Independent evaluation and repair for the KATANA 10 HP picosecond platform. Call us when output has collapsed, a setpoint no longer changes the beam, or the model designation does not match what the system is doing. The configuration is verified from the physical labels first, and you get written findings before any repair.
More KATANA evaluations go wrong at this step than at any point later. The model designation a laboratory has on record is frequently not the configuration sitting on the optical table, and on a platform that shipped in OEM and custom variants, that difference changes the entire diagnostic approach.
In the published Onefive KATANA HP family, the model number identifies the wavelength band, not the output power. A system described as a 10 HP but producing green output is therefore a contradiction that has to be resolved before anything else is measured. It is normally one of four things:
Photographs of every product label, read individually: laser head, controller, power supply, and any separate conversion or trigger module, each showing the complete model and serial number. Photographs of the rear panels and connections. The original configuration documents if they still exist.
No repair quotation is finalized on this platform until the exact configuration has been confirmed from the physical equipment. A quotation written against the wrong configuration is worse than no quotation.
As published in Onefive and NKT Photonics KATANA HP material, the designation number corresponds to the output wavelength band. Use this to check what your label is actually telling you.
| Designation | Published wavelength band | Typical role |
|---|---|---|
| KATANA 05 HP | 512 to 532 nm (green) | High-power visible source, excitation and general visible work |
| KATANA 06 HP | 556 to 620 nm (orange) | Visible excitation |
| KATANA 08 HP | 775 nm | Depletion source in STED microscopy, paired with a visible excitation unit |
| KATANA 10 HP | 1012 to 1320 nm, commonly 1030 to 1064 nm | Infrared fundamental, and the pump for downstream conversion |
| KATANA 15 HP | 1550 nm | Telecom-band infrared |
Band definitions vary slightly between the manufacturer's own documents across model years, and individual units were built to order. Treat this table as an orientation aid for reading your label, not as a specification for your specific serial number.
| Parameter | Published platform behavior |
|---|---|
| Pulse duration | Approximately 30 ps to 10 ns, configuration dependent |
| Repetition rate | Pulse-on-demand up to approximately 100 MHz, continuously tunable |
| Triggering | Internal or external source, master or slave operation |
| Architecture | Seeded, amplified, alignment-free pulsed module |
| Visible high-power example | KATANA 05 HP at 532 nm published up to approximately 5 W and 5 µJ, with peak power in the hundreds of kilowatts |
Your unit's actual configuration takes precedence over any published family figure. Where original acceptance data exists, that is the reference we work against.
The platform name covers several generations and power classes, and the service implications differ.
For evaluation purposes the meaningful difference is not the marketing tier but what is physically inside the box: the seed architecture, the number of amplification stages, whether a wavelength-conversion stage is present, and whether the control electronics are the original generation or a later revision. Those are determined on the bench, not from the model name.
The system powers up and reports normally, but measured optical power is a small fraction of what the laboratory expects, or is at the detector noise floor.
Commanding a new output or current value from the front panel or from software changes the displayed value, while measured optical power stays where it was. Covered in detail below.
Repetition rate and synchronization outputs behave normally. This confirms the timing electronics are running. It does not confirm that optical pulses exist, because the timing path and the optical generation path are separate.
Measurable infrared power at the fundamental with disproportionately low visible output. This is the genuine conversion-stage case, and it is distinguished from the others by the fundamental power actually reaching the stage.
Crystal oven temperature that will not settle, oscillates, or drifts against its setpoint. A different failure from a stage that simply holds a flat temperature.
Pump-current driver failures, protection trips, degraded pump diodes, or a fault in the amplification chain that the control layer does not report as an error.
Serial or software control that appears to work while the lower-level controller is unresponsive, and readback values that turn out to be echoes of commanded values rather than measurements.
An internal interlock or protection condition limiting power with no error, no warning and no alarm presented to the operator.
Send the model, a photograph of the product label and the symptoms. We review it and respond with the next step.
This is the pattern we are contacted about most often on this platform, and it is worth setting out the reasoning, because the obvious conclusion is usually the wrong one.
Second-harmonic generation cannot produce more converted power than the fundamental delivered into it, and in practice it produces a modest fraction of it. If the visible reading and the infrared reading are of the same order, the two numbers cannot both describe a working conversion process fed by that fundamental. There are three ways that arithmetic resolves, and all three have to be tested rather than assumed:
A temperature-stabilized crystal oven exists to hold its setpoint. When significant optical power passes through the crystal, the resulting heat load is visible as a disturbance the controller has to work against. When little or no fundamental arrives, there is no meaningful load to disturb. A temperature that does not move when the current command changes is therefore consistent with an absent fundamental, and it is a weak indicator of crystal failure. A genuine conversion-stage fault more often shows as temperature that will not reach or hold setpoint, or as poor conversion in the presence of a confirmed, healthy fundamental.
On a seeded, externally triggerable architecture, the timing and synchronization electronics run independently of whether the seed diode is emitting and whether the amplifier chain is delivering gain. A clean repetition-rate indication and a normal sync output tell you the trigger board is alive. The presence of an actual optical pulse train is a separate measurement, made with a suitable fast photodetector on the real beam.
Systems of this generation frequently display the commanded setpoint rather than a measured value. If the command reaches the display layer but not the current driver, or if the driver is sitting in a limited or protection state, the panel will read exactly as expected while nothing changes physically. The test is to measure the drive current at the hardware and compare it against the command, and to establish whether the readback path is a measurement or an echo.
On this symptom pattern, the system usually retains its control and timing functions while optical power generation is absent or strongly suppressed. The fault is far more likely to sit upstream, in the seed, the seed driver, the pump, the pump driver, the amplifier chain or the fiber path, than in the frequency-conversion stage. The conversion stage cannot be judged at all until the true fundamental power reaching it has been measured.
The order of work on this platform is set by one principle: eliminate the cheap explanations before touching anything expensive. The configuration is established from the physical labels, the measurement setup is validated before any reading is trusted, and the fundamental power reaching the conversion stage is established before the conversion stage is judged. That single measurement separates a conversion problem from everything upstream of it, and it is why a case can resolve differently from the way it is first described.
From there the work is systematic rather than exploratory. The command path is compared against actual drive current at the hardware, the optical pulse train is confirmed independently of the timing electronics, and the seed, pump, amplifier and fiber path are divided until the fault sits in one subsystem. The feedback path is checked in both directions, because a healthy laser reporting a false low value and a degraded laser reporting a healthy one are both seen on this generation of controller.
You receive written findings before any repair is quoted, and no repair, substitution or modification happens without your written approval of the scope and the price.
A KATANA system that cannot be brought to its normal operating state on the bench cannot be diagnosed. Send:
On a discontinued platform, the customer's own paperwork is frequently the only surviving record of the configuration. Send it even if it looks irrelevant.
We receive systems from outside the United States, including from Canadian universities and research institutes. We will provide the commercial documentation a repair shipment normally requires and coordinate with your shipping office. Customs classification, duties and clearance timing remain the sender's responsibility, and we do not give customs guarantees.
What we offer on this model is a defined diagnostic evaluation with written findings, followed by repair engineering within a scope you approve. Stated plainly:
The measurements that matter here are the visible and infrared components separated with wavelength-selective optics and read on calibrated free-space detection, the real optical pulse train confirmed with fast photodetection, the commanded seed and pump current compared against actual current at the hardware, the spectral condition of whatever light is present, and the fundamental power arriving at the conversion stage. PhaseX defines and provides that measurement capability for the project, drawing on owned, rented or qualified third-party measurement resources arranged as required, and the proposal states where each measurement comes from before anything ships.
Pulse-duration and beam-quality measurement, where this platform requires it, is performed with the measurement resource named in the proposal. We do not offer manufacturer-equivalent calibration, and we do not certify to original published specifications where the original acceptance data is unavailable.
Class 4 work is planned and executed under a project-specific laser safety plan. The required engineering controls, personal protective equipment, beam management and facility conditions are defined in writing before any energized optical work begins, and the infrastructure that plan calls for is provided within the project scope. Where a laboratory's own controls are involved, final hazard analysis and approval stay with the institution's LSO or EHS function.
The KATANA platform originated with Onefive in Switzerland, moved to NKT Photonics, and the associated product line now sits within Hamamatsu Photonics. Hamamatsu's published application material identifies KATANA HP as obsolete.
For a laboratory, that means the ordinary route has closed: no current service contract, no factory module exchange, and in many cases no surviving contact who knows the configuration. It does not mean the instrument is finished. Faults in these systems concentrate in areas that respond to component-level engineering: drivers, supplies, thermal control, feedback paths, fiber interfaces and connectors, and control electronics. Those are the areas we work in.
Where an original part no longer exists, the options are substitution with a functionally equivalent component, a redesigned driver or interface, or a documented finding that the system is not economically repairable. All three are legitimate results, and all three are delivered in writing.
PhaseX Laser Services is an independent third-party service provider and is not affiliated with, authorized by, or endorsed by Onefive, NKT Photonics, Hamamatsu Photonics or any other manufacturer unless expressly stated in writing. Brand names and model names are used only to identify equipment for which customers may request evaluation.
Manufacturer and distributor documentation used for the family and platform descriptions on this page. Where our reading of a document differs from the configuration in front of us, the physical equipment governs.
The full KATANA and Origami family, service scope and what we ask for at intake.
Read the pageHow a conversion stage is actually tested, and why it is so often blamed for an upstream fault.
Read the pageSeed, amplifier chain and pulse-train diagnostics across picosecond platforms.
Read the pageWhat remains possible after end-of-support, and where the real limits are.
Read the pageOn this platform the first useful answer comes from the labels, not from the symptom. Send photographs of every module label with the complete model and serial number, the operating parameters you expect, and what you actually measured. We will tell you whether the configuration is one we can evaluate and what the evaluation would involve.
Request service