Description
Product Overview
This is the 10-piece pack of the PF20-03-M01; the optic is identical to the single-unit version. Ten 50.8 mm (2 in) protected gold mirrors, each 12.0 mm thick on fused silica, each specified at 96 percent average reflectance from 800 nm to 20 µm, each flat to lambda/10 peak-to-valley at 633 nm, each carrying its part number laser engraved on the rear face. What differs from a single-unit order is not the optic. It is the fact that all ten come out of one coating run and one packing operation, which changes what you can do with them.
The problem a pack solves is not price, it is baseline
An infrared build of any size uses more mirrors than its drawing suggests. A single 4f relay with two fold mirrors and a pickoff is already five surfaces; a pump-probe bench with a delay line, a periscope and two diagnostic taps runs to twelve or sixteen. Buy those one at a time over three months and you receive three or four different coating lots. Every lot has a slightly different average reflectance – all inside the 96 percent specification, all perfectly good optics, and all subtly different from each other.
Then something goes wrong. The signal at the end of the path is 8 percent low. You cannot tell whether the loss is a misaligned fold, a contaminated surface, or simply the accumulated difference between four coating lots that each sat at a different point inside the specification window. Debugging becomes guesswork, and guesswork in an infrared bench is expensive because every diagnostic costs a detector and an afternoon.
A single-lot pack removes that variable. Ten mirrors from one run share one reflectance curve, so a loss you measure is a loss you caused. That is the entire engineering case, and it is worth more than the packaging saving.
What is in the box, and what it lets you do
- Ten PF20-03-M01 optics, one coating lot, individually protected, packed as a single shipment.
- One reflectance baseline, so a throughput measurement taken on any of the ten can be compared with any other.
- One purchase order, one goods-in inspection, one entry in the asset register.
- Nine spares while the first optic is in the beam, which is the difference between a planned swap and a lost week.
How to keep the lot together
The value evaporates the moment the ten are separated. Put a lot code on the storage case, not just on the invoice. When a mirror goes into a system, record which position and which lot; when one comes out, record why. A laboratory that can say “position 4, lot B, replaced after 14 months because reflectance dropped” has a maintenance history. A laboratory that can say “we changed a mirror” has nothing.
Key Features
- Ten identical optics from one coating run – a shared reflectance baseline across a whole optical bench, which is what makes throughput debugging tractable.
- 96 percent average reflectance, 800 nm to 20 µm – the protected gold curve, flat enough that pump, telecom and thermal beams can share the same mounts.
- Pack size: 10 mirrors – sized so that a twelve-to-sixteen surface infrared build can be equipped and still leave spares in the cabinet.
- Silicon dioxide overcoat – a surface that tolerates cleaning, which matters when the spares will sit in a drawer for a year and then be put into a dirty environment.
- 50.8 mm diameter, 12.0 mm thickness, 45.7 mm clear aperture – the standard 2 inch footprint, so the pack can be consumed across different mounts and different experiments.
- Lambda/10 peak-to-valley at 633 nm, 40-20 scratch-dig – laser grade on every unit, not a selected subset.
- Part number engraved on the rear face of every optic – the practical basis of an asset register; a mirror without a legible marking cannot be audited.
- Pulsed threshold 2 J/cm² at 1.064 µm – about 16 mJ on the 1.000 mm test spot, so the pack is usable for pump beams and not only for low-power diagnostics.
- CW thresholds 500 W/cm at 1.070 µm and 750 W/cm at 10.6 µm – both quoted per centimetre of spot, to be scaled with your own beam size.
- Angle of incidence 0 to 45 degrees, parallelism 3 arcmin or better – fold, periscope and pickoff duty from a single stocked part number.
Technical Specifications
| Parameter | Specification |
|---|---|
| Pack size | 10 mirrors |
| Optic | PF20-03-M01, identical to the single-unit version |
| Coating | Protected gold with SiO2 overcoat |
| Band | 800 nm – 20 µm |
| Average reflectance | 96 percent |
| Substrate | Fused silica, CTE 0.55 x 10-6 per degree C |
| Diameter and tolerance | 50.8 mm (2 in), +0.0 / -0.1 mm |
| Thickness and tolerance | 12.0 mm (0.47 in), plus or minus 0.2 mm |
| Clear aperture | 45.7 mm, 90 percent of diameter |
| Surface flatness | lambda/10 peak-to-valley at 633 nm |
| Surface quality | 40-20 scratch-dig |
| Parallelism | 3 arcmin or better |
| Angle of incidence | 0 to 45 degrees |
| Marking | Part number laser engraved on the back of every optic |
Damage thresholds, as published for this coating
| Condition | Value |
|---|---|
| 1.064 µm, 10 ns, 10 Hz, spot 1.000 mm | 2 J/cm² |
| 1.070 µm, CW, spot 0.089 mm | 500 W/cm |
| 10.6 µm, CW, spot 0.339 mm | 750 W/cm |
The threshold is a property of the coating and does not change with aperture, so all ten optics in the pack carry the same limit. It was measured on a clean optic, and contamination reduces it; a spare that has been sitting uncovered on a shelf is not the same as the optic that was tested.
Applications
Equipping an infrared bench in one purchase
Twelve to sixteen surfaces is the normal count for a pump-probe or OPO bench with diagnostics. Ordering a pack means the whole bench is populated from one lot, the alignment is done once against a consistent reflectance, and the surplus stays in the cabinet as known-good spares rather than as an urgent next-day order.
Multi-station teaching and shared-user laboratories
Where several users share instruments, mirrors get swapped between setups and occasionally go missing. A pack with a recorded lot code gives the facility manager a way to answer “which mirrors are in which experiment” without opening every enclosure. The engraved part number is what makes this possible at all, because a bare gold mirror has no distinguishing mark once it leaves its packaging.
Integrators building several identical units
If a small OEM build repeats the same optical train four or five times, a pack maps cleanly onto the build: one lot, one set of acceptance measurements, one set of spares proportional to the fleet. It also means a customer complaint about throughput can be checked against a mirror from the same lot rather than against a new purchase.
Labs with a maintenance backlog
An older infrared setup often has mirrors in it that nobody can identify and nobody wants to remove, because removing one means realigning a path that took a week to set up. A pack changes the economics of that decision: the replacement is already on the shelf, already from a known lot, so the swap can be scheduled rather than postponed.
Selection Guide
Sizing the pack to the system
| Optical path | Surfaces in service | Spares at 1.3x | Packs to hold |
|---|---|---|---|
| Single fold plus pickoff | 3 to 5 | 4 to 7 | 1 |
| 4f relay with diagnostics | 6 to 8 | 8 to 11 | 1 |
| Pump-probe with delay line | 12 to 16 | 16 to 21 | 2 |
| Two benches, shared stores | 20 to 30 | 26 to 39 | 3 to 4 |
The 1.3 multiplier is a working rule, not a specification: it covers one in-service failure plus one optic out for cleaning at any time. Facilities that run unattended overnight or that cannot tolerate a shutdown should use 1.5. Facilities with a clean, filtered, sealed enclosure can work at 1.15.
Choosing the coating for the pack
A pack only makes sense for a coating that the whole bench can use. Protected gold, 800 nm to 20 µm at 96 percent, is the natural choice when the path carries more than one infrared band. If every beam on the bench is above 2 µm, the mid-infrared enhanced gold pack reaches 98 percent and is the better purchase. If any part of the path runs below 800 nm, no gold coating will do and the pack should not be bought at all.
Series Comparison
The seven coatings share the same 50.8 mm, 12.0 mm fused silica substrate. The right-hand column describes what each coating contributes to a multi-path infrared build, which is the context in which a pack is normally bought.
| Coating | Band | Ravg | Role in an infrared build |
|---|---|---|---|
| -F01 UV enhanced aluminum | 250 – 450 nm | 90 percent | Not applicable to an infrared path |
| -G01 protected aluminum | 450 nm – 20 µm | 90 / 95 percent | Visible alignment beams and broadband work |
| -P01 protected silver | 450 nm – 20 µm | 97 / 95 percent | Highest visible throughput, needs care |
| -M01 protected gold | 800 nm – 20 µm | 96 percent | The general-purpose infrared surface, stocked as a pack |
| -M02 MIR enhanced gold | 2 – 20 µm | 98 percent | Thermal infrared only, also stocked as a pack |
| -M03 unprotected gold | 800 nm – 20 µm | 97 percent | Polarization-critical paths, not cleanable |
| -L01 line optimised gold | 10.6 µm line | 99 percent | Single-line CO2 work |
FAQ
Is the mirror in the 10-piece pack the same optic as the single-unit version?
Yes. The optic is the PF20-03-M01, identical in substrate, coating, figure, cosmetic grade and marking. The pack changes the quantity, the packaging and the fact that the units are drawn from one coating lot. It does not change the specification, and it does not introduce a lower grade of optic.
Why does a pack cost less per optic if the mirrors are the same?
Because the cost being removed is packaging, handling and order administration, not coating quality. Ten optics in one protective case use less material than ten individual cases, one goods-in inspection replaces ten, and one shipment replaces ten. The coating run is the expensive part of any mirror and it is unchanged. Any supplier who offers a large per-unit discount on a pack and also claims identical coating quality is either subsidising the sale or being loose with the word identical.
Will all ten have exactly the same reflectance?
They come from one coating run and will be close, but 96 percent is a specified average across 800 nm to 20 µm, not a single measured number stamped on each optic. What you gain from the pack is that the lot-to-lot variation which normally sits between purchases is absent. If you need a measured curve per optic, that is a separate metrology service and should be requested explicitly.
How do I decide that a mirror should be retired?
Three signals, in increasing order of urgency: the throughput of the path drops measurably against a recorded baseline; a visual inspection shows a haze, a film or a scratch inside the clear aperture; or the coating shows a lifted edge where the overcoat meets the substrate. The first is the only one that needs instrumentation, and it is the reason to record a baseline when the system is new. Do not retire an optic because it looks slightly different under a torch from a new one; gold coatings vary in apparent colour with lighting.
How should the spares be stored?
In the original case, coated face up, in a dry cabinet if the laboratory humidity is above about 60 percent, and never stacked. Do not wrap them in tissue that sheds fibres. Label the case with the lot code and the purchase date, because a spare that has been in a drawer for four years is still good but should be inspected before it goes into a beam path.
Can I split a pack between two sites?
You can, but you lose the main advantage. The point of the pack is a shared baseline within one build. If two sites need mirrors, buy two packs, or accept that the two sites will be working to slightly different reflectance curves and should not compare throughput numbers with each other.
Usage Precautions
- Keep the lot together until it is consumed. Write the lot code on the case, not only on the paperwork. A pack that has been mixed with older stock is no longer a controlled baseline.
- Clean with reagent-grade acetone, methanol or isopropanol and lens tissue only. Confirm against the coating handling notes before the first clean. Do not use ethanol-based laboratory wipes, and do not dry-wipe a gold surface.
- Handle by the edge, gloved. The overcoat protects the gold from atmosphere, not from a thumbnail.
- Do not stack or slide optics against each other. Contact damage on a gold film is permanent and sits inside the clear aperture where it scatters.
- Mount with a nylon-tipped retaining ring, finger tight. Twelve millimetres of fused silica is stiff, but point loading from a metal set screw can still print into a lambda/10 surface.
- Record position, lot and date on every install and removal. The engraved part number exists so that this record can be reconstructed later; use it.
- Inspect a spare before use, not after. A mirror that has been in storage for years should be checked for film and edge damage before it goes into a path that took a week to align.
- Respect the invisible beam. At 1.5 µm and beyond, four percent of the beam is still leaving the mirror in a direction you cannot see.
Disclaimer
Specifications quoted here are the manufacturer’s published values for the PF20-03-M01 coating and apply equally to every optic in the pack. Average reflectance is a band average. Damage thresholds are measured values on clean optics under the stated conditions and do not guarantee survival in a specific system. Pack contents, packaging format and lot grouping are as supplied at the time of order; confirm with us if a single-lot shipment is essential to your acceptance procedure. Gaofei Laser is a distributor and does not manufacture this optic.
Contact Us
Send us the number of surfaces in your build, the wavelength range and the power level, and we will tell you how many packs to hold and whether protected gold is the right coating for the whole path.
Hotline: +86 159-2722-0467
Email: 859818709@qq.com
Website: www.soar-laser.cn









Reviews
There are no reviews yet.