Controlling AC Xenon strobe power

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Hi All,

I know xenon strobe beacons have used the same basic circuit for years and years and most of the time they are fine but I am working on a strobe and I would like to be able to fully control its power and efficiency.

Attached is a general method of powering an 230ACV xenon, using drop capacitors and the normal calculation of current / frequency etc. I would like to try and use some kind of dimmer circuit instead of the drop caps and use a PIC to control the power to the circuit. I havent seen that used anywhere so I am wondering if at all it would be feasible. Effectively, a dimmer circuit of a kind with a PIC feeding trigger control to the dimmer to adjust the incoming power for the required output.

Any input would be appreciated.
 

Attachments

Dodgydave

Joined Jun 22, 2012
11,284
Then why dont you try it on a normal light dimmer, put the bridge rectifier to the dimmer, and see if it will work,then work on contolling the dimmer triac with the pic.
 
Last edited:

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Then why dont you try it on a normal light dimmer, put the bridge rectifier to the dimmer, and see if it will work,then work on the contolling the dimmer triac with the pic.
That is the general idea - looking for input from someone who has tried it or knows a better way.
 

Dodgydave

Joined Jun 22, 2012
11,284
It will work upto a point, all that will happen is when the strike voltage reaches its limit, the pulse can fire the strobe, any less voltage and the strobe wont work, any more voltage and the stobe is a brighter flash.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
I am just wondering why no one seems to have used this method, there are a lot of AC xenon circuits that pass the drop caps they use high wattage resistors both to control the power and to protect components downstream, a lot of heat is generated in these resistors which is a complete waste.

Will give it a try, thanks for your input.
 

AnalogKid

Joined Aug 1, 2013
10,986
No one?

Almost all hot-shoe-mount camera strobes and all pro AC powered units have variable power. The flashtube is essentially a dead short when it is flashing. To decrease the perceived brightness, decrease either the current through the tube or the time it is on. I have a vague recollection of a circuit with a GTO - gated turn-off SCR. Today, either/both can be done with a high voltage power MOSFET in series with the tube.

ak
 
Last edited:

Thread Starter

seanstevens

Joined Sep 22, 2009
251
AK, thanks for that. My application is not photography, its warning signals (xenon), generally they dont have any power control and they never want to reduce the power! The idea of power control is to just give enough power to produce the required Jules, nothing more to burn in caps or resistors which not only heat up surrounding components and the housing but also just waste energy.

The idea of using a dimmer was a simple way of conveying what I want to do. If TRIAC or MOSFET can be used and successfully control the power, then I can knock up a circuit to put the theory in to practice to see if it works efficiently.
 

Sensacell

Joined Jun 19, 2012
3,432
I have used an IGBT in series with the tube to great effect.

You limit the duration of the discharge to control the energy released, the trick is getting the trigger voltage synchronous with the gate pulse.
The caps don't discharge all the way, it can recycle faster too.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
251
I have used an IGBT in series with the tube to great effect.

You limit the duration of the discharge to control the energy released, the trick is getting the trigger voltage synchronous with the gate pulse.
The caps don't discharge all the way, it can recycle faster too.
Excellent, this is what I was after. Dimmer was a representing either a thyristor or triac but MOSFET and IGBT was being considered but I have never used IGBT. I was also looking to control the discharge of the main charge cap, I am also doing something similar with a high power LED to simulate a xenon tube.

Do you have any circuits?

Thanks,
 

Sensacell

Joined Jun 19, 2012
3,432
Here is what I did- forgive the quick-and-dirty notebook schematic.

I used an external source of triggering pulses.

The strobe lamp is one of those monster coiled glass tubes.
Beware of overheating the tube and capacitor with rapid flash rates!

Flash SCH.JPG
 
Last edited:

Thread Starter

seanstevens

Joined Sep 22, 2009
251
Thats great, thank you very much for that.

Couple of questions:
  • Top left PSU 315VDC output connects (parallel) to the main charge cap bottom right?
  • I take it the zener by trigger coil is to drop the 315V down (minus its own value) down to what?
  • The 7W resistor in series with 315V bridge, that I expect will get pretty hot?
The last point (7W resistor), these resistors appear in almost all circuits, I think as a protection slowing down the inrush to protect components down stream when switched on - they become a source of heat and waste of energy, my target with this kind of circuit and control method is remove these resistors and fully control the power electronically. Have you done anything like that?
 

Sensacell

Joined Jun 19, 2012
3,432
I took apart an old monster photo flash- this was a quick-N-dirty project, hacked into the existing flash unit. The unit had a head with the trigger circuit inside, I added the opto and gate driver to suit my purposes.

1) Yes- the PSU connects to the charge cap as you describe.

2) I don't know the zener voltage off hand, it was part of the original flash-head circuit, I think you could eliminate it.
The zener is there to reduce the voltage on the trigger cap.

3) The resistor gets hot for sure. Some designs use AC type capacitors in series with the bridge input
(along with a small resistance to limit the inrush current ) this idea seems sound, but the ultimate would be to have a switch-mode current source to charge the caps- much more complex, but could substantially increase the system energy efficiency.
 

itit3000

Joined Sep 26, 2016
1
I have used an IGBT in series with the tube to great effect.

You limit the duration of the discharge to control the energy released, the trick is getting the trigger voltage synchronous with the gate pulse.
The caps don't discharge all the way, it can recycle faster too.

Hi Sensacell, I just stumbled across your post. I'm working to build a similar IGBT controlled cut-off for some professional studio strobes. The system I'm modifying uses 600+ VDC to the flashtube and the inrush current at strobe trigger time has been tested around 1000A. My concern as I am designing, is when the IGBT opens shortly after firing, what happens to all that unused voltage/current that the flashtube would normally consume? When the IGBT opens , what keeps from blowing it apart with a flashover?
 

Sensacell

Joined Jun 19, 2012
3,432
The IGBT seems to be standard design practice for strobes.

The excess energy remains in the capacitor, that's the beauty of the concept.

The key it to use a device rated for the application (or maybe multiple devices in parallel) and use a gate drive scheme that is really fast.
Fast gate drive minimizes the device dissipation, did I forget to say drive the gate fast?
 
Top