The Level 2 method for endoscopy: logs, swap tests, the image chain map and the repair line
Collect error codes and logs from an endoscopy tower, design swap tests between scopes, processors and cables that separate two suspects, place the fault on the image chain map, and decide what you may repair and what must go to the manufacturer.
Where Level 1 ends
In Endoscopy Equipment Service — Level 1 you learned the tower, the construction of a flexible and a rigid scope, the leak test, white balance, safe lamp handling, the reprocessing chain and the first symptom map in lesson 6. That map is still the starting point. Level 2 begins where it stops helping: the scope that is 'not recognised' on one processor and fine on another, the noise that comes once a day, the reprocessor that fails a cycle every Thursday, the light source that strikes only when warm.
The expensive mistake at this level is believing the unit that shows the message. An endoscopy tower is a chain, and the unit that notices a problem is rarely the one that caused it. A processor reports 'no connection' because the scope plug is wet. A light source reports a temperature fault because the dust filter on the cart side is blocked. A reprocessor reports a channel blockage because the connector on the scope is the wrong adapter. Replace the reporting unit and you have spent the customer's money and kept the fault.
Collect the evidence before anyone reboots
Modern processors, light sources, camera control units, insufflators and reprocessors keep messages and logs, and most of them classify messages by level. The Fujifilm EP-6000 service manual, for example, sorts its on-screen codes into information, warning and error levels, asks the user to write down the condition and the code when requesting repair, keeps a light source log and separate logs for one-connector scopes, and lets the engineer export logs and the customer's configuration to external memory. Other brands have their own versions; older units may show only an LED pattern or a code on the front panel. Whatever the system offers, capture it first.
The order matters. A burst of messages usually starts with the event nearest the cause. A processor that first reports a scope communication fault and then an image fault and then a recording fault has one problem, not three. A reprocessor that reports low water flow and then a disinfectant temperature error may simply not have had enough water to heat. Check that the clock is right before you trust the sequence; several units keep their time and settings on a backup battery, and the EP-6000 manual notes a lithium battery in the unit.
Then ask the staff the questions that a log cannot answer. Which scope, which tower, which room? During which part of the procedure: at connection, after white balance, when the knife was used, after the scope came back from the reprocessor? Since when, and what changed: a new disinfectant, a new cart, building work, a repair elsewhere, a dropped scope? The answers often decide the first swap.
- Photograph or export the error list and logs before rebooting or clearing anything.
- Note scope serial numbers, processor and light source serial numbers, software versions and the exact moment of failure.
- Save the customer's configuration by the manufacturer's method before you open or reset anything.
- Ask what changed in the last days: chemicals, staff, location, other repairs, power events.
The image chain map
Draw the chain for the system in front of you, on paper if needed. For a flexible video tower it runs: mains and earth, processor power supply, light source (separate or built in), light guide socket, scope light guide, tissue, objective and sensor at the tip, sensor cable along the scope, scope connector and its memory or communication board, the processor's patient-side board behind the isolation barrier, image processing board, output board, video cable, monitor, and in parallel the recorder, printer, capture card and network. For a rigid tower replace the scope part with telescope, coupler, camera head, camera cable and camera control unit, and add a separate light cable.
Every symptom lives somewhere on that chain. 'Not recognised' lives between the scope connector and the patient board. 'Recognised but black' lives at the sensor, its cable, the light or the output side. 'Dark but correct colour' lives in the light path. 'Colour wrong on every scope' lives in the light source, white balance data or the monitor. 'Lines only when cutting' lives in the coupling between the high-frequency surgery unit and the video chain. Put the complaint on the map before you choose a tool; it tells you which swap will split the chain in half.
Some systems change the map. On one-connector scopes, such as the 700-series scopes used with the EP-6000, the scope receives its power without contacts and sends video and control data optically through a window in the plug. A dirty or scratched communication window then behaves like a bad electrical contact: the manual's error list points the user to cleaning the connector first. Know which type of connection your system uses before you blame a board.
Swap tests that prove something
Endoscopy systems are modular, and swapping is the cheapest strong test you have. But a swap only proves something if it changes one thing and you know what the result means. The classic pair is: suspect scope on a known-good tower, known-good scope on the suspect tower. If the fault follows the scope, the scope is the suspect. If it stays with the tower, the tower is. If it appears on both, look at what both towers share: the room supply, the electrosurgical unit, the reprocessing, the operator.
Go further down the chain one step at a time. Change the video cable, then the monitor input, then the monitor. Change the light cable on a rigid tower, then the telescope, then the camera head. On a tower with a separate light source, swap only the light source. When the department has two identical processors, the swap of a single board between them is a workshop test and belongs to people authorised for that work; the swap of whole units with the staff's agreement is a field test you can do.
Two rules protect you. First, every scope you take for a swap is handled as contaminated until you have evidence it was reprocessed, and no swap is done with a patient connected. Second, a scope that is suspected of fluid inside is not connected to a second processor before you have done a leak test and looked at the connector. Moving a wet scope from tower to tower can take one board with it on each tower.
Reproduce, then measure
A fault you cannot reproduce you cannot prove fixed. Find the smallest action that makes it happen: flexing the insertion tube near the tip, flexing the universal cord near the plug, touching the plug body, leaving the scope connected and running until it warms up as it does in a procedure (do not heat it in water or with a heater: a scope with a hidden leak takes in fluid), running the light at maximum for a while, starting the light source cold. Write down the condition that triggers it; that condition is half the diagnosis.
Measure what can be measured safely: mains voltage and earth at the outlet, the supply to the cart's isolating transformer or multiple socket outlet, the temperature of the air leaving a light source, the pressure on the reprocessor's water inlet with the hospital's gauge, the light output of a light cable compared with a good one. The Electronics and Power Supplies course covers the instruments and how to use them safely on the bench; inside an endoscopy unit, measurements on open mains equipment are done only by the manufacturer's procedure and by trained people.
For intermittent faults, set a trap rather than guessing. Ask the staff to note every occurrence with the scope number, the time and what they were doing; leave a simple log sheet on the cart. Ten entries with conditions are worth more than an hour of looking at a fault that is not happening.
What you may repair, and what goes to the manufacturer
Level 1 drew the basic line: you repair what you have the documentation, parts, tools and training for, and what you can verify afterwards with tests you are able to perform. At Level 2 you will be closer to that line more often, so add three questions. Can I return the device to its original specification with original or specified parts? Can I prove that with the manufacturer's tests? Who carries the liability if this device harms a patient after my repair?
For processors, light sources, camera control units, insufflators and reprocessors, module-level repair by the manufacturer's procedure is often within reach of a trained independent engineer: power supply units, fans, filters, lamps, connectors, interface boards, valves and pumps, followed by the specified checks and electrical safety test. Component repair on a board is bench work for the electronics workshop. Work that needs service software, a password, a licence, an EEPROM or calibration write, or a manufacturer jig you do not have stops at the line; you hand over a clear package instead.
For flexible scopes the line is sharper. External parts the user can change are yours; opening the sealed body is specialist work, discussed in lesson 4. Never open a scope that is going to a specialist or under a manufacturer contract: you may destroy the evidence, void the estimate and take on liability for a device you did not finish. When you stop, the hand-over package is the same every time: the complaint, the logs, the swap results, your measurements, what you excluded and what you suspect.
'No connect' that lived in the scope
A Pentax video processor started to show 'no connect' when a particular gastroscope was plugged in, and the image did not appear. The first ideas on the phone were the processor's contact group and the cables. Colleagues advised the cheap split first: another scope, cable or processor shows whether the fault follows the scope, and the suspect scope is leak tested before it goes on a second processor. Their most likely explanation was fluid inside the scope, for example after reprocessing without a proper connector cap or with an unnoticed leak. The advice was a leak test, then opening the scope's video plug, drying it and cleaning its boards, as a scope repair rather than a processor repair. The final result was not reported.
Two swaps take ten minutes and can move the suspect from an expensive processor board to a wet scope. Leak test first, swap in both directions, and only then suspect the unit that shows the message.
Good on the gastroscope, bad on the colonoscope
A colonoscope connected to an endoscopy tower gave a defective image, while a gastro-fibrescope on the same tower gave a clear picture. The light guide connection was checked and changed nothing. The engineers who looked at it reasoned along the chain: the tower and its light were proven by the other scope, so the fault lived in the colonoscope. The most likely explanation was fluid inside the scope, a 'drowned' instrument, and the advice was a leak test and a repair in a specialist workshop rather than more work on the tower.
A second scope on the same tower is the quickest way to clear the tower. Once the fault follows one scope, the next test is the leak test, not the processor.
Shows yesterday's data, but not today's scope
A video processor would not detect a connected endoscope, although it still kept and displayed the data from the previous session. That detail mattered: the processor's memory and display were working, so the problem was more likely in the recognition path than in the main board. The advice from colleagues was to inspect the contact group on the scope and the socket on the processor for dirt or oxidation and to clean them by the manufacturer's method, and only then move to board diagnosis. The final cause was not confirmed in the discussion.
Use what still works to narrow what does not. A processor that remembers and displays is alive; look first at the contacts between it and the scope.
On-site checklist
- Save logs and configuration before any reboot or reset.
- Handle every scope as contaminated until reprocessing is confirmed.
- Leak test and inspect the plug before connecting a suspect scope to another tower.
- Draw the image chain for this system and place the complaint on it.
- Swap in both directions; change one thing at a time.
- Find a reproducible trigger before repairing.
- Check the repair is within your documents, parts, tools and tests.