Old camera speedlight low voltage trigger conversion

[StillTrying]

When the flash's trigger transformer and 380V charged cap are shorted to gnd to create the 4kV spikes, the current is 10s of Amps for a few us. So don't worry about it.

[cprobertson1]

When the flash's trigger transformer and 380V charged cap are shorted to gnd to create the 4kV spikes, the current is 10s of Amps for a few us. So don't worry about it.

Those numbers scare me...

I haven't managed to get a good reading on the trigger current I'm afraid. I was getting figures between 30 and 100 milliamps with the multimeter on min/max mode, but if there are spikes I wasn't able to read them properly... the TRIAC is rated to 1A though, so if it's less than that we should be okay for the most part (excusing voltage spikes) and the transistor driving it is rated at 300mA, so, hopefully, I'm not driving anything into the ground (or rather, with the low duty cycle, it might survive the abuse long enough that it won't matter.)

Time will tell xD

[StillTrying]

Most things in a flash are at unusual levels. When even a small tube lights it got about 90V across it and about 100A through it, which is why it makes a nice pop! The main 350V storage cap doesn't like it much either.

The pulse currents of thyristors and similar are very high, and when over their rated voltage they tend to just switch on prematurely with no lasting damage.
I'll try capturing the few us of trigger transformer current later.

[cprobertson1]

I do have to say, that pop is amazing! ::grins to self::

Oh, and the coil whine when it charges! 

Actually... er... two moments... yeah, that whine doesn't stop in the Hanimex unit even after the "ready" light comes on. I wonder if it's _meant_ to do that; or if its leaking charge and needs to continually recharge the flash capacitor once it's at voltage...

Actually, I might refurbish all the caps anyway: I'm sure it couldn't hurt. Unless I touch them - youch!

[Zero999]

When the flash's trigger transformer and 380V charged cap are shorted to gnd to create the 4kV spikes, the current is 10s of Amps for a few us. So don't worry about it.

But the energy delivered to the trigger transformer is tiny. A modest sized TRIAC should be able to handle those numbers for a short period of time, without being damaged.

[floobydust]

I did use small TO-92 SCR's in a conveyer belt warning strobe design (flashing 24/7) and they didn't last on the trigger circuit.
Better to use a medium sized SCR like C106D  (TO225AA screw you JEDEC it's TO-126) rated 400V 4A common in strobes. It handles the trigger transformer peak currents.

I'm not sure why everyone is putting the silly RC circuit delay in. Sometimes you need a burst of flashes and the RC delay ensures some will not happen. Leave it out.

[Zero999]

I did use small TO-92 SCR's in a conveyer belt warning strobe design (flashing 24/7) and they didn't last on the trigger circuit.
Better to use a medium sized SCR like C106D  (TO225AA screw you JEDEC it's TO-126) rated 400V 4A common in strobes. It handles the trigger transformer peak currents.

I'm not sure why everyone is putting the silly RC circuit delay in. Sometimes you need a burst of flashes and the RC delay ensures some will not happen. Leave it out.

I agree, a TO-92 sized package would probably be too small.

Are you talking about the schematic I previously posted, regarding the RC circuit? That was only there in case the camera was providing a ridiculously long pulse, which wasted power. I agree should be taken out, if that's not the case, as it will limit the repeat time.

[cpt.armadillo]

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!

The Minolta Maxxum 7000 hotshoe is standard shape and size with a few extra contacts for the fancy auto features. They changed to some proprietary thing in the 90's IIRC. I have a Vivitar (I think) flash with a switch to choose between the various sets of contacts for different fancy hotshoes. I would be surprised if the current standard hotshoe is smaller when manufacturers might have just decided to use a shorter one as I run into the same problem with cameras from the 1960s.

[floobydust]

... Are you talking about the schematic I previously posted, regarding the RC circuit? That was only there in case the camera was providing a ridiculously long pulse, which wasted power. I agree should be taken out, if that's not the case, as it will limit the repeat time.

There's a lot of conversion circuits, like in post #5 with a RC various different places. If we don't need multiple flashes right away- I saw a camera do a dummy flash to read back light intensity, then flash again to take the picture, then go to the next level?

Capacitive-discharge (batteryless) scheme is like Jean-Paul Brodier build then Robotroom.com Safe-Sync added to it. pic below.

I think the camera's have open-collector (LV 6V) or triac outputs (HV 250V).

Flash hot-shoe voltages but all over the place, I wonder if people are measuring nothing or leakage.

[cprobertson1]

There's a lot of conversion circuits, like in post #5 with a RC various different places. If we don't need multiple flashes right away- I saw a camera do a dummy flash to read back light intensity, then flash again to take the picture, then go to the next level?

For those interested (but not into photography per se), the double-flash observed is indeed for metering (measuring the light levels). It is part of a metering scheme called TTL (Transistor-Transistor Logic "Through-The-Lens") - it flashes twice - once with the mirror down to perform the metering, and then it flips the mirror up and flashes again to take the photo.

With the advent of adjustable flash output (I believe they do this by electrically quenching the flash tube to limit the flash duration, though I can envision a few other ways to get similar effects - such as limiting the discharge current) - some fancier cameras/flashes adjust the flash output on-the-fly between these two flashes to achieve a desired exposure.

Now, interestingly, I believe some cameras can sometimes do away with the first metering flash by performing the metering without the flash, and then making a reasonable estimate as to what flash output would be needed to achieve the desired exposure (or what settings would be required to achieve the desired exposure with a known but fixed flash output). Obviously, guesstimating the reflectivity (albedo? Can I call it albedo if I'm not doing astronomy? Does it work for things smaller than planets and moons?) isn't as accurate as actually measuring it with the flash firing - but for smaller cameras it's typically good enough - I would assume that the pros would rather use the metering flash though, as it gives more accuracy and precision in the metering (as it takes the camera's guesswork out of the equation).


Back to business! In my case, the TZ7000 has a 3-5 second cycle time on full power, and about 0.5 seconds on low output (but you only get two flashes and then it's a full 3-5 second cycle time again) - either way, the recovery time in my isolation/sync circuit is negligible compared to the recharge time on the flash

I did use small TO-92 SCR's in a conveyer belt warning strobe design (flashing 24/7) and they didn't last on the trigger circuit.
Better to use a medium sized SCR like C106D  (TO225AA screw you JEDEC it's TO-126) rated 400V 4A common in strobes. It handles the trigger transformer peak currents.

I'm not sure why everyone is putting the silly RC circuit delay in. Sometimes you need a burst of flashes and the RC delay ensures some will not happen. Leave it out.

Hrm - that might be a problem - it's a TO-92, 400V 1A, TRIAC I used in mine... hrm... that might die soon then =/
I modded in that diode bridge last night but space is now even tighter: could barely get it back in - assuming it's going to die eventually, I'll replace it with something beefier - though I'm not sure where it's going to go!)

[StillTrying]

If I bend a 0.5R resistor into a U shape and use it to trigger the flash at its hot shoe I nearly always get 2 narrow +ve peaks of around 55A and 40A. It would probably be different with an electronic switch rather than a metal to metal contact.