Laserscope giant pulse question

[weartronics]

We all know that high frequency giant pulses (e.g. using a Q-switch without FPS for blanking) will quickly destroy KTP and/or other optics, but I have a few questions.

* Say the giant pulse has a duration of one microsecond, and a mechanical intra-cavity shutter opens in one millisecond. On a relative timescale, the shutter is opening very slowly. Would the stored energy be released gradually as the shutter opens, thus preventing the generation of a giant pulse?

* Even if a giant pulse is generated by opening such a slow mechanical shutter, is an occasional giant pulse (e.g. once or twice at the start of a show) likely to damage the KTP? I understand that the giant pulse creates a hot spot in the KTP, and the crystal suffers some internal fracture. Can the crystal absorb the energy of one giant pulse from time to time, or is it damaged instantaneously?

[Laserman532]

Weatronics - you understand this more than most - info provided is for others benefit.

In reference to optical damage of the KTP - the crystal will be damaged if the intra-cavity circulating pulse exceeds the damage threshold of the particular crystal (peak power per unit area). Some use cheap crystals, some use expensive crystals so the damage threshold is dictated by the type of crystal used. Grey tracking, coating delamination and surface anomolies are other methods of "crystal damage" which could have nothing to do with giant circulating pulses.

FPS- your shutter scenario should be fine and sounds logical, it may be a bit noisey tho (slow in comparison to your blanking signal and hard on the shutter). The giant pulse scenario generally occurs when a normal operating q-switched laser is operated in "modulation mode" (ie. blanking or external modulation of the RF). Especially when the Pulse Repetation Rate (PRR) is operating at lower frequencies (10khz vs. 25khz for example)

If the RF drive is peroidically interupted, the laser will emit a short high power pulse. In general, the per-pulse optical energy is dependent on pulse repetition interval (PRI). Further, this dependence may be non linear. If the pulsing is suspended for a time which is long compared to the normal PRI, the next laser output pulse may be much larger than the normal train pulse energy. This situation is unacceptable, means must be employed to reduce this first pulse to the standard energy level. In other words First Pulse Suppression.

(information taken from many sources of documents I have)

[DMills]

Yag has a **very** long florescence lifetime (300 microseconds or so IIRC), so you can store a huge amount of energy in the upper laser level if the cavity is not resonating, it is this stored energy that forms the destructive pulse.

Interestingly this 300us figure gives us (using a horribly simplified model) the result that a Q switch operating at say 10Khz (100us between pulses, with the pulses short compared to the PRF) will reduce the pulse size to about 1/3rd of the figure for the saturated gain medium case, and that one operating at 25Khz will push it down to about 1/8th or so.

A Q switch operated at below 3Khz will (abscent careful pulse shaping) cause a giant pulse every time it fires.

These examples ignore the fact that florecence lifetime expresses a timeconstant not an absolute time to saturate (because I could not be bothered to do the math), but they give a feel for the parameters.

One thing to watch is that there will be a lot of gain available as the shutter opens and if the shutter has even slightly reflective surfaces it may allow brief lasing along an axis that would not normally lase.

I have a laserscope which due to needing realignment, lases TEM02 (just, at high current), so you can get odd modes if you are not careful.

I take it you are considering the solonoid mounted on the underside of the wavelength select assembly? Be aware that this ias a 'latching' action due to a permanent magnet, design your driver appropriately.

Regards, Dan.

[Laserman532]

Dan, was my statement correct for the 10Khz vs. 25Khz? I thought I understood it correctly. I do not know the math...only real world operational parameters.

[DMills]

Higher PRF will reduce the peak intracavity power at the cost of reduced doubling efficiency, however turning up the lamp to get the green power back may be couter productive as the average IR will rise for a given amount of green compared to the lower PRF case.

For a given lamp power, higher PRF should be safer, but will make less green, for a given amount of green, things are not quite that simple....

If running CW (Q-switch off) I would expect optical damage due to average power and thermal effects at only a little above 10W or so (Probably the KTP would suffer, being in the beam waist), again wild arsed guess without running the numbers, particularly as the thermal lensing of the rod is so important to that resonator design.

REgards, Dan.

[LesioQ]

I have a laserscope which due to needing realignment, lases TEM02 ...

That would be too perfect , I think You meant something looking like TEM02 ... You see 3 lines in output ??

I have a feeling that intracavity slow shuttering of continuously q-switched z-fold is safe for cavity, but not for the shutter

Piotr.K

[DMills]

Nope, cylindrical TEM02, but it seems to flick the axis of the lobes at random by 45 degrees every few seconds. Really weird.

Personally, I would settle for it doing a reasonable amount of output power at a reasonable threshold (A watt or so at balls to the wall lamp current is not useful), I must get around to sorting it out.

Yea, slow mechanical shuttering is **probably** safe for the cavity, but will be kind of hard on the shutter.

Regards, Dan.

[Laserman532]

15Khz is the recommended PRF

generally CW operation is hard on the OC KTP seems to be no problems (unless it is a crappy KTP then grey tracking phenon will be worse)

[weartronics]

Hi guys, thanks for all your comments. There is lots of useful information here and I'm a bit more confident about my design now. Dan: I read a lot of your posts on alt.lasers and some contributions to Sam's FAQ, so it's good to see you on PL too now. Best regards, weartronics

[Laser Ben]

am thinking a way you could deal with this problem would be to use a longer crystal and move it out of the beam waist. This way you would be able to maintain the conversion efficency while reducing the risk of damage to the coatings due to high circulating power.

This is a common practice on high power scientific DPSS systems, my SP Millennia uses a 20mm LBO crystal and keeps the beam waist outside of the crystal. On this laser, it is vital to fully heat the crystal as it acts as the output coupler, without this, the circulating power causes enough losses in the crystal that it can damage the coatings, however, it I'd not to the damage threshold of the LBO. On my Verdi, I was warned not to risk giving the cavity any pump power as the circulating power will greatly exceede the damage threshold of nearly every component. My friend at coherent told me the most common failure mode of the laser is the head circuitry failing to heat the LBO and distroying the mirror coatings.

I think it would be very interesting to stick a much longer doubler crystal, or perhaps even two of the original KTP crystals in the cavity slightly out of the beam waist where you will get lower conversion efficency per crystal, but higher overall damage threshold.