How to reset SCR used for triggering camera flash

First some quick background.

For studio photography, it is typical to use several flash units to illuminate the subject. One of the ways to syncronize all of the flash units is by using an optical trigger also known as an optical slave trigger. The setup works like this - you have one flash that is triggered by the camera - all other flashes are attached to an optical slave trigger so that they will go off in unison with the master (camera triggered) flash. This optical slave trigger simply shorts two leads from the flash when the flash of the master light is detected.

Now the problem/question.

Most of these inexpensive slave triggers reportedly use an SCR (silicon controlled rectifier) that acts as the switch or trigger. Apparently from what I've learned, this SCR must have the current brought to zero in order for it to reset.

This is not a problem for most flash units on the market. But, the top of the line Canon Flash units will not function properly with a standard optical slave trigger. It will flash once, but after that it cannot reset. My guess is that there is some residual current flowing that does not allow the SCR to reset.

My question - is there a circuit I can place between the flash unit and the optical slave trigger that will bring the current to zero long enough so that the SCR can reset without interfering with the general operation of the slave trigger and flash?

The flash circuit is a relatively simple DC circuit - just 2 leads from the flash to the optical slave trigger. When these two leads are shorted together, the flash goes off. I'm not really sure exactly how either the flash or the optical trigger works, but the theory seems simple enough.

There was an individual that I found online that reported that he had placed an RC circuit in the system, but my attempts to duplicate his efforts have been inconsistent and I am unable to contact him.

Any help would be greatly appreciated!

Please let me know if you need any additional information. I have a multimeter and can take some measurements as necessary.

Thank you,
shootfirst

You are right, once a SCR is triggered, it will stay on unless current is stopped from flowing through it.

Unfortunately I can't think of a circuit though :cry:

:shock: I must say, this is the best post I have ever read, so much detail, and complete sentences too.

Unfortunitly, I dont really know how to help. I know alittle bit about strobes from my failed strobe light project, but not enought to be able to suggest anything that I would be certain of.

Only thing I can think of is that maybe the slave trigger is picking up ambient light which is causing some current flow and preventing the SCR to reset?

jrz126 said:

:shock: I must say, this is the best post I have ever read, so much detail, and complete sentences too.

Click to expand...

Well golly!

Thanks.

Looks like that 5 years of college was worth something! I just wish I would have paid attention during my Electrical Engineering classes. I sure did hate those classes at the time. Didn't understand much of it either - that's probably why I didn't like it.

Anyway - anyone else have any ideas on how to help me with this circuit?

Thanks,
shootfirst

Try this:

Connect a 10k resistor and a 10uF nonpolarized capacitor in parallel. Connect this combination in series with one of your trigger wires. It shouldn't matter which one.
I don't guarantee that this will work, but it gave me a short pulse when I simulated it. I don't know how long your pulse needs to be, or what the Canon flash load is, so it was a bit of a SWAG.
It also takes several hundred milliseconds for the circuit to recover before it will reliably fire again. I'm hoping that's not too long.

There are certainly other sources of nonpolar caps. I just found this one with a quick search.

I suspect that the Canon strobe's recharge circuit supplies a high current so it can recharge quickly. Therefore the SCR in the optical slave keeps conducting the recharging current. Maybe the strobes that can be retriggered don't have such a high recharge current which allows the SCR to turn off.
Ron's suggested 10uF capacitor might need to have a value as high as 100uF to steal enough recharge current away from the Canon strobe in order for the SCR to turn off.

Yes- I know this one because I'm working on one right now!

As said the circuit needs a capacitor in parallel with a bleeder resistor in series with the SCR. The cap will support a significant current until it is all charged up and then the current is only the flash cap voltage divided by the bleeder resistance. If this is below the commutation threshold of your SCR you're good.

A question occurs if the tube fails to fire. The flash cap will still be at 350v or so which means a much higher current through the resistor after the SCR has triggered and the cap has charged. If the SCR fails to commutate it cannot fire again unless the main power is cycled off. So, best thing is to size the bleeder resistance high enough so that the full flash cap voltage/R is less than your SCR's "holding current". Be sure to take into account whether the spec sheet is giving you the highest or lowest guaranteed holding current across all rated conditions. Higher temps=lower holding current which is more of a problem. Usually they give the nominal value for the 25 deg C rating, at 105 deg C it's basically half the nominal value.

For example if you have a 300v cap and a 30mA holding current SCR, 10k is too close. At 100 C junction temp it will hold down to 15mA. So 20k+ would be a good idea.

Ron H said:

Try this:

Connect a 10k resistor and a 10uF nonpolarized capacitor in parallel. Connect this combination in series with one of your trigger wires.

Click to expand...

Thanks Ron and audioguru.

I've had some success with the parallel RC circuit.

Would you mind quickly describing why this works? :?:

The individual who inspired me to try this suggested a 47k resistor and a .1uF cap in the same configuration you described. I tried that combination and also a 47k resistor w/ a 1uF Cap. Do I need to keep a relationship between the R and the C? FYI the voltage running through the trigger wires is about 4.6VDC.

Both solutions work (1uF and .1 uF), but the sensitivity of the optical trigger has dropped way down (w/ either of the caps). The master flash has to be pointing directly at the trigger whereas without the RC circuit on a non-Canon flash, the flash could be pointing anywhere in the room and the trigger would sense the light.

Any ideas on how I can reduce the impact of this RC circuit on the sensitivity of the trigger?

Thanks again for your help!

shootfirst

https://www.botzilla.com/photo/G1strobe.html#wide

It's tempting to think you could use the Powershot's built-in flash to fire slave strobes — this is a very appealing idea and in fact it works very well with many other cameras that have built-in strobes. It's not dependable with the Powershot, however.

The Powershot's built-in flash uses Canon's E-TTL metering system. E-TTL fires twice for a single exposure — once at low power to meter the scene, and then once at full power (or whatever) to expose the image. Upside: very good range of exposures. Downsides: (1) a delay between the time you press the shutter and the exposure, and (2) the first flash will set off your slave units, which will fire and overexpose the meter reading but be unable to recycle quickly enough to make the second, real, exposure.

The safest way to use optically-slaved strobes with the G1 is to trigger them from another external unit. I typically use my old Canon 199A (a speedlite made in the era of the A-1) for this task — it can be swiveled off to bounce from the ceiling while the slave units do the bulk of the work (recently, to avoid any chance of being bitten by the wide-angle mistiming bug, I've been using the 550EX instead. Pretty pricey for a slave trigger!).

Click to expand...

is this what is happening...
(2) the first flash will set off your slave units, which will fire and overexpose the meter reading but be unable to recycle quickly enough to make the second, real, exposure.

williB said:

http://www.botzilla.com/photo/G1strobe.html#wide
[is this what is happening...
(2) the first flash will set off your slave units, which will fire and overexpose the meter reading but be unable to recycle quickly enough to make the second, real, exposure.

Click to expand...

No - that is another common problem that many people run into. In my case, I am disabling the Canon ETTL system and using the flash in manual mode so there is no preflash. It only flashes once. The cheaper version of the Canon ETTL flashes like the 420EX and the built-in flash on most newer Canon cameras will not let you disable ETTL so you are stuck trying to work around the preflash.

That provides a unique electronics challenge as well. How would you get the slave to ignore the first flash but fire on the second.

I'm not interested in tackling that problem. I'm shooting in full manual and trying to deal with the other issue I described in my first post.

Thanks,
shootfirst

Hmm .. How long between flashes is optimal .??
i mean you said that your slave flashes once.. How many times , or how often do you need to flash ??

williB said:

Hmm .. How long between flashes is optimal .??
i mean you said that your slave flashes once.. How many times , or how often do you need to flash ??

Click to expand...

Which issue are you discussing?

The first one that I brought up regarding the flash only flashing once and then locking up or the issue you brought up regarding the pre-flash disturbing the sync of the master and slave flash?

The flash may be asked to flash as much as 5-10 times per second for a rapid sequence - but that is not a typical usage and requires that you flash at a reduced power so that there is ample time to recharge between flashes. More likely you will be triggering the flash every 5 seconds or so. At full power a more frequent rate of flash will overheat the unit.

Thanks,
shootfirst

Hiya Shootfirst,
On reading your post brought back memories of a project siliconchip did it was called "A smart slave flash trigger" and looking in the index it was back in july 2003. Here is the link to the article https://www.siliconchip.com.au/cms/A_30754/article.html
I reckon it could be just what your looking for and it is avaliable in kit form if you don't want to source all the gear and make a circuit board.

Hope this helps

Cheers Bryan

Edit:- Don't pay for the rest of the article if you like it PM me and I'll e-mail it over to you

I got the entire project from Google. I entered Silicon Chip A Smart Slave.
It doesn't talk about the problem where a strobe can't be retriggered due to the SCR in the optical slave staying on.

tuning the RC circuit

This thread is getting slightly off topic. I am interested in the multiple flash issue, but I'd like to get the flash lock-up (SCR) issue solved first.

I've been able to get the SCR to reset using the parallel RC circuit described above - but the sensitivity of the optical sensor has gone down. I'm wondering if I need to tune the RC to get the sensitivity back.

I'm currently using a 47kOhm R and a 1uF Cap wired parallel to each other and inserted in series with one of the trigger leads. The voltage is 4.6 VDC.

Any ideas if I require a larger or smaller resistor or cap? Do I need to keep a specific relationship between the R and the C?

Thanks!

4.6 V at which side ?. ( the cheapo detecter or the canon flash..?? )
Without a schematic of both units , i can only guess as what to do next..

when you touch both terminals of the canon flash unit it goes off ..correct..?
you could use a FET to make this connection..but you would need to know what happens , when the slave trigger goes off

williB said:

4.6 V at which side ?. ( the cheapo detecter or the canon flash..?? )
Without a schematic of both units , i can only guess as what to do next..

Click to expand...

The 4.6V is at the canon flash side.

I'm afraid that I can't get a hold of any schematics. But as you said, connecting the 2 terminals of the flash makes it fire.

I'm not sure what the FET would accomplish. I have very little electronics experience. I'm trained as a Mechanical Engineer so I've had some exposure to electronics but not enough to understand your suggestion without further explanation.

That said, I think I am close to having a working solution with the parallel RC circuit, I'm hoping that I just need to dial in the values of the resistor and capacitor.

Any help is greatly appreciated.

Thanks,
shootfirst

thats the easy part.
triggering the FET is going to be harder, without knowing what the slave flash trigger is doing..

williB said:

thats the easy part.
triggering the FET is going to be harder, without knowing what the slave flash trigger is doing..

Click to expand...

I'm not sure that we are working on this problem from the same perspective.

I have an optical trigger that shorts the two leads. I want to modify this trigger (not redesign a new one) so that it will work with my Canon flashes.

I've had some success with adding the parallel RC circuit but need help dialing in the behavior of this circuit so that it doesn't have such a large impact on the sensitivity of the optical trigger that I plan on using.

On the other hand, the optical trigger I have looks very simple (maybe only 6 or 7 components) and a total redesign would allow me to place the trigger and the additional RC circuit on a single PC board for a much cleaner installation. I'd try to reverse engineer the trigger I already have but it's potted in epoxy and could be a problem.

If someone has a schematic of a similar optical trigger circuit, I'd love to see it.

Thanks again,
shootfirst