flash.
This is where Q3, Q4, and C1 come in. They get triggered by that short pulse and "latch" on until the charge on C1 is used up. In effect That part of the circuit is doing what an SCR does.
But an SCR can remain latched on if the current through it doesn't drop low enough to extinguish itself. This may be why some optical triggers will trigger certain flashes once and never a second time. Then you have to turn the power off on the flash and then back on and it is ok for one shot again.
Like I said, most of my circuit is just acting like an SCR. Once triggered it would stay on indefinitely but since it relies on C1's charge it has a limited time to stay on and it will reset itself.
So once again, the inductor allows us to see only sudden changes in light (typically only the rising edge of the light from a flash). This is too short to trigger a flash and we extend the pulse with the right half of the circuit.
Now there actually is a flaw in this circuit. If the continuous light is high enough Q1 by itself will trigger the flash. This is where you have to balance maximum sensitivity with operation in bright ambient light.
On my original circuit at the top of the page the 27k resistor would prevent the photo transistor from triggering the flash with high ambient light (the 82K resistor also).
I though of including a series resistor and capacitor as in the collector of the photo transistor in the "new" circuit but while that would get rid of the ambient issues it would also reduce the potential maximum dynamic range. I won't go into an explanation on this.
In the end I had to compromise on a little more range or total ambient light rejection. I went one way but I'm still not sure that was the best choice. Maybe I'll try the other option at some point and compare the two.
Posted 43 months ago.
(permalink)
|
|
|
gpphoto I see where you got the "hold open" confusion from. Actually ttstam said it. I replied to two of his questions and it may look like I answered yes to the hold open question where the yes was for the previous question followed by No for the "hold open" question. I also explained how didn't hold open the circuit.
Posted 43 months ago.
(permalink)
|
|
|
Rudy, I'm old enough to have learned electrical engineering from Medusa, I still used slide rulers when I went through college :) I used to do some pretty complex analysis on rather large circuits and doing that without software would have taken some significant time.
I do agree that your circuit is easily analyzed without software unless you want to analyze for noise, stray caps and other such nuisance, which would certainly be overkill in that application.
Posted 43 months ago.
(permalink)
|
|
|
I actually would love to have an analysis done on this but I don't have a program to do it.
In the mid 80's I made a development board for an IBM PC. The board had a Z80 microprocessor on it. The processor lines were brought out and plugged into our target hardware of the product we produced. On this interface board I had bus arbitration hardware that allowed the IBM PC processor to sneak in and view the memory contents of the Z80's memory without interupting or delaying the Z80. With this we were able to see the contents of the z80's memory space in real time. It was a great tool.
A company who sold analysis tools (PC based) came to town and offered to demo me their software. I went to the meeting and I brought a copy of the schematic of the bus arbitration circuit I used. After they made their pitch and showed me their program I pulled out my schematic (all digital logic) and asked to put in a real circuit.
They agreed and we spent some time putting it in. Then we let the software do its work. It showed a potential for a glitch to cause extra delay on the PC side (but it made no real difference in the application).
While I can't remember exactly the detail but whatever the software came up with (and it was correct) the two people doing the demo didn't know it was capable of doing what it did.
Anyway I was impressed with the software. We never did buy any software since it was quite expensive. The company I worked for was real small and we couldn't justify the expense.
Now I work for a somewhat bigger small company (happily) and still don't have any analysis tools. Maybe I should check to see what is currently available. We bought out another electronic manufacture last year and it came with a young electrical engineer. I could always come up with the circuit design and he can learn the program and do the analysis for me. :)
Currently I have come up with the electronic design on paper like in the first schematic at the top of the page. I pass that off to one of my other guys and he does the schematic capture. The above schematic I threw together with Photoshop.
Originally posted 43 months ago.
(permalink)
rudy__ edited this topic 43 months ago.
|
|
i have a broken Sigma 500 super. Do you think there is a way to use its optical slave system and mod it to externally fit an sb-600?
Posted 43 months ago.
(permalink)
|
|
Man, it's stuff like this makes me wish I took electrical engineering in college... (My EE friends always made the coolest gadgets)...
good stuff!
Posted 43 months ago.
(permalink)
|
|
|
Rudy, I did run an quick transitional analasys on it but can't remember the exact data. I will post or send you a plot of the turn on speed vs. input signal. I will not be able to send you this data until tomorrow, I got all my electrical stuff on my home computer. It's pretty fast, but I'm not absolutely positive about the right side of the diagram, need to do more looking around :) It looks it could cause trouble with some flashes depending on the impedance of the flash trigger circuit. Did you notice if the circuit locks up? I have not done SPICE for quite some time so I could have done someting wrong, I'll play some more and see.
P.S I will also send you a link to the SPICE program I use, it's free for personal use with some restrictions and only costs $175 for the full version, it's quite impressive. In the old days I used ORCAD but at almost $10000 I have no use for it anymore.
Originally posted 43 months ago.
(permalink)
Alfredk edited this topic 43 months ago.
|
|
|
Here is the link to the SPICE program.
www.5spice.com/
I would not attempt to analize the space shuttle with it, but it works great on most things.
Posted 43 months ago.
(permalink)
|
|
|
It looks it could cause trouble with some flashes depending on the impedance of the flash trigger circuit. Did you notice if the circuit locks up?
I convertered a few flashes with an optotriac/scr circuit. The supply was the battery voltage and there was a series resistance of 680 ohms in series with the LED in the optotriac. I had no problems with lockup on that. On the other end I have a Canon 420EZ that has an open circuit voltage of 4 volts and a series pull up resistance that is higher than 5K (could even be 22k but I can't remember) and I had no lockups with it or any other flash I used with it..
Other than the high ambient light issue it has been very reliable. And with the high ambient light situation I figured that if there was that much light to saturate the phototransistor I wanted to know, and the light being triggered indicated that.
Thanks for the link. I'm real interested in trying it out.
Posted 43 months ago.
(permalink)
|
|
|
i have a broken Sigma 500 super. Do you think there is a way to use its optical slave system and mod it to externally fit an sb-600? It depends on how it is made. Being a newer design it could use surface mount components. Even if it were through hole components the circuit would likely be integrated on a more complex board.
The work to extract the circuit (if it can be used) could be as much work or more to fix it.
Posted 43 months ago.
(permalink)
|
|
Well done. Congrats on this project.
Also, you could connect one of those optical slaves to the transmitter of your radio triggers. Then just have that duo somewhere where the flash from your A570 compact can see it, and voila, radio triggering flexibility and distance for your compact camera.
Posted 42 months ago.
(permalink)
|
|
|
jeffrey_bloch Also, you could connect one of those optical slaves to the transmitter of your radio triggers.
Yup, done that.
Actually that is quite handy sometimes since a P&S camera only directs light forward. If I want my flashes off to the side the camera's flash may not be able to trigger the flash. Using the optical slave to trigger a radio trigger overcomes this limitation. Since all my cameras, rf triggers, and optical triggers have a 1/8" mini connector this is a piece of cake.
Posted 42 months ago.
(permalink)
|
|
|
Rudy did you try the Spice simulation yet? I have some real concerns about the right hand side of your circuit, the delay part, it does not have enough juice to turn Q3 on in my simulation.
Posted 42 months ago.
(permalink)
|
|
|
I just quit. I was unable to find a spice model for a photo-transistor. Since I had never used the program before I was just winging it. Not bad until I was looking for models and couldn't find them. Also for some reason I couldn't open the doc file on the program. Might be my copy of word (came with the machine and I have not used it yet).
As far as the Q3 turn on, it sure does work. I've tested it quite a bit and verified it with a digital scope. That was how I chose the capacitor value.
At first I didn't have the Q3/Q4 circuit. The trigger would work if I triggered it with a Vivitar 283 but not my Canon A570. The photo risetime of the 283 was long enough for Q2 to turn on and trigger a flash. With the A570 the light output jumps to a maximum imediately and that time was not long enough for Q2 to trigger a flash.
Then I added the Q3 & Q4 circuit. After Q2 collector went low this turns on Q4 and then kept Q3 on until the charge on C1 drops to low.
Like I said, it works and I have observed it working with a scope under different conditions. I don't know how to get a digital image out of the scope. I'll ask one of the younger guys. It might be a couple of days before I could test this again though.
Posted 42 months ago.
(permalink)
|
|
|
Alfred Just thinking about it some more. I wonder if the capacitor for C1 is really .0022uF or if it is .01uF.
edit No, just checked (with reading glasses) it is .0022uF
Originally posted 42 months ago.
(permalink)
rudy__ edited this topic 42 months ago.
|
|
|
You can find models by Googeling for the part number followed by spice like "2n3904 spice".
I never bothered to use an actual photo transistor, I used a regular pnp driven by a square wave source.
As I said before I have not used Spice for some time and it will probably take a few trials to get the hang of it again.
I'm actually having fun with it, so I will try to test that right side some more. I assume that there is some peculiarity in Spice which does not allow for a capacitor to be considered as a power source.
What is funny is that it is always the small and simple circuits which throw this programs off :)
Posted 42 months ago.
(permalink)
|
|
|
I took some scope waveforms for the above circuit. Since they were big, and of limited interest, I decided to put it on a web page. I also included much of the information posted here. I have more work to do to clean up the page but for anyone interested in the waveforms you can see then on my DIY Optical Slave trigger page.
Posted 42 months ago.
(permalink)
|
|
|
Hi Rudy and the others,
hope you don't mind me posting weird schematics again :o)
I guess you remember my goal of designing a very simple yet powerful optical trigger.
Here it is:
The photodiode can be any silicon type, the potentiometer is not necessary, only if you like to decrease sensitivity.
And it IS sensitive. It triggers my Metz 45, when I fire the master at minimum power (1/128) through a half-closed door! The diode "looks" to the ceiling, in a mostly white room, about 6m away.
In other words: my commercial (HAMA) trigger fires with an IR-remote in ~20cm distance.
This circuit here triggers in ~1.5m distance with the same remote!
Nevertheless, placed underneath a 36W Neon tube (50Hz) doesnt trigger the flash.
How does it work?
The large reverse voltage of maximum 215V (for the Metz) puts the diode in kind of an "avalanche mode". That means the PIN diode works a little like an avalanche diode. The large resistor converts the current into a fairly high voltage. Together with the sensitive SCR (threshold voltage of ~0.8V) this yields a super simple and sensitive optical trigger.
It even works with a hot-shoe voltage of down to 3.5V! (I have no strobe with less than that)
Costs about 1€.
Comments welcome!
Posted 41 months ago.
(permalink)
|
|
For that point and shoot, take some old slide film (a slide that was taken with the lens cap on then developed) and put the developed slide film over the pop up flash.
The slide film has will no allow visible light to pass through but will allow infrared light to pass through. Thus your optical slaves will be triggered but no light will be cast upon your subject.
Posted 41 months ago.
(permalink)
|
|
|
dadita_1603 Yes that is a good standard circuit for a trigger.
The part that I don't like is that with certain loads the SCR can turn on and latch on and never turn off. This is a problem that has commonly been seen with Canon flashes. To reset the SCR you have to turn the flash off then back on of the next shot.
One thing you may want to test is having the flash in a bright room or lit by a bright light and then try to trigger it with a remote flash. Sensitivity is great but if your sensing device (photo diode, photo transistor) saturates it will not trigger. Sure it will not give a false trigger (because of C1) but it may not trigger because the voltage at the Anode of D1 can't go any higher.
Posted 41 months ago.
(permalink)
|
|
|
Rudy, i needed quite some time to figure out that circuit :o). Actually, I found no good passive optical trigger on the net (apart from yours). They all use fancy microcontrollers and stuff (which will be my next project), but none is simple enough and works from low voltage strobes up to high voltages.
Concerning bright ambient light: unfortunately, i haven't had a chance so far to test it under sun light, as the sun is quite rare these days ;o) But as i said, placed underneath a strong fluorescent tube doesnt trigger the flash and it still works from the IR- remote. In fact, the photodiode should not saturate even when placed outside on a bright sunny day (when not in direct sunlight), because the photo current is more or less linear over seven decades or so of illumination.
The problem i see is that the photocurrent might bee too high for the flash to maintain the trigger voltage on the hot shoe. After all, the hot shoe can source only a few microamps.
Nevertheless, such a simple design, it's a feasible result.
Concerning the latch-up: That's quite interesting, I have one single flash that doesnt work twice with the commercial trigger nor my DIY one. I always have to disconnect the trigger and then reconnect it.
It's a Carena Speedlight I bought for 1€ on ebay :o)
I guess the reason is that the hot shoe voltage does not fall to zero long enough when triggered. Is that right?
Posted 41 months ago.
(permalink)
|
|
|
The problem i see is that the photocurrent might bee too high for the flash to maintain the trigger voltage on the hot shoe. After all, the hot shoe can source only a few microamps.
When you scavange power like this you have to make some compromises. In order to be able to handle a higher ambient range I decided to go for the inductor rather than the series blocking capacitor as in your and my first circuit at the top of the page.
I guess the reason is that the hot shoe voltage does not fall to zero long enough when triggered. Is that right?
No. It is the current that keeps the SCR turned on not the voltage across it. This is where selecting the right SCR is important. The problem is that the "one size fits all" does not always work. This is why there has been a lot of frustration with commercial optical triggers.
Posted 41 months ago.
(permalink)
|
|
I am very eager to put these into use, but I am not sure which phototransistor to use for Q1. Can somebody say which component or Radio Shack part number to use? Also, how does one incorporate the SCR and the OptoCoupler into the circuit? I don't know a whole lot about electronics, but can follow schematics and solder them to make usable circuits. I have a few older flashes that could use these.
Thanks!
Posted 39 months ago.
(permalink)
|
|
|
Unless you have a convenient cheap source of parts you may be better off buying ready made optical triggers. They are not that expensive.
www.flashzebra.com/opticalslaves/index.shtml
Flash Zebra has this one that fits on the flash.
They have others that can be used with trigger inputs as well. Also not expensive. When you add up all the costs you may not actually save anything if you have to buy parts from places like Radio Shack.
But if you want to make your own, and you need to trigger a flash with a higher trigger voltage, then look at photocell-SCR based design. My circuits above are for low voltage triggering only.
As far as what photo-transistor to use. I think any would work. But they all may be too sensitive. You may have to restrict the amount of light they see. I have had so success taking them out of old flashes. Some auto thyristor controlled flashes have a sensor to cut off the flash. Since I replace that part with a variable resistor to get power control, I end up using the part in the optical trigger circuit.
Originally posted 39 months ago.
(permalink)
rudy__ edited this topic 39 months ago.
|
|
im thinking of making this optical slave, so I'm wondering what's the value of Q1 (as I have these available in the store: BPW 34, BPW 40, 41, 43, 50, 76) and is C1 really ,0022uF or is that 0,22 uF. Couldn't really find 0,0022uF in my local electronic store. Thank you very much
EDIT: Now I see that you have comfirmed that it actually is 0,0022uF...that would be 2,2 nF right? the best I can find is 1,8nF, would that work.
tnx a lot
Originally posted 37 months ago.
(permalink)
Winnito edited this topic 37 months ago.
|
|
|
1.8nF should be fine. I picked 2.2nF because that was all I had in that range.
As far as Q1 that was a salvaged photo transistor from opto-interrupters used in old floppy disk drives. You list a photo diode. They are more expensive. I don't know how it would work. Might be OK.
Before you build this add up what it would cost first. Often you can buy complete electronic devices for less than the cost of building them yourself. And then you have to include the time involved. But if you have a burning desire to do it then go for it.
Posted 37 months ago.
(permalink)
|
|
Hi Rudy, i bought the slave hot shoe triggers on ebay but they get stuck after one flash, they are canon 430ex... Any idea on a diy fix??? Have you ever heard of this problem?
Thanks
Posted 37 months ago.
(permalink)
|
|
|
Most likely the part they used to act as the switch was a SCR. This part is designed to turn on and stay on until the current stops. So as long as there is current it stays on indefinitely.
If you put a resistor and capacitor in line with the center pin contact it should work. The values are not critical. I would try a 0.1 uF capacitor (10 - 50 volts. anything higher will be bigger but will work) and a 47k resistor. If you can open the slave you might be able to put the parts inside and connect to the center contact.
I don't show the side contact line in the diagram. That connection stays the same.
Posted 37 months ago.
(permalink)
|
|
Hi Rudy, Thanks for the detailed descriptions on the optical trigger. Very nice.
You said you modded your high voltage flashes to 6v or less. Do you have a circuit diagram for reducing the voltage? I've seen one that used a zener diode but I'm not sure about that one. Is there something better to use?
Posted 36 months ago.
(permalink)
|
|
I just browsed through this post and went like
:O
reminds me about my electrical machinery class... that I barely passed haha
Posted 36 months ago.
(permalink)
|
|
|
HeliGeek I don't have a schematic on the net so I found the following. www.carlmcmillan.com/Optoisolated_Adapter.htm
I used a MOC3023 opto-isolator as that is what I had and it requires half the LED current. Because of this I used a 620 ohm resistor rather than the 330 ohm for R1. I didn't bother with R2 as I didn't think it was required. The SCR I used was a C106D. It is smaller.
I make two different physical configurations depending on what I have for room (shown below). I cut the top off the SCR in order to reduce the size. The SCR part is in the bottom of the package and the top is only for mounting and heat dissipation and not required here.
Posted 36 months ago.
(permalink)
|
|
|
mmm that solution sems like i will loose stability on the swivel if i have an extra someting between the swivel and the flash... thanks anyway!
I sold a 430EX to get 2x SB-28... they will arrive soon and i should get better reception with the poverty wizards untill i get cybersyncs or radiopoppers
but thanks
Posted 36 months ago.
(permalink)
|
|
|
MAX www.barrosinc.com If you don't need an on camera TTL flash then you are better off with both SB-28s. The more quality flashes you have the better.
Posted 36 months ago.
(permalink)
|
|
|
im super stoked i got them because i thought the same would be better... but now that the SB work way better with the poverty wizards and i have 3 flashes is a cool deal!
so im happy! I even have a couple of light slave triggers that will work with the SB28!
nikon flash system pwns canons flash system
Posted 36 months ago.
(permalink)
|
|
|
I'll review this and find some parts. Thanks for the link.
Posted 36 months ago.
(permalink)
|
|
rudy__,
Thanks for your optical slave circuit. I tried it on a breadboard and it seems to work. I said 'seems to work'; I need to confirm a few things from you if you will. First, where I stay in Nigeria I cannot easily find inductors to buy. I then went to our commercial city, Lagos. The highest value I go was 1mH. So I ended up borrowing what looks like a bigger inductor from an old pcb, no way of knowing its value. I can observe a delay in the triggering effect. Any substitute component in lieu of inductor? Second issue: I tried a Sony DSC W300 camera with your design (in fact with other similar circuits); although I can sense some delay in triggering, the camera does not record properly exposed pictures. There appears to be a lack of synchronization. Could this be a result of a different pattern of pre-flashes in the camera design? If you know, how many preflashes does this model emit before image recording? I will appreciate every assistance. Meanwhile I will look for other point and shoots to try on your design. Thanks a lot for sharing the circuit.
Kunle
Posted 15 months ago.
(permalink)
|
|
|
This circuit was not meant to deal with pre-flash. There is no delay before turn on but a small delay before turn off.
10mH would be desirable but 1mH will also work with less sensitivity. Inductors can be hard to get. One possible source is to use a transformer in a one time use (disposable) film camera. The transformer used to generate the high voltage for the flash capacitor will work OK. Use the highest resistance winding.
Posted 15 months ago.
(permalink)
|
|
|
rudy__,
Again, many thanks for such a quick reply. You seem to have lost me on your circuit principle. Why am I not getting correctly exposed images on my dsc w300 point and shoot? If I understand you, it should not matter how many preflashes my camera emits? I need more help, please.
Posted 15 months ago.
(permalink)
|
|
I found your circuit after searching for a non battery operated dumb slave, I was curious if I could substitute a 2n3906 for q4 or if there was a reason you picked the 2907a. I have a ton of the 3904s and 3906s in my junk box, would love to build it for free (which I have a feeling is what you have done) thanks caleb
Posted 12 months ago.
(permalink)
|
|
too complicated..
Posted 9 months ago.
(permalink)
|
|
I know it is an old topic, but let me add to it as well:
I ahve used the so called Carl Vogt slave trigger with much satisfaction. It really is the easiest trigger to build:
The top two I have actually built and used. The top one uses a small solarcell from an old pocket calculator (the ones that can be charged by sunlight). It functions pretty good even without the 1Mohm resistor.
If one does not have access to a solarcell, that can be replaced by 1 or more photodiodes in 'conductive' mode. The way they are used actually makes them generate a voltage. In fact, I found that 1 photodiode already works well enough, but two just gives a bit more sensitivity.
In terms of surface area, I'd say the photodiodes are a bit better than the solarcell.
I have used them on a flash with 6 volts over the hotshoe and on a flash with 250 volts over the hotshoe. Worked fine in both. I can trigger from several meters away. Inside, ambient light is not a problem. Have not used them outside.
The advantage of both is that the have minimal drain on the sync contact, coz there is no extra circuit to feed
The 3rd one I have not used myself, but it also was an original design of Carl Vogt.
Just to make clear, none of these designs works with a preflash, well, they work, just a bit too early.
Posted 7 months ago.
(permalink)
|
|
All very interesting. I gave up electronics as a hobby in 1977 just before I took up serious photography. The reason being that it was getting harder and more expensive to buy components and I had to travel to London to have a real choice back then.
I recently bought a small mixer for audio because I can't really rely on in camera microphones if I am to do serious video work. I used to make these things but there is no way that I could have built it for the same price as the Chinese do.
One of the reasons for this is that the sockets and knobs on their own bought separately here would cost more than the whole unit.
Originally posted 7 months ago.
(permalink)
Roger Blackwell edited this topic 7 months ago.
|
|
|
@Ed_X Yes those are simpler, and I have used them as well, but they don't always work. Some Canon flashes can cause the SCR in the circuit to stay latched on. The optical slave will trigger the flash the first time and then you need to turn the power off on the flash to remove the current through the SCR so they will turn off. Not really a good solution. The circuit I showed latches on initially but will automatically reset.
Posted 7 months ago.
(permalink)
|
|
|
I understand Rudy. It was no criticism on yr schedules, just expanding on the subject. Till now they have served me well and they may help some others. I do not have a Canon flash though :-)
Nevertheless, I kept yr schedule as I may have need for it myself one day :-)
With regard to the Canon flashes, I have tested circuits that would work well with an SCR except for the Canon flash, but replacing the thyristor with a transistor such as 2N6517 or the BD series, interestingly enough did work
Originally posted 7 months ago.
(permalink)
Ed_X edited this topic 7 months ago.
|
|
|
@Roger, you are right. Same happened to me. Self built is hardly worth it anymore, I just do it for small projects or if it is something that I can't find. Parts stores have almost gone and eventhough the elctronic parts might not be expensive, the paraphernalia are (a jack, a switch, a plug) and then you got to be creative to keep the housing cheap :-)
Posted 7 months ago.
(permalink)
|
|
|
Using a transistor does work, sometimes. That is what I had in my initial circuit at the top of the page. Unfortunately this caused a very short pulse when triggered by my P&S camera (that has a very sharp rise time) and the short pulse didn't trigger the flash. That's why I had to latch the signal.
The circuit I made was a solution to a specific problem. There are a number of other designs that are simpler and are great if they work with what you need to use them with. I would have prefered simpler but it didn't work for me.
When I first developed this circuit there were a number of people that had bought commercially available optical slave only to find they didn't work with their flashes. Since that time some of the optical slave manufactures have come up with EX friendly models.
Posted 7 months ago.
(permalink)
|
|
|
"@Roger, you are right. Same happened to me. Self built is hardly worth it anymore, I just do it for small projects or if it is something that I can't find. Parts stores have almost gone and eventhough the elctronic parts might not be expensive, the paraphernalia are (a jack, a switch, a plug) and then you got to be creative to keep the housing cheap :-)"
I work for an electronic manufacture doing hardware design. Even with my sources it is rare that a DIY project will come out costing less than a commercially available product. You have to do it for other reasons.
Posted 7 months ago.
(permalink)
|
|
FYI, YN560 also latches SCR
Posted 7 months ago.
(permalink)
|
|
I know this is an old topic here, but I was stimulated by this to find my solution.
Here I show you what I use with a Canon A620 point-and-shoot camera, with its flash at minimum power and a Canon 580EX as optical slave.
It works, without the persistent latching effect !!!
SCR can be even a TIC106 (I guess).
Originally posted 6 months ago.
(permalink)
Donato Greco edited this topic 6 months ago.
|
Would you like to comment?
Sign up for a free account, or sign in (if you're already a member).
|
Strobist.com /
Discuss
Feed – Subscribe to Strobist.com discussion threads
