Help making an camera flash trigger

[mribble]

The http://www.CameraAxe.com  has flash triggers using an optoisolator (http://www.mouser.com/ProductDetail/Lite-On/LTV-847/?qs=sGAEpiMZZMteimceiIVCBx3Y662pemTNEFNr76lpuig%3d) .  This works great for modern low voltage flashes, but older high voltage flashes exceed the component's 25V max.

The standard solution online for high voltage flashes is to use an SCR to trigger flashes.  I've tested this one and it works for high voltage flashes (http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=497-3321-1-ND).  The problem with SCRs is that modern low voltage flashes never drop their current to zero so the SCR never turns off.  This means you can trigger a modern flash once and then you need to disconnect it from the circuit before you can trigger it again.

I would make a circuit that would trigger both low voltage and high volt flashes.  I thought a high voltage transistor would be the solution.  First I tried a standard 2N2222 NPN transistor and that worked fine with low voltage flashes.  I thought all I'd need is a high voltage transistor so I tried this one http://www.mouser.com/ProductDetail/Supertex/LND150N3-G/?qs=sGAEpiMZZMv4eh0jmGe02%2ftg6hfkwCDeGhYm1QIW7Ls%3d.  This didn't work.  It would trigger the flash (ie transistor was switched on) when I had the gate hooked attached to ground or 5V.  I thought attaching the gate to ground would turn the transistor off, but it doesn't seem to do that.  I admit there are a lot of things on that datasheet I don't understand...

Could someone help me understand what is going on here?  If you have another idea how to trigger low and high voltage flashes that would also be appricated.  I thought it would be cool if I could use the voltage from the flash to power my circuit, but I couldn't get that working either so I've decided this will need a battery.

[Zero999]

If it's a high voltage flasher, it's a good idea to have isolation so if I went for an SCR I would've used a pulse transformer or opto isolator at the gate.

How about just using a relay?

[mribble]

Hero, I agree that you should have isolation.  Since I'm using it with the Camera Axe it will have the 8000 V of isolation through an opto isolator I linked above.  In theory this isn't needed, but I agree it's a good safety mechanism.

This is used for very high speed photography so relays are too slow.  I need something with less than a 1 us delay.

[Zero999]

A solid state relay?

Still might not be able to get down to 1µs though.

[mribble]

Solid state relays are interesting idea, but when I looked at them for the voltages I needed their on times ranged from 10s to 100s of microseconds so I don't think they are fast enough unless someone knows about a really fast one.  I kind of doubt that though since for really fast things like this people don't usually use relays.

[Simon]

perhaps a schematic of how all this should work will help, I can't say I'm familiar with flash and flash trigger circuitry even if I do dable in photography

[Zero999]

I don't see why it needs to be that fast.

What are you filming?

Even if the object is travelling at the speed of sound, 340m/s, it will only travel 3.4mm in 10µs.

[Kiriakos-GR]

It looks to me, that some one is trying to cheat  and not pay , the 400$  usual price ,
for an true fast "digital flash unit" .  

Well its not impossible to make it , but it is not works by just picking, some fast single parts.  

The master and slave flash units , hides lots of theory and complex technology behind it ..


.

[mribble]

I'm photographing bullets.  Here's a article I wrote about it: http://diyphotography.net/bullet-photography-at-home

Bullets travel up to 1000 m/s.  The trigger system adds a few µs of delay.  The sensor adds about 1/2 a µs of delay.  The flash adds < 1 µs of delay.   I suppose a few more µs delay from this high voltage protector would be ok, but my target is less than 1 µs.

Below is a schematic of what I'm doing.  I got this from here: http://www.carlmcmillan.com/Optoisolated_Adapter.htm

Flashes units use a current source to trigger a flash.

Code: [Select]

                +---------------------+      R2
   o-----------(1)---+         +-----(6)----/\/\----+----o Flash Trigger (+)
                |    |  OPTO1  |      |     5.6K    |A
To CPU          |  __|__   ____|____  |           __|__
pin             |  _\_/_-> _\_/_/_\_ (5)NC        _\_/_   SCR1
                |    |         |      |           / |     400V,6A
       --------(2)---+         |      |          |G | K   (RS#276-1020)
       |     NC(3)             +-----(4)---------+  |
       /        +---------------------+             |
   R1  \              Optotriac                     +----o Flash Trigger (-)
  330  /               MOC3010
       \             (RS#276-134)
       |   
      GND   

The problem is the newer flashes don't work well with the SCR (reasons explained above).

I don't really follow Kiriakos's claim that I'm trying to avoid paying for a digital flash unit.  I guess I want this because older flashes tend to use high voltages and that's the reason I need this, but it has nothing to do with digitial.  Actually the main reason I'm building this is because I've built a custom flash with a flash duration about 50 times faster than you can buy commercially that has a trigger voltage of 400 volts.

[Jimmy]

Try using a relay to open circuit the scr using a nc contact. The relay will take some time to open hopefuly long enough to trigger the flash

[Mechatrommer]

I'm photographing bullets.  Here's a article I wrote about it: http://diyphotography.net/bullet-photography-at-home
code...

nice schema! hi-V not only around for "punny" flash units, there are lot more lighting setup including the bigger and more powerful studio strobes etc that can be priced easily $1K++, even without remote and TTL capability built in, just a dumb manual power setting knobs. i'm building a kind of flash trigger myself, the available trigger on the market priced at close to the price of the Auto-TTL flash unit itself! crazy profit maker(s)!

how about changing the scr to igbt, mosfet, bjt etc? but i'm yet to discover the real difference between them.

[Zero999]

I think you need to post the rest of the circuit,

Normally it's not the SCR which connects the flash to 400V, it's permanently connected. The SCR normally discharges a small capacitor into a high voltage pulse transformer which produces a short high voltage pulse (about 20kV) which is coupled to the tube's trigger terminal. Before the tube is triggered, it holds back the 400V, when the gas inside the tube is ionised by the 25kV trigger pulse it forms a conductive plasma which allows the 400V to flow causing it to flash.

It sounds like the resistor in series with the trigger capacitor has too lower value so the SCR never turns off. If you increase its value so that the DC current is below the holding current of the SCR, it should solve the problem. Interestingly, as the amount of energy stored in the trigger capacitor is so small, you could probably remove the SCR and use the TRIAC in the opto-isolator to directly trigger it, which will make it faster.

[jahonen]

Your fet seemed somewhat bad for the application, Rds(on) is high, 1 k?. Furthermore that LND150 is a depletion mode type (quite unusual). So it is on by default and can then be turned off using suitable gate voltage. Try something like FQP6N40C. IGBT would perhaps also work.

Perhaps it is a bit clumsy but if above approach does not work, then the conventional SCR could be turned off with a parallel fet. That would divert the current away from the SCR, assuming that fet on-resistance would be sufficiently low.

Are you connecting the flash trigger circuit to anything potentially sensitive and expensive electronics? If not, then consider dropping the isolation entirely. Just put a TVS in trigger line to prevent high voltage damage in case the switching component fails catastrophically. If isolation is a must feature, then consider faster optoisolator (like 6N137) or digital isolator like ADuM1200C that could reduce the propagation delay to less than 100 ns (guaranteed maximum propagation delay is 45 ns for ADuM1200C). One complication of both that they would need voltage in isolated side, but i think that ADI makes some types (ADuM6200 etc) which also supply small amount of isolated DC-power to isolated side.

Regards,
Janne

[Zero999]

Here's the schematic for the typical camera flash I was talking about above.



C2 is the trigger capacitor, R2 is the resistor and S2 is the trigger switch which needs replacing with an SCR or opto-TRIAC.

It's possible that the circuit could be different.

If all you need is a short pulse, you could try using a MOSFET coupled via a small pulse transformer to provide isolation but the pulse transformer must be driven from a reasonably low impedance push-pull source otherwise the winding inductance will resonate with the gate capacitance.

[mribble]

I think Jimmy's idea of using a relay and an SCR would work, but I'd like to go a simpler route if possible.

I will try Janne's idea of using a FQP6N40C transistor.  I don't quite follow the suggestion of using a fet in parallel with the SRC.  That sounds like it could be better backup plan than a solid state relay, but I'd need help picking the proper fet.

Thanks Hero for posting an example schematic.  The problem is that I want this to work with lots of different flash and the circuits are all different, but having a single example schematic is useful.  But I'm assuming that's there.  What I want to do is to allow different types of flashes to all be triggered by a single trigger that only allows low voltages.

Thanks for all the help you guys have given so far!

[jahonen]

I think Jimmy's idea of using a relay and an SCR would work, but I'd like to go a simpler route if possible.

I will try Janne's idea of using a FQP6N40C transistor.  I don't quite follow the suggestion of using a fet in parallel with the SRC.  That sounds like it could be better backup plan than a solid state relay, but I'd need help picking the proper fet.

Thanks Hero for posting an example schematic.  The problem is that I want this to work with lots of different flash and the circuits are all different, but having a single example schematic is useful.  But I'm assuming that's there.  What I want to do is to allow different types of flashes to all be triggered by a single trigger that only allows low voltages.

Thanks for all the help you guys have given so far!

The idea of parallel mosfet was that you turn the fet on to turn the SCR off (mosfet steals the current from SCR). After the SCR has turned off, you can turn the mosfet off. Now both devices are off. That mentioned FQP6N40C mosfet should work ok for that also.

Regards,
Janne

[Zero999]

Here's the circuit I was talking about before.

The transformer is just a pulse transformer. You could probably make your own with a ferrite bead.

The primary of the transformer should be connected to an astable multivibrator circuit such as a 555 timer or logic gates not just a switch because the pulse mustn't be long enough to saturate the core and the secondary needs to be shorted to 0V when the pulse goes low.