Sam's Laser FAQ - Solid State Laser Power Supplies (original) (raw)

Sam's Laser FAQ, Copyright © 1994-2021, Samuel M. Goldwasser, All Rights Reserved. I may be contacted via theSci.Electronics.Repair FAQ Email Links Page.

Sub-Table of Contents

Introduction, Basic Structure and Characteristics, Safety

Introduction to Solid State Laser Power Supplies

The major emphasis in this chapter is on power supplies for pulsed solid state lasers. Continuous wave (CW) solid state lasers utilize totally different types of power sources - usually either arc lamps or high power laser diodes. Power supplies for arc lamps share much in common with those for ion lasers. In fact, a unit designed for an arc lamp can sometimes be used to power an ion laser and vice-versa. See the section:Arc Lamp Power Supply for Ion Laser?. Drivers for high power laser diodes and arrays of laser diodes are fancy, expensive, current controlled sources. There is some information on this topic in the chapters starting with Diode Lasers.
### Basic Structure and Characteristics of SS Laser Power Supplies
A typical pulsed solid state laser consists of three parts:
1. Laser head/cavity - This includes the lasing medium (ruby or Nd:YAG rod), flashlamp, reflector, mirrors, Q-switch, etc.
2. Pulse Forming Network (PFN) - This is really the heart of the SS laser power supply and includes the main energy storage capacitor or capacitor bank, and components to control the discharge duration and peak current through the flashlamp (typically at least one inductor though for very small units such as those driving tiny 1 inch rods from disposable camera flash units, these components may be unnecessary.)
The design of the PFN can significantly affect the performance of a solid state laser including the lasing threshold, peak power, pulse consistency, and so forth. There is some general information on PFN design at thePlastic Capacitors, Inc. Pulse Forming Network Page. Some of the basics are covered in subsequent sections below.
3. Capacitor charging unit - A means of charging the main energy storage capacitor(s) in the PFN. These will have both a maximum voltage rating and a power rating in watts or J/second. The voltage rating is obviously critical to achieve full power but also to prevent destruction of the energy storage capacitor(s) by exceeding their voltage ratings. The power rating determines the maximum pulse repetition rate (prr) of the laser (subject to the maximum power ratings of the flashlamp and cavity).
4. Trigger circuit - A means of initiating the discharge of the xenon flashlamp. Three types are commonly used: external (or proximity) and series or parallel:
* External triggering is the same technique used to fire a photographic strobe - a wire or metal structure near or in contact with the flashlamp is driven from a high voltage pulse transformer. The required specifications for the external trigger circuit will be dependent on the physical arrangement (e.g., how close the trigger electrode is to the lamp and how much stray capacitance is present) and the lamp specifications. The required trigger voltage will be somewhat greater for external triggering but this is usually not a major problem since the required current is neglegible. The basic circuit typically consists of a small 300 to 600 V capacitor (.01 to 10 uF), an SCR, and the pulse transformer with a 10:1 to 100:1 stepup ratio. Trigger transformers for/from photographic electronic flash units may be adequate for smaller flashlamps.
* Series triggering puts the secondary of the pulse transformer in series with the anode connection of the flashlamp. The pulse transformer's secondary leakage inductance then becomes part of the PFN in so far as discharge duration calculations are concerned. The basic circuit is otherwise similar to that for external triggering. The trigger transformer may be a separate unit or part of the PFN. (It may also be possible to modify the existing PFN's inductor enabling it to do double duty as the trigger transformer - see the section:Modifying the PFN Inductor for Series Triggering. A typical separate pulse transformer would consist of a ferrite core with a 1 or 2 turn primary and a few dozen turn secondary of thick (since it needs to carry the full flashlamp current) well insulated wire.
* Parallel triggering - Blocking diodes isolate the output of the PFN and trigger transformer secondary. The diode in series with the PFN must be able to handle the maximum voltage of the trigger pulse and the peak current of the PFN discharge pulse through the flashlamp.
External triggering has the advantage of decoupling the PFN from the trigger transformer. The circuit design is simpler since allthough the peak trigger voltage needs to be somewhat greater than with series or parallel triggering schemes, the trigger circuit doesn't need to carry the flashlamp current. However, since the trigger wire/electrode is outside the flashlamp and exposed to the enviroment, triggering may be less reliable and there is some evidence that flashlamp efficiency is slightly lower. For these reasons, commercial pulsed lasers tend to use series triggering. There must be a way of locating a well insulated trigger electrode in close proximity to the flashlamp. Otherwise, series or parallel triggering must be used. Examples include a situation where the cavity reflector and chassis are electrically connected or where there is simply no space for a trigger electrode or no way of adding one easily (e.g., to an existing liquid cooled cavity).
What we call the solid state laser power supply is thus items (2) to (4). Depending on the details, these may be combined into a single unit. However, large pulsed lasers will generally have them separate. Looking through the classifieds sections of the lasers and optics trade rags, you will see many advertisements for "capacitor charging units" since this function can easily be dealt with independently of the rest of the laser.
There is a great deal of practical information on using flashlamps and designing pulse forming networks inDon's General Xenon Flash and Strobe Design Guidelines Page which also includes some basic design equations.
### Solid State Laser Power Supply Safety
WARNING: Even the smallest solid state laser power supply can and will be lethal under the wrong circumstances. This even applies to the flash circuit from a disposable camera that runs off of a 1.5 V AA Alkaline battery!
* Main energy storage capacitor - These may store anywhere from 5 J (W-s) toTHOUSANDS of J at 200 V to 10s of THOUSANDS of V! For anything but the smallest, a lethal outcome is more the norm than the exception should almost any part of your anatomy provide a nice easy path for the capacitor to discharge, particularly if it is through your chest! High voltage protocol must be understood and strictly followed. There are no second chances. The dangers of other power supplies in both lasers and consumer electronics pale in comparison compared to a modest size pulsed laser power supply capacitor bank.
* Capacitor charging unit - Depending on its implementation, portions of this circuitry may be nearly as lethal including line connected (non-isolated) front-ends, 150 to 300 VDC on main filter capacitors, and voltage multipliers.
* Trigger circuit - This will include some components that can give you a painful, if not lethal shock either when just sitting there or when activated.
Some links:
* Additional important safety information regarding shock, excessively bright light, ultraviolet radiation, heat and fire hazards, and other hazards related to xenon flash systems is available from Don'sXenon Strobe and Flash Safety Hints page.
* See the document: Safety Guidelines for High Voltage and/or Line Powered Equipment. Study it forwards and backwards and follow it! To repeat one statement: NEVER ASSUME ANYTHING!
* See the chapter: Laser Safety for general information.
* See the section: Solid State Laser Safety for more details including those dealing with pulsed lasers.
### Interlocks and Firing Controls
Nearly all SS lasers systems will have a variety of interlocks on the laser head and power supply to prevent operation if covers are removed and in some cases, if the laser head isn't in the proper orientation (e.g., a tilt sensor to detect if the head isn't vertical). These will (or should) interrupt primary power to the high voltage portions of the power supply in a fail-safe manner (where a broken wire can't result in the system failing on). They should also rapidly discharge any high voltage capacitors if this doesn't happen automatically when power is removed.
Some systems will also have foot or hand operated "fire" controls. As with interlocks, these must also be designed to not fail on due to a broken wire or loose connector. They will generally include both a normally open and normally closed input and require a transition on both to activate the laser. In other words, just closing or opening a single circuit isn't enough - one circuit must be closed and the other opened within a short time (e.g., using a SPDT switch).
In all cases, with interlocks and firing controls, simpler is better. Relay and switch contacts are more reliable the low level inputs to a microcontroller which can crash. :(

SS Laser Power Supplies Using Photographic Strobes

What Types may be Suitable?

It may be possible to (relatively) easily adapt photographic electronic flash units and 'speed lights' for pumping ruby and YAG rods. Suitable units range from the mini-flashes in pocket 35 or 110 cameras to huge studio systems with multiple flash heads.
The nice thing about this approach is that everything needed for the SS laser power supply is already contained in a well constructed, compact, and low cost (possibly free) package. Flash units for disposable (single-use) cameras are often available for the asking from your local 1 hour developing kiosk or similar place (if you can convince them you won't kill yourself) but you'll probably have to lug away a carton of used cameras (the AA batteries alone are worth the trouble - they usually have a lot of life remaining). WARNING: It is likely that the energy storage capacitors will still be at least partially charged - ouch! Broken external and shoe mounted electronic flash units may be available for little or no cost as well (Often, there is only physical damage to the shoe mount mount and the relevant parts of the unit are fine.) If you know a friendly professional photographer or photo nut who has since moved on to videography, they may have an old large unit gathering dust which you could 'borrow' permanently. :)
For this approach to be effective, the following 3 conditions must be satisfied:
1. The flash energy is sufficient (J or W-s). Figure that the available energy into the flashlamp is about 80 to 90 percent of the calculated value based on energy storage capacitor uF and voltage rating.
2. The flash (pulse) duration is short enough. To be effective, this must be less than the fluorescence lifetime (FL) of the lasing medium. For ruby with its 3 ms FL, this usually isn't a problem. However, Nd:YAG's FL is only 230 microseconds and typical flash units using electrolytic capacitors produce full energy flashes in the millisecond range.
Photographic flash units rarely have any pulse forming components beyond the energy storage capacitor (at least I haven't seen any inductors except as required by energy conserving flash designs). Thus, shortening the output pulse usually isn't an option except by replacing the energy storage capacitor with proper pulse forming network.
3. The shape of the flashlamp/reflector is such that coupling to the rod is efficient. The linear flashlamps used in common small to medium size electronic flash units can usually be adapted easily. However, for larger units using helical or circular flashlamps, it may be necessary to replace the lamp with a linear type and suitable reflector or else most of the light will be wasted.
### Using One or More Disposable Camera Flash Units to Pump a Small YAG Rod
With an optimum cavity configuration, even a single flash unit (typically 160 uF at 320 V for about 8 J) will be more than sufficient to pump a 1 inch rod and maybe even a 2 inch rod (the SSY1 laser with a similar size rod had a threshold of around 7.5 J). Ruby, with its higher threshold, may not be appropriate for pumping with the low budget approach. :)
However, I would suggest remounting the flashlamps and reworking their reflectors into the desired ellipsoidal configuration if possible (relatively easy if they are chrome plated sheet metal) so they can be positioned very close to the rod rather than attempting to collect the light and refocusing it. I disagree somewhat with Chris's comments about the unsuitability of Kodak flash units - there are ways of dealing with controlling multiple flash units and I actually consider the flashlamp not being an integral part of the reflector to be an advantage so that it can be mounted properly. (Note that the newer Funsavers, if they are still made, are probably similar to MAXes; the old ones used a somewhat different design.
See: Kodak Funsaver Flash Unit Schematic, Fuji Flash Unit 1 Schematic, and Fuji Flash Unit 2 Schematicand the descriptions inSam's Strobe FAQ in the PART IV in the chapter: "Schematics for Pocket Camera and Externally Mounted Compact Flash Units".
See the section: Power Supply for Micro YAG Laser (uYAG) for details on this approach using a single flash unit from a Kodak MAX camera.
(From: Chris Chagaris (pyro@grolen.com).)
This set-up may actually be capable of pumping an Nd:YAG rod of small dimensions (2" X 3/16"?). There are a number of different cameras that may be utilized for this experiment. Some lend themselves to simpler manipulation than others. The Kodak 'Funsaver' is the least desirable of the many models you may encounter for a number of reasons. The first being: the flash unit has a logic circuit to initiate the inverter. This circuit is very sensitive and will be rendered inoperative when the units are connected together and triggered, leaving the charging system useless. Secondly; the small xenon lamps and the reflectors are separate entities causing much difficulty in securing them in the appropriate positions on the laser unit. Better results can be obtained with the cameras manufactured by FUJI, which do not have these drawbacks.

Capacitor Chargers

Capacitor Charging Circuits

Unless firing rate/cycle time is critical, this is one area where there are many many solutions using all sorts of circuits which may have been designed for something else.
Basically, anything that can charge a capacitor to a specified voltage (without going over it) will work. At most, a rectifier or bridge will be needed to adapt circuits originally designed for AC (e.g., fluorescent lamp inverter). For the following, even if not mentioned, unless otherwise noted, a voltage multiplier can be used to boost the output up to several times (doubler, tripler, quadrupler, etc.).
Commercial capacitor chargers are usually rated in terms of output voltage and charging rate in joules/second (J/s). For typical home-built systems, a rough idea of the requirements/capabilities can be found by determining the maximum energy to be placed on the energy storage capacitor and the power/wattage capabilities of the circuit or those that are needed based on the desired firing rate. Then derate by a factor of 3 to 5 to account for the exponential charging characteristic of your system (since it probably doesn't have fancy constant current or constant power regulation! For example, if your capacitor charger is rated at 100 W (100 J/s) and you have a 50 J capacitor bank to charge, figure it will take 6 to 10 seconds, not the 2 seconds that would be nice. :) However, home-built capacitor chargers are often way overdesigned (or gross overkill if you prefer) considering that the ultimate limitation on cycle time is likely to be cooling of the cavity components, particularly the flashlamp.
Here are some possibilities for capacitor charging circuit:
* Battery or low voltage DC powered capacitor chargers.
* Up to 350 V, up to a few J/s: Inverters from pocket camera and external flash units. Some of these will simply not work if modifications are made to the output circuit (like a voltage multiplier). The only way to know for sure is to try it! Especially useful are the flash units from single-use (disposable) pocket cameras because they are often free! Of course, the energy storage capacitor and trigger circuits may be useful as well.
* Up to 1 kV or more, up to a few dozen J/s: Fluorescent lamp inverters, flyback based high voltage generators, etc.
* Up to 1 kV, up to a few hundred J/s: Automotive, RV, and marine power inverters (12 VDC in, 115 VAC out) with voltage multipliers added to their output. For squarewave types, a voltage tripler will produce about 350 V, a quadrupler about 460 V, and so forth. For sinewave or modified sinewave types, a doubler will result in about 300 to 350 V, a tripler about 450 V, etc. For 220 to 240 VAC output, fewer multiplier stages are required (and/or greater output voltage possible. For 300 to 350 VDC to capacitor, no multiplier is required for sinewave and modified sinewave types).
* AC line powered capacitor chargers. All of these can be fitted with voltage multipliers to boost output beyond what is listed.
* Up to 1 kV, up to a few hundred J/s: Direct connection to line using voltage multiplier. WARNING: No line isolation as if this matters where the energy storage capacitor alone can kill you a day later! AS above, fewer multiplier stages and/or higher output from 220 to 240 VAC line input.
* Up to 1.5 kV, up to a few hundred J/s: Power transformers from vacuum tube type audio amps and old TVs.
* Up to 5 kV, up to 1 kJ/s or more: Microwave oven transformers.
* Up to 22 kV, up to several hundred J/s: Neon sign/luminous tube and oil burner ignition transformers.
* Up to 500 kV, up to several MJ/s: Pole pigs and substation transformers wired up backwards. :)
Relevant circuits that will work directly or with minor modifications can be found in:
* Sam's Strobe FAQ
* Sam's Schematic Collection
* The chapters: Complete HeNe Laser Power Supply Schematics, The Home-Built Laser Assembly and Power Supply, and others in this document.
### Sources for Capacitor Charging Circuit Components
See the chapter: Laser and Parts Sourcesfor all sorts of suppliers of everything needed to construct the capacitor charger (and trigger circuits). The other chapters listed above may also list sources for specialized items like high voltage diodes, resistors, and capacitors.
### Recycled Defibrillator as Capacitor Charger
The following may come in handy when building that next high speed high power strobe or pulsed laser.
WARNING: Defibrillators are at least as good at stopping beating hearts as restarting misbehaving ones. The charge in their energy storage capacitor (typically 300 to 400 joules) is enough to kill a half dozen healthy adults instantly. The operating voltage (up to 5 kV) doesn't respect common wire insulation and can jump 1/4" or more in air. There are no second chances.
(From: Steve Roberts.)
Older defibrillators are now showing up as inexpensive surplus because their ancient edmark waveform is being replaced with newer computer controlled biphasic waveforms.
So what do you get in a typical edmark waveform defibrillator:
1. Switching HV supply up to 5 kV with programmable shutoff voltage.
2. 32 uF Maxwell energy storage capacitor rated at 5 to 6 kV.
3. 47 mH inductor rated for the above cap's current.
4. A high pressure gas (SF6?) or BIG vacuum glass relay.
5. A second smaller vacuum relay.
6. A 50 ohm resistor for bleeding the cap and simulating a patient's chest.
7. Usually a vectorscope with an amber phosphor CRT and a GM20 or similar galvanometer if the unit has a chart recorder.
Notes: The relay is usually a 5 kV 50 A DPDT which has a short across one set of contacts to protect the patient. The other set of contacts goes to the capacitor common leads and to the patient via the paddles. So, presto! - apply 12 volts to the relay and you get up to 360 joules dumped into the victim or patient via the inductor to control the waveform. A patient's chest is assumed to be about 50 ohms impedance via the conductive cream to the paddles so the test circuit monitors what happens when the second smaller relay dumps the cap into the 50 ohm air cooled test resistor. The cap is also dumped during power-down.
I can't overstress the absolute need for safety when handling a 33 uF 5 kV capacitor. Newer defibrillators have a MOSFET H-bridge for bipolar switching and only go to two kV with smaller caps.
### Electrophoresis Power Supplies as Capacitor Chargers
Relatively inexpensive electrophoresis high voltage power supplies are popping up on eBay. Search for "electrophoresis". Depending on model, these may be capable of producing up to 2,500 VDC or more with a current capability of a few mA to 100s of mA or more. They are often fully adjustable and protected against faults. However, isolating the energy storage capacitor with a current limiting resistor is a good idea.

Energy Storage Capacitors

Calculating the Required Capacitance

The equation that should be engraved on the inside surface of the front of your skull is: E = 1/2 * C * V2. :) Given a desired flashlamp voltage, this tells you how much energy (in joules or watt-seconds) is available from a certain uF value of capacitance. For all intents and purposes, nearly all the available energy will be deposited in the flashlamp even for a less than optimal circuit.
So:
2 * E
C = -------
V2
To obtain this uF value and voltage rating may require a capacitor bank consisting of a series/parallel combination of multiple capacitors in an array. This creates problems of its own in guaranteeing that voltage is (nearly) equally distributed or balanced among multiple series capacitors. See the section: Series Banks of Capacitors. Such issues are particularly critical for the rapid discharge circuits being discussed here.
Also note that normal vanilla flavored electrolytic capacitors may self destruct if called upon for this service. Or, they may have too much inductance (increasing pulse duration) or too much resistance (resulting in excessive losses inside the capacitor(s). Electrolytics are probably adequate for ruby pumping with its long fluorescence lifetime (3 ms) but would be marginal for Nd:YAG (230 us) with a desired 100 us pulse duration.
### Sources for Flashlamp Rated Capacitors
WARNING: I can't over-emphasize the risk of instant and total electrocution with even small flashlamp capacitors. Where you have acquired some really large ones - say 12" x 18" x 8" weighing 60 pounds each - you may be better off selling them - you'll get a lot of nice cash - and building a power supply with something just slightly less lethal! Each one of those caps will kill dead-dead a few dozen or more people after sitting with the power off for a week! Seriously, they are deadly and should be treated with the same respect as an equivalent size block of sticks of dynamite (or C4 if you prefer) wired to detonators sitting out in a thunderstorm. :)
For long duration pulses, ordinary high quality electrolytics capacitors may work without self destructing and/or dissipating too much energy inside the capacitor. However, for high-joule flashlamps operating with short pulse duration, you won't get away with this for either or both reasons. No, wiring 10,000, .01 uF, 1 kV disk capacitors in parallel probably isn't a very viable solution! :)
If you want to get an idea of what the "big boys" use, check out:
* HiVolt Capacitors, Ltd. Product Selection Guide
* General Atomics (was Maxwell) Energy Products
High quality motor run capacitors should be fine for most long to medium pulse length applications including flashlamps for ruby and YAG lasers. A motor run capacitor rated for 370 VAC is *probably* good for at least 500 VDC. In addition to simply the peak of the AC being 1.414 times the RMS, there are corona and other effects in the dielectric which limit the maximum reliable AC working voltage, resulting in a more conervative AC rating.
Aerovox High Voltage AC and DC Capacitors seems to show a relationship of between 1.75 and 1.82 (take the link to the datasheet).
Robin says below that it's much higher but I'm not sure this applies to all types.
(From: Robin Bowden.)
I have found that motor run capacitors are quite suitable. They are made from metalized polypropylene and there are models available up to at least 440 VAC.
If you look at dual rated AC/DC metallized polypropylene caps - then a cap rated at 400 VAC will likely be rated at 900 VDC. So I guess these 440 VAC capacitors are good up to 1 kV.
These are readily available and are relatively inexpensive. Here in the UK, a 40 uF (top end of available values) costs around 5.75 pounds (less than $10US).
I have a bank of three of these in parallel and use them to just over 900 VDC (50 joules).
Looking at the manufacturer's datasheet the LCR MRP series 40 uF caps (the ones I have), tan delta of less than 0.001 at 50 Hz. This tan delta figure is the worst case for all caps in the series. For a 40 uF cap this gives ESR of less than 80 milliohms.
A metal cased Arcotronics one stated tan(d) to be less than 0.002 (160 milliohm) but this incorporates an overpressure disconnect device (wire that breaks when the metal can deforms due to over-pressure). I guess this is responsible for most of those extra milliohms.
Today I took one of those caps into the lab to measure. Real world figures for the 40 uF LCR MRP series:
* Capacitance: 38 uF.
* Self resonant frequency: 78 kHz.
* ESR measured at self resonant frequency: 15 milliohm.
* ESL derived from self resonant frequency: 0.11 uH.
The caps are of wound construction with the termination in bulk from each side of the roll (not start and end of winding) I guess this explains the low ESR and ESL I don`t know how this compares to a real laser cap but it looks reasonable to me.
The datasheet states a maximum dV/dt of 20 V/us. I'm not sure how far you can push this - I assume it is to prevent overvolting the dielectric due to excitation of the self resonance rather than a maximum current issue.
(From: Don Klipstein (don@donklipstein.com).)
Sprague (a division of Vishay) 36DX and probably the similar Durocap/Mallory CG, CGH, etc., work fairly well for discharges that take a few hundred microseconds or more, and work well for discharges that take at least a millisecond. With a 1 millisecond discharge, maybe a couple to a few percent of the energy is dissipated in the internal resistance of the capacitor. (Note: Duracap International, Inc.is the actual manufacturer of the Mallory product line now.)
For shorter flashes, the Sprague TVA series (axial lead) is a little better than electrolytics that have both leads on the same end. But for flashes under about 100 microseconds, I know that the usual electrolytics may not be the way to go.
(From: Boris Mohar (borism@interlog.com).)
And then there are also: ECI Capacitors, Inc.Unlytic UL30 film capacitors.
They are packaged in the same form factor as electrolytics, computer style 3" diameter cans. Voltage range is from 300 to 1,000 WVDC (!!) with capacities from 50 to 1,600 uF. These are not electrolyics (I guess you figured that out, huh?) so they are non-polar. Impedance dip of 0.01 ohms or so is in the 30 to 100 kHz range. They are specified to handle thousands of amps peak currents.
I dare not ask the price.
(From: David Linsley (david.linsley@baesystems.com).)
Norfolk Capacitors Limitedis a UK supplier of all types of capacitors including those suitable for pulsed lasers. See Norfolk Capacitors Limited Energy Discharge Capacitors.

Pulse Forming Network Inductors

Calculating the Required Inductance

As a first approximation, the specific value of inductance in the pulse forming network will be determined mostly by the desired pulse duration from the flashlamp and the value of the energy storage capacitor(s). Assuming a critically damped system, the pulse duration is approximately: T = 3 * sqrt(L * C) which is close to 1/2 period of the resonant frequency of the LC combination. So, given a fixed uF value for the energy storage capacitor and desired pulse duration, L = T2/(9 * C). This assumes critical damping - part of the advanced course. :)
Whether a system is critically damped, under-damped, or over-damped will depend on how the resistance of the discharge path (capacitor, inductor, flashlamp, and wiring) compares to the reactance of the LC network. However, it may not be possible to achieve both the desired pulse duration and critical damping.
Note that with a critically damped or over-damped system, there will be no undershoot which is the desired state of affairs. Undershoot damages flashlamps. In any case, it is a good idea to add a reverse biased high current diode across the output of the PFN. This will conduct if there is any undershoot bypassing the current around the flashlamp.
As an example, the energy storage capacitor in PFN1 (see the section: spspfn1">Pulse Forming Network 1) is 36 uF and for a Nd:YAG laser, a pulse duration less than 230 us (the fluorescence lifetime of Nd:YAG) is desired. The 30 uH inductor then results in: T = 3 * sqrt(3*10-5 * 3.6*10-5) or about 100 us.
### Constructing Pulse Forming Inductors
These typically need to have a value between a couple of microHenries and 100 uH. While commercial inductors are available at inflated prices, in most cases, it is easy to construct your own from a ferrite core and a few turns of wire. While the peak current through the inductor may be several hundred amps, unless you are operating at a high repetition rate, you don't need to construct the inductor from #0000 AWG wire - #16 to #20 will do fine as long as its total resistance is small compared to that of the flashlamp and ESR of the capacitor.
While there are messy equations to calculate inductance based on number of turns; core type, style, diameter; number of layers, packing, etc. The easiest may be to construct a, say 10 turn, test coil and perform a 'ring test' to find its inductance. Then, scale as appropriate for your needs. Inductance is a squared function of the number of turns where the packing doesn't change much. Err on the high side and then remove turns to tune it precisely.
The 'ring test' just uses a pulse generator (almost any type) to excite the parallel combination of a high quality capacitor (I use polyester capacitor around 1 uF) and your test inductor. This circuit will resonate at:
1
F = ----------------------
2 * pi * sqrt(L * C)
So:
1
L = -------------------
(2 * pi * F)2 * C
With a high quality capacitor, each pulse from the pulse generator will result in many cycles of a decaying sinusoid - enough to accurately measure frequency or period.
(From: Chip Shults (aichip@gdi.net).)
I usually wind inductors out of 18 gauge magnet wire for this, and the resulting current doesn't melt them because it's so brief. I use a ferrite about 3/4" in diameter. 60 turns gives you about 35 uH with a single layer coil.
(From: A. Nowatzyk (agn@acm.org).)
For the 3" x 1/4" Q-switched ruby tank range finder, I used two 100 uF Maxwell caps (low ESR for pulsed application) and a 120 uH pulse forming inductor. The measured T1/2 is 200 us, peak power is 1.8 MW, peak current is 2500 A and peak discharge voltage is 700 V (caps were charged to 1000 V. The Q-switch lasing threshold is about 690 V.
The inductor was wound from 7 strands of 19 AWG magnet wire that were twisted into a round bundle with about 2 twists/inch. A linear air-coil works fine, but the stray field during discharge is very annoying to any nearby electronic. A toroidal inductor (air-core!) should work better. Forget any core-material other than air: At these currents, anything saturates and L will be much less that you would expect.
### Modifying the PFN Inductor for Series Triggering
Where you have an existing PFN but no means of triggering the flashlamp, it may be possible to combine these functions with the PFN's inductor iff:
1. The inductor's insulation is adequate to handle the high voltage trigger pulse and there are no components other than the flashlamp attached to its output. If there are any (like a reverse polarity prevention diode), their ratings would have to be adequate to handle the trigger pulse. The typical required trigger pulse voltage will be greater than 10 kV, possibly as much as 30 kV or more. And, should there be any capacitors on the output, this probably won't work since the pulse would be filtered out!
2. The number of turns on the inductor can be determined so you will know how many primary turns to use and what voltage to use on the trigger cap based on the triggering specs for your flashlamp. The number of turns can be determined by adding a 10 turn test primary to the inductor's core, driving it at several kHz from a low level source (like a signal generator), and measuring the voltage on the original winding. This will give you the turns ratio. For triggering, a 1 or 2 turn primary is sufficient if the required trigger cap voltage isn't excessive (e.g., 300 V or less).
The trigger primary winding must be connected to the trigger cap discharge circuit such that the resulting pulse increases the voltage on the flashlamp.
### Lou's Comments on Replacing the M-60 Inductor/Trigger Transformer
The following was prompted by a request for information on what to use in place of the original potted combination inductor and pulse transformer used in the PFN of the ruby laser in the M-60 tank rangefinder.
(From: Lou Boyd (boyd@fairborn.dakotacom.net).)
I did some tests on one of the M-60 ruby laser flash injection transformers and was somewhat surprised by the results. I hadn't worked with one before. Here's what I found:
The main winding, when placed in parallel with an 88 uF 5% capacitor (had one handy), resonates at 415 hz at a drive level of about 5 volts. That indicates an inductance for that winding of 1.68 mH when there's no core saturation. I was surprised that it was that large, I was expecting more like 200 to 500 uH. The more surprising part is that the resistance of that winding is only 0.052 ohms (DC). My ohmmeter just showed a short so I used a bridge. I couldn't access the other winding of mine directly since it's potted, but when driven through the .22 uF trigger capacitor at the assembly's resonant frequency of 545 khz (nothing attached to the secondary except a 10M scope probe) the step up was about 16 times. Dumping 500 V with the SCR should produce at least 7 kV to initiate the flash. The cap is rated 600 V. I have no idea at what voltages/rise times the core starts to saturate.
To wind an air core inductor of 1.68 mH inductance and get .05 ohms resistance would create a monster. A quick calculation says it would take about 100 turns of 12 gauge wire wound on a 6" cylinder using three layers. The permeability of the core on the inductor you have gives a dramatic size reduction. If you can dissect the one you have and rewind it I recommend doing so. My guess is that it's in the order to 50 to 100 turns of something around #20 gauge wire. Count them and measure the size of the wire if you take it apart. I don't have any idea what goo is holding it together. If it's an Epoxy it may not be practical to open it.
I presume you're certain the inductor/transformer is bad. Seeing what looks like a dead short with an ohmmeter on the output winding is normal. If there's an open between the heavy orange and heavy clear lead that's bad. If those lead have a DC path to ground the .001 uF, 5kV cap or the inductor could be bad. The only practical way to check for a shorted winding is to measure it's inductance. The resistance difference for a 1 turn short would be less than a milliohm.

Pulsed Solid State Laser Design Resources

Rotorwave Technical Section

There used to be a company called Rotorwave with a Web site at http://www.rotorwave.com/ and technical resources at http://www.rotorwave.com/technical.htm. They had information related to pulsed solid state lasers including case studies/design examples, programs and spreadsheets to assist in laser design, specific information for pulsed holography, and more. Their Web site had always been unreliable but now it seems to have disappeared for good.
One of the useful resources was an MS Excel spreadsheet that implements calculations for a variety of laser design issues including resonator parameters, power supplies, and pulse forming networks. It was on their Web site under "Ruby Laser Designs" named "weights.xls". I have an archived copy at: Rotorwave Laser Design Calculator Spreadsheet. (If someone complains, I may have to remove this file but will happily add any requested authorship or copyright information to it, or link to the original Web site. The only reason I copied it was that the Rotorwave Web site had been somewhat unreliable and the spreadsheet seemed to be particularly useful. Now that it's gone complete, that was probably a wise move.)
The organization of the spreadsheet is somewhat haphazard but there are a variety of useful equations solved for specific parameters. Cut and paste the relevant lines and drop in parameters for your laser and the result may be useful.
CAUTION: I have not verified the accuracy of all of these equations. So use at your own peril. I will not be responsible if the Universe should implode when you fire your laser for the first time! :)
#### Doug Little's PFN Calculations Spreadsheet
This has the various PFN equations implemented in MS Excel providing a convenient way to experiment with variations. Get it from Doug Little's PFN Calculations Spreadsheet (Version 1.1).
(From: Doug Little.)
This updated version has been used to build a working ruby laser and appears to check out reasonably well in the 'real' world. In fact the calculations one way or another brought me to within a few percent of actual performance, which I think is not bad given that some of the efficiency figures are approximate.
The spreadsheet predicted my laser should threshold around 1,350 V at 633 uF, and it actually took place at 1,450V at 633 uF. This factors in the rod dimensions, reflector, lamp spec, lamp current, etc. So not too bad. :)
### Handy Little Programs
The following C programs were contributed by: Eddie Kovelan (kovelan@austin.ibm.com.) I (Sam) did some editing and cleanup. :) Also see the section:Useful Equations for Flashlamp Selection and PFN Design upon which some of these programs are based.
While tested under unix, (Sun OS and freebsd), they should compile under almost any C environment since there is definitely nothing fancy about anything and your minimal Turbo C should be fine! Just save the file and compile it with the command line shown in the program header.
#### Energy Stored in a Capacitor (joules.c)
This program returns the total energy stored in a capacitor of specified value (uF) and operating voltage (V). Note that this will be somewhat more than actually can be delivered to a flashlamp.
* Download joules.c
#### Capacitor Value for a Desired Energy (flashcap.c)
This program returns the required capacitance (uF) of an energy storage capacitor given the desired total energy (J) and operating voltage (V).
* Download flashcap.c
#### Inductance for a Pulse Width and Capacitor Value (inductor.c)
This program returns the required inductor value (uH) for a given pulse duration (us) and capacitance (uF).
* Download inductor.c
#### Pulse Duration of a PFN (pulse.c)
This program returns the pulse duration of a series LC pulse forming network given the inductance (uH) and capacitance (uF) assuming that the system is critically damped.
* Download pulse.c
#### Capacitance, Voltage, and Inductance of a PFN (pfn.c)
The following program returns a capacitance (uF), voltage (V), and inductance (uH) for a series LC pulse forming network given the the desired total energy (J or W-s), pulse duration (us), and flashlamp impedance, Ko (ohms), assuming the system is critically damped.
* Download pfn.c
#### Inductance of an Air-Core Coil (coil.c)
This program returns the required number of turns for an air-core coil of specified inductance (uH), radius (inches), and length (inches). Note that this is an approximation assuming the length is more than 0.8 times the radius.
* Download coil.c

Sam's Laser FAQ, Copyright © 1994-2021, Samuel M. Goldwasser, All Rights Reserved.
I may be contacted via theSci.Electronics.Repair FAQ Email Links Page.