Optical triggers and preflash

[Lello]

Found this Thought it looked quite interesting.
http://mysite.verizon.net/jayandriot/Loop.htm

[robski]

Just to remind folk that my background like Foxy Bob is in electronics. I was an electronics engineer for 25 years and basically lost interest 10 years ago when the majority of equipment became uneconomic to repair. By then most designs were a one chip solution and the manufactures were not going to release any information on its inner workings.

To start with I’d thought I would give you a layman’s guide to the workings of an auto thyristor flashgun. As a word of warning I strongly recommend that you don’t mess with these devices unless you have a healthy respect of high voltages with a large current (250 – 400 Volts DC generated from those tiny AA batteries)

The Bulb is a device called a Xenon tube. It has 3 connections. Applied to 2 of them is the High Voltage (ranging from 250V to 400V dependant on the power rating of the Xenon tube used). This high voltage charge is generated by a voltage converter circuit (the thing you hear whistling) and is stored in a Capacitor.

The camera uses a switch to fire the flash. When the switch operates it sends the trigger Voltage through a step-up transformer to generate high voltage pulse (4KV to 10KV dependant to the Xenon tube used). This high voltage pulse is applied to an exciter electrode on the Xenon tube to ionize the xenon gas inside the tube. The xenon gas suddenly becomes a low resistance and the energy storage capacitor discharges through the tube resulting in a short duration brilliant white light.

Old cameras used a mechanical switch to fire the flash. On flashguns from this era the trigger voltage was tapped off the high voltage supply and was in the order of 100 – 200 Volts. This meant a small step-up transformer could be used to generate the exciter pulse. Flashguns designed for electronic cameras have a much lower trigger voltage (5 – 15 Volts) and use an opto-coupler (electrically isolate) to connect to the former exciter pulse generating circuit.

OK that’s what a cheap and cheerful flash unit does. Then some bright spark came up with away of controlling the duration of the flash with a device called a thyristor. Basically a thyristor is a special kind of electronic switch used to interrupt the current flow through the tube and thereby turn the flash off. On the old auto type flashguns a sensor fitted to the gun provided the turn off signal to the thyristor. On modern TTL metering systems the camera provides the turn off signal when enough exposure has been received.

The above gives us the understanding that the brightness of the flash is not controlled but the duration.

Now we come around to the optical slave trigger device. It is fitted with a sensor sensitive to light. Most appear to use a Cadmium Sulfide photocells (CsS) used in the old cheap and cheerful handheld light meters. There is a whole bunch of theory on light sensors which is important in Gigabit fiber optic links but I won’t bore you with it here. The key thing about this theory is a big sensor has a strong output signal but is less sensitive to very short pulses and the converse is true for a small sensor.
I think the pre-flash is still long enough for the sensor to respond to it well enough to fire the slave.

Now to look at slave flash units that copes with pre-flash. In Saphire’s thread on wedding photography she has such a device. I think Metz also has a unit but it is expensive. These units expect and make use of the pre-flash. In a nutshell the electronics uses the first short flash to generate an enabling signal which is held for a period of say 0.5 second. By which time the main flash has occurred and because its path has been enabled by the pre-flash it goes through to trigger the slave.

[Don Hoey]

Rob,

Burning the midnight oil again I see.

Excellent write up on the workings of a flash unit. You are the electronics boffin, so would not disagree with anything you have said.

Your mention of Cadmium Sulfide photocells (CsS) in the trigger is interesting. My trigger of unknown origin dates from the mid 80's so I would not expect the light sensor to be very sensitive. My tests involed every permutation I could think of. Pop up - big flash - distance from trigger, all to no avail

Camera light meter cells went through the route from CDS to SPD with its faster response and greater sensitivity to light, so one might have expected cells used in flash triggers to have gone a similar route. My thought was that if the pre flash is not as bright as the main flash then possibly we could make use of the inverse square law to limit the light intensity the sensor recieves. Hence the black flags.

Graphic attatched.

May work, may not.

Don

[Don Hoey]

Originally Posted by Lello

Found this Thought it looked quite interesting.
http://mysite.verizon.net/jayandriot/Loop.htm

Excellent read Lello.

Don

[robski]

Don

It looks like I may have get my Oscilloscope out of the loft and rig up a sensor circuit to test this out. The Boss has jobs for me at the weekend so it may take awhile.

Another thing to think about while on the subject. If the brightness of the flash was to change then also would its colour. The brightness is very much down to the voltage on the capacitor at the time of discharge. As the batteries age then the Voltage would be a bit lower.

I see on the recent models of Canon flash 430EX & 580EX they attempt to compensate for this by telling the camera what colour temperature to set the camera to.

[Don Hoey]

Originally Posted by robski

It looks like I may have get my Oscilloscope out of the loft and rig up a sensor circuit to test this out.

Nothing like getting a techie involved.

Interesting thought on colour temp. Must be something in it if Canon recognise a problem. I will have a look when the workshop is freed up, athough I do not have a colour temp meter, so a white card job is the best I will manage.

Don

[robski]

It does appear you can get digital versions of the optical slave triggers. (peanut) A tad more expensive.

"WEIN products has duplicated the entire line of its popular solid state batteryless slaves in a digital format that can perfectly synchronize with digital preflash cameras."

The Wein PN-XLD for example.

This talks about coping with a 2 preflash system used on Nikon. From what I recall from my past reseach is that Canon has 1 preflash. So I don't know if they will cope with both or only Nikon. Info on this product is a one liner and a bit vague.

[Canis Vulpes]

I think the spirit of this thread is to provide studio quality results using bare minimum equipment and lots of brain power.

I have been reading with interest some attempts to use i-TTL or equivalent with optical slave triggers. My initial experiments confirmed that preflash triggers the optical triggers leaving the test shot very much underexposed. A result found above by others.

I thought of a few things to solve the situation by inhibiting preflash from triggering the slaves. I wont bore with the ideas but the final solution was simple kitchen roll used to attenuate light picked up by the optical triggers. I found that a single layer laid across my triggers did the trick. Attenuating enough of preflash not to be effective but allowing main flash to trigger the slaves. Of course a positive result would be a whited out exposure becuase i-TTL could not predict the output of the slaves.

Photos below....

Pic 1 wrapped double layer of kitchen roll, this proved too much attenuation as main flash did not trigger, I knew I was on to a winner.

Pic 2 Jackpot! using a single layer of kitchen roll placed over the optical triggers

Pic 3 as pic 2 but slave flashes switched off.