Old camera speedlight low voltage trigger conversion

[cprobertson1]

Good afternoon ladies and gents!

I've just discovered an old speedlight that I want to use in my modern DSLR. Sure, I could, just buy a new one, but this is more fun!

The only problem is that the sync/trigger voltage on the speedlight sits very comfortably at 380V - and modern DSLRs use SCRs that are spec'd at less than 24V (and in the case of Canon cameras, less than 6v, for some reason.)

I had a plan to use a zener and a TRIAC to trigger it while keeping the voltages less than 5.6V, and a series of high value (megohm) resistors to keep the current low - but the more I look at it, the more I think that if something fails, it will bite the camera.



What I'm worried about is the failure mode in which a zener fails open circuit - the voltage will raise to the flashgun's switch voltage of ~380V = I have a single point of failure, and it doesn't necessarily fail safe. Sure the current/power may be low, and it may fail closed (in which case it will just retrigger the flash continually) - but if it fails unsafely and I hook that up to my camera it won't be a happy camper!

I considered adding redundancy in the form of multiple zeners in parallel - if one fails open, the rest will stop the voltage rising - however, if the failure is repeatable (i.e. if something is going wrong and will consistently cause zeners to fail open), then it fails silently until the last one, at which point it then fails hot and the voltage goes up again.

My next idea was to add an optocoupler between the camera and triac - but that leaves me with the problem of powering it. I could either power it from the flashgun itself (using the batteries) or add a small external battery (I don't believe I can borrow any power from my hot shoe for this one I'm afraid... do correct me if I'm wrong though!)

So - my questions!

• Are the zeners reliable enough to outlast the flash gun under these circumstances (especially if redundancy is added)?
• Is powering the optocoupler from the 4xAA batteries a bad idea (I'm thinking along the lines of ground loops and capacitive coupling from the voltage multiplier)
• Erring on the side of caution, is there an easy way to indicate that a high voltage (exceeding, say, 6v) on the sync terminals so that I can test before plugging it into the camera? Maybe a simple zener + transistor >> LED to indicate an overvoltage condition? (powered form the flashgun AA batteries)
• Is there a better way of doing all of this that I'm missing? (not including buying a modern flashgun!)

Ps - the attached circuit is borrowed from a circuit seen on a number of flashgun/strobelight-fan forums (some fun reading there! Looks like some of it is back from the usenet days too!)

Thanks for the help folks!

[Zero999]

You must use an opto-coupler, to isolate the lethal voltage in the flash, from the camera. Zener diodes and a TRIAC just won't do. You could try an opto-isolated TRIAC, which should work, assuming the leakage isn't a problem and the trigger current required by the flash isn't too low.

Ensure there are sufficient creepage and clearances, between the hazardous and non-hazardous voltages, on the PCB.

[cprobertson1]

You must use an opto-coupler, to isolate the lethal voltage in the flash, from the camera. Zener diodes and a TRIAC just won't do. You could try an opto-isolated TRIAC, which should work, assuming the leakage isn't a problem and the trigger current required by the flash isn't too low.

Ensure there are sufficient creepage and clearances, between the hazardous and non-hazardous voltages, on the PCB.

Optocoupler it is! I'm afraid I don't have any optotriacs to hand, but I have a number of optocouplers from an old project I can use.

I'm going to need to figure out a way of powering the camera-side of the optocoupler... an external cell would be best, but I'm not sure where I could fit a cell-holder on the existing enclosure... not without securing it to the outside at least!

How is this for an idea: SPDT switch that charges a capacitor from the AA batteries of the flash gun - when switched it completely disconnects the AAs and powers the optocoupler instead: this way the circuits are always kept mostly isolated (though, only by the dielectric strength of the switch)... maybe that's not such a great idea - single point of failure and we potentially lose the isolation.

Right - back to the button-cell idea... wait a minute - optocouplers don't use that much power do they - and it's not like my camera is keeping it on for that long... I might be able to get away with just using an internally mounted cell and not worrying about it until it drains, rather than treating it as a consumable...

I'll look into that too!

Any other suggestions would be most welcome too!

[StillTrying]

Have a search for your camera, most of the 6V are really 250V. I've been using a 270V flash on a 6V/250V camera for years.
http://forum.mflenses.com/canon-flash-voltages-t15474.html

I wouldn't recommend 380, 270, or even 250V, but I don't think you have to worry too much about 6V. If it's powered by it's own internal batteries I don't think galvanic isolation is needed at all, just arrange for a 5 to 12V pulse to trigger the 400V triac or even thyristor.
You could do with the internal circuit of the flash to see where to get the 6V trigger from, rather than using the 380V.
YMMV

[NiHaoMike]

Just get a wireless flash extender.

[floobydust]

Here are typical schematics of opto-isolated low voltage camera flash trigger circuits. They are not great, one uses wrong MOC3010 250V part, and the other a long recovery time with no thyristor.

I have seen really old camera flashes have no SCR- the camera switch directly drives the hot shoe and a trigger coil/discharge cap.
So I'm not sure an opto-coupler alone is strong enough, best to have an SCR after it. MOC302x is rated 1A 100usec at 400V.

[Peabody]

What I've done with older questionable flashes is to relegate them to off-camera use, triggered by a radio trigger receiver or optical slave adapter.  I don't know if either would be available at your high voltage.  My old flashes are Vivitars, but only 12V or so.  But better to put a cheap receiver or adapter at risk than the camera.   But if you're lucky, your flashes will have a built-in optical slave function, so you don't have to put anything at risk.

[cprobertson1]

Have a search for your camera, most of the 6V are really 250V. I've been using a 270V flash on a 6V/250V camera for years.
http://forum.mflenses.com/canon-flash-voltages-t15474.html

I wouldn't recommend 380, 270, or even 250V, but I don't think you have to worry too much about 6V. If it's powered by it's own internal batteries I don't think galvanic isolation is needed at all, just arrange for a 5 to 12V pulse to trigger the 400V triac or even thyristor.
You could do with the internal circuit of the flash to see where to get the 6V trigger from, rather than using the 380V.
YMMV

Let's see - this one is... maybe ~24v, but the manufacturer (Pentax) also says that HV won't kill it instantly. Most things won't kill me instantly either - that doesn't mean I go around tasting random things though! Not very often anyway.

I've still to properly reverse engineer this flash - but adding something between the pins on the hot shoe and the trigger circuit seems the best bet so far (as opposed to buying a third party adapter, or building an adapter that goes between the hot shoe and the flash unit)

Here are typical schematics of opto-isolated low voltage camera flash trigger circuits. They are not great, one uses wrong MOC3010 250V part, and the other a long recovery time with no thyristor.

I have seen really old camera flashes have no SCR- the camera switch directly drives the hot shoe and a trigger coil/discharge cap.
So I'm not sure an opto-coupler alone is strong enough, best to have an SCR after it. MOC302x is rated 1A 100usec at 400V.

Aye, that's exactly what this flashgun expects - the (original) camera's hot shoe was literally just mechanically bridging the gap on the trigger circuit: amusingly simple!

I just realised I never checked the trigger current on that flash sync actually - I had better get on that tonight. I had also better get some pics of the circuit boards for all those discharge-tube aficionados!

I have also got a hold of my dad's old flash which is a little bit more modern - I haven't had a chance to check voltages on it or do a teardown yet though - but I would assume it'd have the same high voltage problem in one form or another - so my need for an interface remains!

Just get a wireless flash extender.

Where's the fun in that!? Nah, if I can't get this running safely for less than £5, I'll save up and get a decent/modern entry-level flashgun that's actually compatible with my camera! I can build a simple slave circuit for the other flashes so I can still get use from them though.


Note to self:
-Check trigger voltage + polarity (both flashguns)
-Check trigger current (both flashguns)
-Take pics of the PCBs for the flashgun aficionados

[Zero999]

I thought wireless flash extenders only worked with low voltage trigger circuits?

I don't think I've seen a trigger circuit which directly connects the capacitor straight to the flash tube. Normally the trigger connects a small capacitor, charged to a few hundred volts, to a small transformer which provides a very low current, high voltage pulse and is capacitively coupled to the tube, via an electrode.

A small relay is probably the safest thing to use and will need a small battery and back-EMF diode to protect the camera from the inductive kick-back.

[StillTrying]

I'd probably do something like this, ignore all this opt-isolation nonsense.

[floobydust]

The need for an opto-coupler might be a carry over from the days of mains-powered flashes which are not isolated. I'm in favour of one to block any EMI and ensure the camera body could not get damaged by the flash.

Old flashes are usually HV DC with a trigger transformer and trigger capacitor raw to the hot shoe - no thyristor at all. So the camera shutter switch is to discharge HV trigger capacitor into the trigger coil. Best to add thyristor.
If you put a scope there, you'll see the SCR retriggers a few times due to ringing with the trigger coil's inductance. It's not a nice single pulse.

One question is the configuration of a low voltage camera hotshoe. It seems to be open-collector with a weak pull-up resistor to camera's Vcc?

Schematic is of old Fuji FinePix S100FS camera that can drive a HV hot shoe.

[Zero999]

Even if it's not powered from the mains, isolation is a good idea.

Just use a relay.


This one is perfect: high coil resistance of 500R, a contact rating of 200V (but will probably take a bit more) 1A and a built-in back-EMF diode.
https://uk.rs-online.com/web/p/reed-relays/2919710/

[cprobertson1]

The need for an opto-coupler might be a carry over from the days of mains-powered flashes which are not isolated. I'm in favour of one to block any EMI and ensure the camera body could not get damaged by the flash.

Old flashes are usually HV DC with a trigger transformer and trigger capacitor raw to the hot shoe - no thyristor at all. So the camera shutter switch is to discharge HV trigger capacitor into the trigger coil. Best to add thyristor.
If you put a scope there, you'll see the SCR retriggers a few times due to ringing with the trigger coil's inductance. It's not a nice single pulse.

One question is the configuration of a low voltage camera hotshoe. It seems to be open-collector with a weak pull-up resistor to camera's Vcc?

Schematic is of old Fuji FinePix S100FS camera that can drive a HV hot shoe.

No idea about the hot shoe schematic on my camera (Pentax K-50) (wait, I just realised I never mentioned that earlier in the post?) - I was hoping that my modification could be mostly-universal though: and an optocoupler just makes sense for that given that I can't know the specs of the hot shoe it may or may not be attached to.

Side note - I'm saving up for a modern flashgun too xD I see a £70 model I like the looks of, though it's still only entry level of course (I'm an entry level photographer, after all!). I still want to get this old one working though - at least the Hanimex TZ7000 (I still haven't tore this one down yet or even probed it for that matter). The older flash (the one I was originally working on) is a Sunpak (I can't remember the model name I'm afraid). The sunpak is a little monster though - it's bit me twice now, and it also has reversed polarity on the hot-shoe (centre-negative) - though that's easy enough to fix just by swapping the wires around internally. It also has a big voltage spike (in the opposite direction) when the flash fires... I might need to add some snubbing in there too - but I'll deal with that bridge when I come to it!

I'd probably do something like this, ignore all this opt-isolation nonsense.

Ah! Okay! ... Relay then? 

Even if it's not powered from the mains, isolation is a good idea.

Just use a relay.


This one is perfect: high coil resistance of 500R, a contact rating of 200V (but will probably take a bit more) 1A and a built-in back-EMF diode.
https://uk.rs-online.com/web/p/reed-relays/2919710/

Ha! Brilliant! I went to post that up as a half-serious/half-joke suggestion, and as I hit send, I get a message saying there's another post where Zero999 actually  using a relay - impeccable timing!

I did consider it, but was worried about how best to power it: see, with an optocoupler I can use a primary cell for a potentially very long time - with a relay the current draw is potentally 10x greater. That said, it's not like we'll be drawing current for a long time - I could probably get away with it.

In fact, come to think of it... could I just power the coil from the 4xAA batteries? As I mentioned earlier, I was pensive about the idea as it means things aren't fully isolated and I was worried about loops - but on the other hand, I really don't want to be adding an entire additional battery source into the flash unit if I can help it (not a big one, anyway).



Side note: do relays switch fast enough? The part you linked to lists 0.5ms including bounce - so if the shutterspeed is as faster than 1/2000 I may actually miss the flash (plus any other delays in the trigger circuitry on the flash itself)! I mean, it shouldn't be a problem - if I'm flashing I probably won't be using shuterspeeds that high - but it might be fun to play around with all the same! I wonder how that trigger-time compares to the TRIAC circuits, actually - I never thought to work out the trigger-time for them.

[cprobertson1]

Sorry to come back to this again but I can't wrap my head round this for some reason - or rather, I keep flip-flopping as to whether it's a good idea or not.

Powering the optocoupler or relay (coil) from the flashgun's 4xAA batteries would be ideal... but, because we don't have total isolation, I have some fears - which may or may not be well-founded!

• Can the HV voltage multiplier form a capacitive loop, causing a high voltage to appear on the hot shoe contacts?
• Can EMI cause a HV pulse on the hot shoe contacts?
• Under reasonable conditions, can the HV multiplier fail in such a way as to expose the hot shoe contacts to HV?
• Am I worrying over nothing?

I'm rather annoyed with myself that every time I draw the circuit out/look at the block diagram I've drawn, I end up bouncing back and forth: I feel like this is something I should just know!

--EDIT--
The power draw is actually pretty low, even for the relay - and I'm not running it continuously - I reckon I could get away with a number of different primary cells, while still keeping everything neatly isolated from the camera itself - so I may have addressed my own concerns!

--EDIT--
Corrected "nearly isolated" to "neatly isolated"

[Zero999]

The circuit I posted fully isolates the camera from the flash gun, as the relay's contacts are isolated from the coil. A 6V battery or four AA batteries in series, to make 6V will be fine and should last for ages, as the relay will only be activated for short periods of time.

I don't know whether it will be fast enough. How fast is your flash gun?

[StillTrying]

Side note: do relays switch fast enough? The part you linked to lists 0.5ms including bounce - so if the shutterspeed is as faster than 1/2000 I may actually miss the flash (plus any other delays in the trigger circuitry on the flash itself)!

When you enable flash your camera will ignore the 1/2000 and fix it at 1/100 or 1/200, because the shutter has to be fully open for all of the 30us to 2ms of the light flash, I don't think an extra 0.5ms delay to the 5ms or 10ms shutter time would make any difference.

Just wire a thyristor to the flash and apply 4 or 5V to it's gate, you'll find there's nothing to it! But be very careful the 380V doesn't go through you or the scope.
On my cheap flash it's possible to make contact with the outside ring of the ext trigger, and the center pin of the hot shoe at the same time, which occasionally gives me a surprise.

The problem with isolation is some power is needed on the camera side.

One question is the configuration of a low voltage camera hotshoe. It seems to be open-collector with a weak pull-up resistor to camera's Vcc?

The middle of this page suggests there's a weak 6V coming out of the camera which can be stored and used to trigger the opto's input side, I've never heard of it myself.
https://electronics.stackexchange.com/questions/242088/connecting-a-camera-hot-shoe-to-a-homemade-strobe

[cprobertson1]

When you enable flash your camera will ignore the 1/2000 and fix it at 1/100 or 1/200, because the shutter has to be fully open for all of the 30us to 2ms of the light flash, I don't think an extra 0.5ms delay to the 5ms or 10ms shutter time would make any difference.

Ah, nice! That makes sense! I'm forgetting the camera will override settings - I've not actually used a flashgun with a modern camera before - the last one I used was an entirely manual film camera, and even then I was pretty young!

The middle of this page suggests there's a weak 6V coming out of the camera which can be stored and used to trigger the opto's input side, I've never heard of it myself.

Now that's interesting! I'll probe around with that as well - though Im currently leaning towards just using two CR2032s in series for the camera-side power.

With the CR2032 x2, running for <1s at 10mA coil current, I'm reckoning I'll be able to get 10'000 discharges from it.

Code: [Select]

~235mAh*0.66 (to account for poor battery conditions) =155mAh
155mAh/10mA=15.5 hours=55800 seconds >> 10'000
So, let's assume for argument's sake that the coil runs for a whole second each time the flash is fired, those cells should be more than happy for a good 10'000 shots.

Let's just run that again with Watt-hours just as a quick sanity check:

Code: [Select]

~235mAh = ~650mWh (according to the datasheet's energy density figure - I'm actually getting 470-to-705mWh when I perform the calc. myself, depending whether I use the final voltage or the nominal voltage.)
650mWh*0.66=430mWh
Relay @ 10mA, 5V = 50mW
430mWh/50 = 8.58hours = 30888 seconds >> 10'000
Okay, I'm not too happy I got a different figure there... I must have done something dumb. Still over 10'000 shots should be possible even if that calculation hasn't raised my confidence any!

I'm now scratching my chin trying to figure out why those numbers aren't particularly close to each other... it may just be that mAh are a bad way to measure battery capacity, but I was hoping the figures would have been a bit closer than that...

[StillTrying]

Have you blown anything up yet.

235mA/10mA = 23 hours = approx. 82,000 1sec triggers.

The flash's 380V only has to be shorted for about 4us so you could reduce the camera's trigger to just a 30us pulse through an opto or ~3ms for a small relay and even smaller batteries would last forever.
You can usually see the trigger transformer ringing by bringing the flash near to a scope probe, no physical connection is needed, it only rings for a few us, then there's another 5-8us delay before the tube lights, I've watched it many times!

[Peabody]

I believe your Pentax uses a standard hotshoe, but my memory is the Hanimex is for a Minolta hotshoe.  If that's true, the Hanimex will not fit on your Pentax.  I mean it physically will not fit.  You didn't say what model your Sunpak is.

I would just encourage you again to use these old flashes off-camera.  You will find that the results you get with off-camera lighting make it look like you know what you're doing, even if you don't, and of course your camera would not be at risk from high voltage.  You would need a light stand, an umbrella adapter and an umbrella, plus some way to deal with the high voltage.  And you can get a basic wireless trigger set for $20.  This would be manual triggering of course - both camera and flash(es) in manual mode.  None of this ETTL rubbish.  :-)

[floobydust]

A relay adds a huge delay 3-10msec to operate, are we really thinking a flash that late is going to work?

I took apart an old bounce flash I have, it's got a 800uF capacitor (!) and about 160VDC at the shoe and 320VDC for B+. Like an old Metz I repaired, serious sized capacitor.

[OldEE]

I bought one a Wein adapter a number of years ago for a similar application.  Looks like they are not being made anymore and BHPhoto in New York is clearing them out for about half price $25.  B&H does ship internationally.

https://www.bhphotovideo.com/c/product/245292-REG/Wein_W990560_Safe_Sync_Hot_Shoe_to.html

Larry

[cprobertson1]

Have you blown anything up yet.

Not as part of this project!

235mA/10mA = 23 hours = approx. 82,000 1sec triggers.

Ah, good, my method was fine (though I deliberately chopped a third of the capacity off for "undisclosed reasons").

The flash's 380V only has to be shorted for about 4us so you could reduce the camera's trigger to just a 30us pulse through an opto or ~3ms for a small relay and even smaller batteries would last forever.
You can usually see the trigger transformer ringing by bringing the flash near to a scope probe, no physical connection is needed, it only rings for a few us, then there's another 5-8us delay before the tube lights, I've watched it many times!

So I modified the Sunpak. It's a Sunpak 134 - which, according to this site should have a trigger voltage of "only" 50-ish volts... so I don't know why my unit sits at 380V (were there different revisions of the same flash? ::shrugs:: )

There was barely enough space for my modifications - I decided to go for the optocoupler since I had it to hand, and it was smaller. Optocoupler powered by a CR2032 stuffed down a tiny gap between the PCB edge and the case, triggers the TRIAC on hard via an NPN transistor. Ended up using SMD parts on some stripboard and then cutting the remainder off with side cutters and filing the edges down to keep the size as low as I could manage.

The only problem I see is the creepage distance - I cut the board with a hacksaw and soldered the optocoupler across it (about a 5mm gap) - but because it's wedged in a small space, there may be creepage across the side of the case (I put a sil-pad in there, but I don't know how that performs when it comes to surface creepage.)

I think I'll keep it away from my camera for now just in case: I'll relegate that one to wireless triggering once I can afford that!


As for the Hanimex TZ7000... turns out it runs with a trigger voltage of 4.4-4.5V - I'd call that a win!

I believe your Pentax uses a standard hotshoe, but my memory is the Hanimex is for a Minolta hotshoe.  If that's true, the Hanimex will not fit on your Pentax.  I mean it physically will not fit.  You didn't say what model your Sunpak is.

I would just encourage you again to use these old flashes off-camera.  You will find that the results you get with off-camera lighting make it look like you know what you're doing, even if you don't, and of course your camera would not be at risk from high voltage.  You would need a light stand, an umbrella adapter and an umbrella, plus some way to deal with the high voltage.  And you can get a basic wireless trigger set for $20.  This would be manual triggering of course - both camera and flash(es) in manual mode.  None of this ETTL rubbish.  :-)

Aye, I didn't realise the wireless units were that cheap - I may partake once I get enough money together for a new flash.

The Sunpak is a "134" - though other people are only measuring around 50V on their trigger pins, so I don't know what mine is doing (different board rev maybe? ::shrugs:: ).

The Hanimex is indeed supposed to fit a Minolta (Maxxum 7000?) camera, but it seems to fit my pentax and fujifilm no bother. There is an extra pin for the AF illuminator and the functions of the other two pins are flipped round - BUT the foot of the flash is actually longer than the modern ISO hot shoe, so they don't make contact anyway. The trigger pin makes contact though, so it ends up working fine anyway: I'd call that a win!

I might add in a similar optocoupler + battery anyway, just to be on the safe side. It's thyristor triggered (I think?) and there is some control circuitry in the bottom module. Oh, there's an interesting point about the construction - the Hanimex flash is two modules that clip together - the bottom module contains what I assume is some control circuitry, and connects to the rest of the flash via a board-edge connector. It's entirely enclosed and can be disconnected with no tools from the rest of the flash - I assume they reused the control circuitry across several different flash units - just a matter of slotting it in: interesting!

[Zero999]

I you use a relay, there is absolutely no point in having an opto-coupler as well.

I forgot it said 380V, so the relay I previously linked to won't do. This is better. It has two contacts rated to 220VDC, which can be connected in series. It doesn't have a built-in diode, so that will need to be added separately.
https://www.digikey.com/product-detail/en/te-connectivity-potter-brumfield-relays/2-1462039-9/PB1165CT-ND/1828459

The battery life can probably be boosted by connecting a capacitor across the battery. As the battery discharges, its internal resistance will rise. Adding a capacitor will enable it to provide a large current pulse to fire the relay, even if the battery's resistance is too high for it to happen otherwise. Try a 100µF capacitor.

If you go with a TRIAC opto-coupler, then it only needs a very short pulse to fire, so set the pulse so it's quite short. If the shutter switch produces a much longer pulse, than necessary, power could be saved by using a high value resistor to charge a capacitor and the camera shutter switch to discharge it into the opto-coupler, via a lower value resistor.

Note that the TRIAC/BJT side of the opto-coupler will need to be rate to the full flashgun voltage. If you use a BJT, then it will be polarity sensitive, unless a bridge rectifier is connected between the opto-coupler and TRIAC.

[StillTrying]

Have you blown anything up yet.

Not as part of this project!

If you're not going to blow the camera or the flashes up, make a sound triggered flash and blow some other things up.

[cprobertson1]

I you use a relay, there is absolutely no point in having an opto-coupler as well.

I forgot it said 380V, so the relay I previously linked to won't do. This is better. It has two contacts rated to 220VDC, which can be connected in series. It doesn't have a built-in diode, so that will need to be added separately.
https://www.digikey.com/product-detail/en/te-connectivity-potter-brumfield-relays/2-1462039-9/PB1165CT-ND/1828459

The battery life can probably be boosted by connecting a capacitor across the battery. As the battery discharges, its internal resistance will rise. Adding a capacitor will enable it to provide a large current pulse to fire the relay, even if the battery's resistance is too high for it to happen otherwise. Try a 100µF capacitor.

If you go with a TRIAC opto-coupler, then it only needs a very short pulse to fire, so set the pulse so it's quite short. If the shutter switch produces a much longer pulse, than necessary, power could be saved by using a high value resistor to charge a capacitor and the camera shutter switch to discharge it into the opto-coupler, via a lower value resistor.

Note that the TRIAC/BJT side of the opto-coupler will need to be rate to the full flashgun voltage. If you use a BJT, then it will be polarity sensitive, unless a bridge rectifier is connected between the opto-coupler and TRIAC.

Ooop! It was just the the optocoupler that I added, I don't have any relays small enough so I decided to just stick with the opto. Come to think of it, I'm not actually sure I'd have been able to fit the ones you linked me to in the space available (in fact, I could barely fit the SMD parts I ended up using. Plenty of length, but very little height: I actually need to go back and make double-check the leakage paths on it - I thought of another path across the edge of the existing PCB that might cause even bigger problems. Sil-pads to the rescue? I that allowed?)


Good shout on the bridge rectifier for the optocoupler's BJT on the flash side, btw! There's only one diode in there just now (and I debated adding a second to act as a snubber, but being unable to probe it with the scope, I decided to wing it and go without...) - I'll modify a proper bridge in tonight.

The transistors (And optocoupler) are rated for 400V. I'd rather have a 600V or higher rating just to be safe, but the 400V ones are what I have to hand - and for 380V they should be fine for a while - they're not exactly putting much current through them, so they're within their power spec.

Have you blown anything up yet.

Not as part of this project!

If you're not going to blow the camera or the flashes up, make a sound triggered flash and blow some other things up.

I reckon if I string a number of these flashes out I could visually record and work out the speed of sound from visual cues!

"flash-flash-flash-flash-BANG!" - okay, I don't have that many flashes, but still!