Need a help about Xenon flash tube

[Warpspeed]

The supply authorities do not like direct half wave mains rectification.
Just one of the problems is that dc in the neutral line from similar loads spread across three phases does not cancel, but the resulting return dc current adds.

This can quickly burn out the neutral line, as its not commonly protected with fuses or circuit breakers.

Normally with three phases, and three thirty amp ac loads, the neutral current would theoretically be balanced resulting in zero neutral current.
Three thirty amp half wave rectifiers might put very roughly 90 amps dc average back into the neutral that may only be rated to carry thirty amps, and nobody would know until the fire starts.

I have a very old German "Mayer" strobe that uses that same curly tube.

This works from 230v through a bridge rectifier and a 2K2 twenty watt resistor into 2uF flash capacitor. That works on the highest frequency range from 0 to 270 Hz flash rate and 0.1 Joules.
Middle frequency range switches in an additional 4uF (6uF total) for 0 to 60 Hz at 0.3 Joules.
Low frequency range switches in 40uF (46uF total) for 0 to 15 Hz flash rate and 2.36 Joules.

Highest energy is on the slowest range, 15Hz times 2.36 Joules which is comfortably within the 40 Joule flash tube rating.
Pushing it much harder than that hardly seems worthwhile.

 

[Dan Mills]

That and the problems caused by triplen harmonics which also sum in the neutral (and are usually a much bigger problem, due in part to the widespread use of phase controlled dimming).

None the less, the Dataflash and martin Atomic strobes are a common fixture at most large music stages, and some of these will actually also do a 'Blinder' mode where they fire on every half cycle for up to a few tens of seconds, supply current can hit 30A or so average per unit and you cannot use magnetic breakers because the peak currents will cause instant tripping....

The limit is wall loading on the tube at least as much as it is available energy, and they run quite close to the explosion limit for an air cooled lamp.

Regards, Dan.

 

[c_mitra]

None the less, the Dataflash and martin Atomic strobes are a common fixture at most large music stages, and some of these will actually also do a 'Blinder' mode where they fire on every half cycle for up to a few tens of seconds, supply current can hit 30A or so average per unit and you cannot use magnetic breakers because the peak currents will cause instant tripping.... .

30A average per unit is really very high. The 30A is a peak rush current that lasts only for a few milliseconds per cycle. 30A for one half cycle is 30*220*0.01=66J per flash and if we count 50 flashes in one second that becomes 66*50W=3.3KW, large but not uncommon. But the tube will be real hot (they are glass, not quartz) and the pressure can be very high (I do not know the exact value). If you count 30A per cycle for a few ms it takes to charge the capacitor, the flash output may be perhaps 1J (more realistic value) and professional photographic flashes (they fire only once in a while) give around 10J (mostly less).

 

[Dan Mills]

3kW average in blinder mode is not unreasonable, but it is more like 1/4 cycle pulses (You cannot ignite the lamp at zero crossing!) and 100Hz not 50Hz.....
These tubes are of the order of 30cm long, and I would not be at all surprised if martin were using Quartz in that application to get a high enough wall loading.

The few tens of seconds in blinder mode is a thermal limit enforced by the control logic keeping track of a thermal model of the tube.

These are not your capacitor discharge photographic strobes, which are low energy but short pulse.

 

[FvM]

None the less, the Dataflash and martin Atomic strobes are a common fixture at most large music stages, and some of these will actually also do a 'Blinder' mode where they fire on every half cycle for up to a few tens of seconds, supply current can hit 30A or so average per unit and you cannot use magnetic breakers because the peak currents will cause instant tripping....

The limit is wall loading on the tube at least as much as it is available energy, and they run quite close to the explosion limit for an air cooled lamp.

Do you say these devices are feeding the xenon tube current directly from the mains like the post #11 "capacitor less" circuit, or do they use classical storage capacitors?

I see that the capacitor charge circuit can still cause high mains current peaks, but it has a certain chance to comply with power quality standards, e.g. avoiding DC current which is absolutely prohibited.

If its continuously fed from the mains, as this circuit is, the arc will be continuous.
As there is no current limit, the arc will grow until it fills the tube and bursts the glass envelope.

An arc does not always extinguish at the zero crossings, as ac arc stick welders weld continuously even when the arc polarity goes through zero and reverses.
Xenon tubes will also do that.

That's why the "capacitor less" circuit uses a half-wave rectifier. It will surely extinguish the arc.

Even a straight xenon lamp a foot long will discharge a capacitor bank down to only a few volts.

Typically 50 to 100V if I remember right.

If you fire a xenon tube from a very high voltage trigger, usually at least 4Kv, the tube conducts and only a fairly low voltage is required to sustain a continuous arc discharge.

The discharge caused by the trigger transformer has too low current to ignite the plasma completely. Having a small capacitor charged to a voltage above the minimal anode voltage and a diode circuit switching to the low voltage, high energy source will however do. The post #11 circuit is implementing this method. It can be used to avoid switching of storage capacitors with strobe frequency although the storage capacitor voltage falls below the minimum anode voltage.

But the tube will be real hot (they are glass, not quartz)

High power xenon tubes (e.g. > 100 W) are often made of quartz/fused silica. Average, pulse power and current ratings have to be observed nevertheless.