A quote prices the cabinet. It almost never prices the parts that get used up while the cabinet earns.
Lamps, rods, doubling crystals, windows, handpieces, arms. Quotes call them consumables; service departments call them wear parts. Same list, and the distinction that matters is whether a part is metered by a counter or replaced on damage. The wear side runs on a different clock from the capital side, and most of it sits outside your warranty on purpose. This piece stays on those parts alone. For the wider picture of freight, duty, registration and service contracts, the ownership cost breakdown covers that ground.
Wear Is Counted in Discharges, Not Minutes
Start with what damages a flashlamp. Every discharge erodes the electrodes and throws a trace of that material onto the inside of the quartz envelope. Repeat it enough and the wall darkens, transmission drops, and the rod sees less pump light for the same charge. The RP Photonics encyclopedia entry on flash lamps sets out those mechanisms plus one figure worth carrying: lamps run well below their explosion energy, the single-shot energy that shatters the envelope, can reach more than a hundred million pulses. Drive the same lamp hard and it dies in a small fraction of that.
Same part. Two very different lives.
So the physical unit of lamp wear is one discharge. Hours are a conversion somebody made by assuming a repetition rate. The consoles know it even when the datasheets don't: our Q-switched platform manuals describe a counter with two lines, a seven-digit lifetime total and a per-session count that resets at power-off. The lamp spec on the same machine talks in hours. Somebody divided.
The rating is a curve, not a number
Here's the clearest example in our own archive. Service documentation for an archived e-light platform in our service library, not a current catalogue model, rates the pulsed lamp inside the treatment handpiece at roughly 60,000 flashes. The same document then says that in hair-removal work, which runs the top of the parameter range, the head stops producing the intended result at around half that count, and it tells the operator to raise energy to compensate. Take those figures as one data point on lamp-in-head architecture, not a spec for anything we ship now. That architecture is still standard on current builds like the MF-05, so the pattern travels even where the count doesn't.
Lamp life isn't a property of the lamp. It's a property of the lamp plus the fluence you need out of it. A pigment room working gently and a hair-removal room working hard burn the identical part at rates that aren't comparable, and the second room hits the wall long before the number on the sheet.
Questions that turn an hours figure into a budget line
- At what pulse energy and repetition rate was the rated life established?
- Does the rating assume simmer current stays on between shots, or does it only count discharges?
- Is there a shot-count equivalent, and will the supplier put it in the contract?
- What's the acceptance criterion for worn out? A stated percentage of original delivered energy is checkable. An operator's opinion isn't.
- Can a trained technician on your side fit the replacement, and does opening the cabinet void anything else? Ask it as a double question: will the maker supply a written service procedure covering capacitor discharge and interlock handling, and is the machine delivered against IEC 60601-2-22, the particular standard for surgical, cosmetic and therapeutic laser equipment?
The Rod, the Doubler and the Windows Age Separately
A laser rod sits in a cavity full of ultraviolet and thermal cycles on every pulse. End coatings and the pump reflector age alongside it. None of it announces itself. What you see is a machine needing more from the lamp for the pulse it used to give for free, which reads on your side as a lamp problem when it may not be one.
When the green fades first, suspect the doubler
Dual-output cabinets make their 532 nm line by frequency-doubling 1064 nm through a nonlinear crystal, and that crystal has its own failure mode. Boulanger, Fejer, Blachman and Bordui reported in Applied Physics Letters in 1994 that gray-tracking damage during 1064 nm second-harmonic generation in KTP depended on the intensity of the 532 nm light itself. The crystal producing your green is the part most exposed to it.
Practical version. On a dual-wavelength platform like the QN-03, if superficial 532 nm work starts underdelivering while 1064 nm work still lands the way it always did, the doubler and its coatings belong at the top of the suspect list, ahead of the lamp. Different part, different invoice, different lead time. Diagnose it as a tired lamp and you've bought a lamp and fixed nothing.
A window fails in three ways and they cost three different amounts
Contamination sits on the surface and comes off, and the part returns to spec. Coating damage, meaning a pit, a burn or a scratch, gets replaced; our manuals are blunt that a damaged lens goes back to the factory rather than getting polished, because polishing changes the surface you were relying on. Bulk damage inside the substrate is the same answer with a worse mood.
The Console Reading Is a Request, Not a Measurement
Two measurement studies matter here. Town, Ash, Eadie and Moseley measured 18 intense pulsed light systems in clinical use across England and Wales and found discrepancies between measured output and the values those systems displayed or their manufacturers claimed (Journal of Cosmetic and Laser Therapy, 2007). Lister and Brewin ran seven cutaneous laser systems through simulated clinic lists and reported that pulse-to-pulse output energy varied considerably between a representative test patch and a full treatment, and across a whole list (British Journal of Dermatology, 2014).
You can't see ageing on the display. You see it with a meter, or not at all. And ageing hides inside ordinary output variation for a while, which is exactly the window in which a room quietly starts working at higher settings without anyone deciding to.
That last bit stops being a procurement question. Chasing a fading lamp with fluence is how post-inflammatory hyperpigmentation and blistering enter the conversation on darker phototypes. Fix the part. Don't outrun it with energy.
Handpieces and Arms Sit Outside the Warranty for a Reason
They're the only components a person holds. Handpieces meet trolleys, get dropped, and get pulled by the umbilical when someone reaches across a couch. Warranty schedules reflect that. On lamp-in-head builds it's a glass envelope centimetres from splashing pigment. A diode head has no envelope, yet the exclusion survives in another form: thermal cycling on the emitter stack, an abraded sapphire contact window, coolant fittings, all at the operator's end.
That archived e-light service documentation shows how field triage actually runs. First step for a dead head is looking for obvious impact marks, then looking inside for a cracked lamp. Both steps assume the machine is off and discharged to the maker's written procedure, and they stay at the level of a look: opening a cavity or touching the optical path is the maker's job, not yours. A head that leaks means an internal water line or the quartz tube has gone, and that one ships back rather than getting opened on your bench. Best trick in the document: if a console drives two heads and one lights while the other stays dark, the simmer board is sound and the fault sits in the dark head.
On arm-equipped builds such as the QE-01 electro-optic cabinet, our Q-switched documentation carries a similar warning. The seven-joint arm ships as a separate item, quoted net weight excludes it, and it's fitted on site. Once the path is aligned, the manuals say leave it alone: no dismantling, no re-adjustment, keep the gun clear of impact. Mirrors inside an arm drift and their coatings age on their own schedule, and a serious knock can undo an alignment nobody in your building can restore. The QN-03 drops the arm entirely, which deletes this whole wear line.
What the commercial question leaves out
Warranty is only half of who-opens-the-cabinet. A flashlamp-pumped machine stores its pump energy in a capacitor bank that stays charged after the mains go off, and Berkeley Lab's non-beam hazard guidance is explicit that capacitors are discharged, shorted and grounded before anyone gets access to that area. Lamp and capacitor housings are also built to contain a disintegrating component, which is a fair description of what a quartz envelope can do. In a service state interlocks may be bypassed, putting whoever is at the cabinet inside a Class 4 direct-beam risk, so the laser safety management IEC 60825-1 underpins still applies, eyewear included. Optical density is specified per wavelength: 1064 nm eyewear is not 532 nm or 10600 nm eyewear.
What to settle before you sign
- Is the handpiece a warranted assembly, or a consumable line item on the quote?
- Is the lamp inside it replaceable, or does the whole head return?
- Is a loaner head available while yours is away, and who pays freight each direction?
- Does each head carry its own counter, so you know which one you're running down?
- What exactly counts as misuse, in writing?
Change the architecture and the wear moves somewhere else. The DL-07 diode platform has no flashlamp at all; our brochure archive quotes more than 20 million shots for the diode hair-removal handpiece, with sapphire contact cooling on the skin side. The emitter stack is the wear part there, and it generally isn't a field swap. On CO2 the tube decides everything: our catalogue archive contrasts glass tubes rated in hundreds of hours against RF-excited tubes rated in tens of thousands, at a higher price up front. That trade is the entire case for the sealed RF source in the CF-01.
Turning Wear Parts Into a Per-Treatment Number
You don't need the supplier's cooperation to do this properly. Six steps and a counter.
- List every wear part on the machine and get a price for each in writing, freight included.
- Get a rated life for each part, in shots where possible, hours where that's all they'll give you. If all you get is hours, convert: shots equals rated hours times the repetition rate the rating assumed, in Hz, times 3600. Demand that rate in writing, because without it the hours figure converts to nothing. And the conversion only describes your room if your working rep rate matches theirs; run slower and you consume the rated hours faster in calendar terms while burning the same shots.
- Derate it. Budget to the point where output no longer reaches your working fluence, not to the point where the part dies. That archived e-light document puts those two points a factor of two apart on demanding work.
- Measure shots per session off the counter across a week of real treating. Not an estimate. The counter.
- Per part, cost per treatment equals price divided by derated life in shots, multiplied by your shots per session. Sum across parts.
- Add the count-blind items as a flat annual allowance: windows, tips, filters, eyewear. Those get replaced on damage rather than on a count, so traffic and staff numbers drive them.
Before you chase every line, note where the money usually sits. On lamp-pumped platforms the head or its lamp dominates recurring spend; the rod and doubler are rarer but arrive as single large invoices; windows, tips and filters are frequent and small. On diode and RF builds that inverts: almost no consumables recur until the emitter stack or tube reaches end of life, and that one item is the whole bill.
Only now do quotes compare. Two cabinets can carry the same sticker and diverge hard by year two, because one of them wants its lamp twice as often at the settings your diary actually demands.
What this does to your session price
Clinics that have never run this number treat discounting as free. It isn't. Every extra pass to an endpoint is wear you already bought, and a protocol needing more passes costs more per result even when the machine was cheaper. It's also why slow decline does more damage than hard failure. A dead lamp gets replaced on Tuesday; a fading one taxes every session until somebody measures it.
The Log That Makes Any of This Checkable
Wear-part accounting is only as good as your records, and the records are cheap.
- Counter reading, logged monthly with the date, not reconstructed from memory later.
- Delivered energy at one fixed reference setting, same meter every time. The trend matters more than the absolute value.
- Serial numbers and fitting dates for every lamp, rod, window and head.
- Every handpiece incident, including the drops nobody wants to write down.
- Water changes and filter swaps, tied to your service schedule rather than to whoever remembers.
This is procurement and engineering background for people specifying equipment, not clinical guidance. Want the wear-part list priced against a specific configuration? Send us the spec and we'll break it out part by part.
Frequently asked questions
My supplier quotes lamp life in hours. Should I insist on shots?
Ask for both, plus the conditions behind them. Hours only mean something once you know the repetition rate and pulse energy the rating assumed, and whether the clock counts firing time or powered-on time. Shots are the physical unit of wear, and the shot count is the one your console can actually verify.
Green output has dropped but 1064 nm still works. Is that the lamp?
Probably not, and the lamp is an expensive place to guess wrong. A fading lamp drags both lines down together. When only the 532 nm line weakens, the doubling crystal and its coatings are the likelier cause, since that crystal takes the full intensity of the light it produces. Get it inspected before ordering anything.
Why do warranties cover the console but exclude handpieces?
Handpieces absorb the handling. On a lamp-pumped head that means a glass envelope centimetres from splashing pigment, plus drops and cable strain. A diode head has no envelope, but it carries the same exposure in another form: thermal cycling on the emitter stack, an abraded sapphire contact window, coolant fittings. Argue the specifics anyway. Length of cover, loaner availability and freight direction stay negotiable even when the exclusion doesn't.
How do I audit wear before it shows up in results?
Measure output at one fixed setting on a schedule, and plot it. Published measurements of aesthetic devices have found real gaps between displayed and delivered output, plus meaningful pulse-to-pulse variation across a normal clinic list, so the display alone won't warn you. A meter and a logbook will.
References
- RP Photonics Encyclopedia - Flash Lamps: degradation mechanisms, explosion energy and achievable pulse counts
- Town G, Ash C, Eadie E, Moseley H. Measuring key parameters of intense pulsed light (IPL) devices. J Cosmet Laser Ther, 2007
- Lister TS, Brewin MP. Variations in laser energy outputs over a series of simulated treatments. Br J Dermatol, 2014
- Boulanger B, Fejer MM, Blachman R, Bordui PF. Study of KTiOPO4 gray-tracking at 1064, 532, and 355 nm. Appl Phys Lett, 1994
- Lawrence Berkeley National Laboratory EHS - Laser Non-Beam Hazards: capacitor discharge, grounding and lamp housing requirements
- IEC 60825-1:2014 - Safety of laser products - Part 1: Equipment classification and requirements
- IEC 60601-2-22:2019 - Particular requirements for basic safety and essential performance of surgical, cosmetic, therapeutic and diagnostic laser equipment