Laser Maintenance Class/Lesson Plan/2017 Oct
From Site 3 coLaboratory Wiki
Approximate talking time: 1.5 hours. Add 45-60 minutes per student for hands-on time. A 4-hour class had just enough time for 3 students.
- 1 Introduction
- 2 Mirror Alignment
- 3 Cleaning Optics
- 4 Miscellaneous Mechanical
- 5 In Back
- 6 Conclusion
- Laser's default state is "not working"; our job is to give the illusion that it always works.
- Specific things that go wrong and get fixed: Mirrors out of alignment, mirrors or lens dirty, red laser out of alignment, air hose kinked, smoke extraction not working, table not level, screws loose, limit switches not working.
- If there are problems, write them down in the logbook. If you do anything to the laser, write it down in the logbook. If anyone says anything laser-related, write it down in the logbook. Check the logbook for recent entries as the first part of any maintenance task.
- If there is a problem that can't be fixed right away, tape a sign to the laser saying that it's out of service. Using it while it has a problem can damage it further.
- Open the logbook to a random page, and describe what happened on those days in your own words.
- Used to be the most common thing wrong with the laser; rare now.
- This produces a gradient or fading across the work area.
- If cutting is inconsistent across the work area, this is probably why.
- How to check: Run "burn-test-paper" at corners, edges, and middle, and check for consistency.
- Note: this test pattern uses "corner power" settings to fade out at the ends of lines.
- Old power supply: 5th line was marked, others cut. New: no 5th line; 4th line is barely cut.
- Use thin printer paper for this test, not cardstock.
- If there's no fourth line, or the fourth line is marked but not cut, there's fading.
- If lines are wide and charred, it's not alignment - it's focus. Alignment causes fade-out, with all lines still sharp.
- Set Z-datum in the middle of the work area for best focus for this test.
- Point out locations of mirrors (rear, front, and lens mount).
- Explain lens mount landmarks: screw is at the top of the slot (mount is all the way down), bottom of mount is almost but not quite resting on the stop (i.e. it's vertical).
- Check alignment before adjusting mirrors. If the laser works, don't fiddle with the optics.
- Check that the lens mount is in the correct position. If it's mis-mounted, it can block the beam (per below).
- Start with the back mirror. Check location of the spot on the front mirror when the rail is at the top of the work area and at the bottom of the work area.
- Close the lid afterwards. If you forget to do this, it'll slam down a few moments later and potentially throw the red laser alignment off.
- Do not go near the dichroic mount. Even looking at this funny is risky, and it's a massive pain to realign.
- Close the lid carefully - if it slams, this may throw the dichroic mount out of alignment (it's held in place by wishful thinking).
- You're looking for consistency, not centering. The dot should be at the same spot on the mirror in both positions.
- Next, adjust the front mirror, looking for the spot on the mirror in the lens mount. Check the location when the lens mount is all the way to the left and all the way to the right.
- Put the rail in the middle of the work area's Y range when doing this.
- Important: there should still be a fair bit of clearance between the spot and the edges of the lens mount mirror. The mirror is inset, and the edges of the hole can block the beam from hitting the lens.
- (Diagram of lens mount beam clearance goes here.)
- If it's impossible to get consistent positioning and adequate clearance, it means the lens mount is in the wrong position (usually too high). Doublecheck that.
- If it's fine and things still can't be lined up, there's a problem in back. Get me and expect lots of swearing.
- Recap: Check the lens mount position, adjust back mirror, adjust front mirror, making sure spot positions are consistent and not obstructed.
- For each person, misalign both of the mirrors and the optics mount. They fix it and then do an alignment check to test their fix.
- It doesn't have to be perfect. Let them make two tries, or take two minutes, then move on for the next person. They're trusted to safely practice further on their own.
- This is now the most common problem, and it's a serious one - it can destroy the lens.
- Downstairs, this was mostly sawdust and similar contamination from nearby work. Upstairs, this is mostly soot produced from the lasering itself.
- Wood is bad for this, leather is horrible for this. Engraving is really horrible for this. If anyone cuts these or does engraving, do a quick lens check afterwards.
- They're allowed to cut these; it's why the laser's at the shop. We just have to do appropriate maintenance as well.
- Lens is transparent to IR, but dust isn't. Dust heats up, and lens is sensitive to heat.
- Mirrors are less sensitive but should still be cleaned if dusty.
- If you can see the red dot on a mirror, it's dirty. A clean mirror has a very faint dot.
- Lens always shows a dot. To check for dust, shine a bright light into the lens mount from above, and look for particles.
- Also check for scuffing (usually from improper cleaning), for chipping (from over-tightening the mount), and for patches of shininess (indicates that the anti-reflection coating has been rubbed off). If someone let dust get too hot, that results in pitting.
- None of these can be fixed, but they should be written down. When they get bad enough, it's time for a new lens.
- The old lens shows all of these; point them out to people. Use good light.
- The goal is to slow down damage accumulation, so that the lens lasts as long as it can.
- Side mirrors have caps that unscrew, allowing access to the mirror without disturbing the mount too much.
- The mount still gets disturbed, so realigning will probably be needed.
- You can wipe these without unscrewing them; you just won't be able to clean the edges, and crud from the edges may smear.
- Mirrors are steel plated with nickel or chromium; they can be cleaned with alcohol or acetone. We'll use acetone.
- Lens is cleaned with acetone only - water or alcohol will eat it, producing poison gas.
- Acetone left uncapped will collect water from the air; leave it capped when possible.
- Use lens paper - otherwise clay particles fog the mirror, or lint is deposited and catches fire.
- Never touch the face of the lens (or any other optics) with your fingers. Only touch the edges.
- Lens mount disassembly:
- Put a piece of lens paper down on the grille.
- Disconnect the Z datum connector.
- Loosen the screw holding the lens mount to the laser.
- Let the lens mount turn so that the lens is on the top, not the bottom. This means a dropped screw won't scratch or chip it.
- Finish unscrewing the lens mount.
- To remove the mirror, loosen the mirror clip screw, rotate the mirror clip, and take out the mirror.
- To remove the lens:
- Hold the assembly with the lens facing up. Your fingers on that hand should always be holding both parts of the assembly together.
- Loosen the screws a little bit at a time, going around in sequence, to avoid pressure on the lens.
- After the screws are out, hold the assembly together by hand, and flip it over so that the lens is facing down. Use two hands for this.
- Hold the lens retaining plate, and lift off the rest of the assembly.
- Carefully pick the lens off the retaining plate, holding it by the edges.
- To clean a lens or a mirror, soak lens paper in acetone, and wipe the optical element in one smooth motion, starting from the middle. Rotate it, wipe again with a different part of the lens paper. Repeat until either it's clean or you need a new piece of lens paper.
- Do not wipe multiple times with the same part of the lens paper; that'll smear grease, and scour with dust. Both are bad.
- Acetone can pull grease with it, and leave it at the edge of the lens when it evaporates. To avoid this, don't soak the entire lens paper; start wiping with the damp part and finish with a dry part to wick away the acetone.
- The lens will take a lot of careful wiping. Mirrors are usually faster. Don't spend an excessive amount of time on either, though, because there's always at least a little bit of dust present, which will eventually scour and fog the optics.
- This it doesn't have a good antireflection coating any more.
- (Point out all of the damage on the old lens, now that it's out of its mount.)
- When putting the lens back into the lens assembly:
- Hold the retaining plate with gasket in one hand. Make sure the gasket is centered; if it overlaps the open area, it'll get burned, which is bad.
- Carefully place the lens on to the gasket, being careful not to move the gasket while doing so. The lens bulges down towards the bed of the machine.
- Carefully place the lens assembly over the lens. Hold the two pieces together by hand, and flip it over. Use two hands for this.
- Doublecheck that the gasket is not overlapping the open area. If it is, start over.
- Finger-tighten the screws most of the way, then tighten each a half- or quarter-turn at a time, moving around the border in sequence or across diagonals. This avoids putting too much stress on any one part of the edge of the lens.
- It's tight enough when you can't twirl the allen key shaft between your fingers.
- Screw the lens mount back on to the XY stage, and reconnect the Z datum connector.
- Remember to check that the Z datum connector is reconnected! If it isn't, the optics mount will crash, which throws a lot of the rest of the system out of alignment.
- Recap: Unscrew front mirror, clean mirror, screw it back in, take off lens mount, unmount lens, clean lens, remount lens, screw lens mount back in place.
- Each person demonstrates disassembly and cleaning of the front mirror, and of the lens mount (mirror and lens). Don't touch the back mirror for the test.
- Optional: Place a drop of grease on the front mirror and check that it's cleaned off properly.
- After everyone has finished demonstrating cleaning, the instructor realigns the front optics.
- Grab-bag of several topics: air hose, air compressor, extraction system, wiggle check, table check, limit switches, chiller check, interlock switch.
- Some of these can be fixed, some of these we just need to know about ASAP if they aren't working.
- Air system blows air on to the workpiece, sending smoke and soot away from the lens.
- Without this, the lens gets damaged very quickly.
- Check airflow by putting hand under the lens assembly; there should be lots of air coming through.
- Poor airflow usually means the hose is kinked (rubber tube connects to thin copper tube near the lens assembly; rubber tube often kinks).
- To fix, gently pull out slack hose from the cable guard. This gives it enough play to no longer kink.
- If there still isn't enough slack, gently pull the hose connection to the right, towards the end of the copper tube. Do not pull it off the tube - that's a pain to fix.
- Air compressor is noisy, but this can be reduced by having padding or a shim or both under it.
- If it's really loud, check to see if the shim has fallen out, or if the mountings are wandering. These are problems (the compressor can walk its way off the window ledge if it's free to move).
- Right now, improvised. Eventually it's worth picking up an isolation mat and mounting/gluing things properly.
- Past improvised solutions include a teddy bear and a piece of wood. Wadded-up cloth (e.g. rags) is probably the best option if these go missing.
- If there's poor airflow (or very loud noise) with no obvious cause, check behind the machine to make sure the hose from the compressor is still connected to the laser. Problems are very rare but have happened.
- Extraction check is done by running "burn-test-wood" on a piece of wood. If the smoke moves quickly towards the exhaust, it's working. If it lingers, that's bad (it will damage the lens fairly quickly).
- (FIXME - Put "burn test wood" settings on the wiki, beside the "burn test paper" settings.)
- Extra check: Put your hand on the housing of the extraction fan to see if it's on. It's very quiet, but you can feel the vibration. If you can't feel vibration, there's a problem.
- No easy way to fix this if the problem isn't obvious. Sign the laser as out-of-service and contact Marc.
- Wiggle check: See if any parts of the translation stage are loose.
- Easiest to check by trying to wiggle the horizontal beam up/down, left/right, and in the z direction, gently.
- If this is loose, people will be complaining about curves showing discontinuous jumps when the XY stage changes direction.
- Test pattern is a circle or set of circles.
- This is on the laser machine as "rattle test". It'll cut paper, which makes a mess; use it to mark wood instead.
- Fix is to carefully tighten the screws, then do another wiggle check.
- Do not over-tighten the small screw on the right of the XY stage. This prevents the X axis belt from moving; it'll make unpleasant noises instead.
- If this (or the other) belt can't move, it'll throw off the laser's homing; hit "reset" on the control panel to seek to the limit switches again.
Table Level Check
- The grille sometimes has crud collect under it. This causes focus to vary across its area.
- This can look like an alignment problem, but isn't. Out-of-focus areas have wider cuts with char around them; in-focus are narrow.
- The table that the grille sits on can be adjusted, but shouldn't be. I've never seen it go out of alignment in all the years I've been here; it's always been a problem with crud collecting under the grille instead.
- To fix:
- Move the XY stage all the way to the back.
- Carefully remove the grille, remove the support grid under it, sweep out everything with the hand brush (kept by the drill press downstairs), and reinstall.
- The support grid is missing screws and flops all over the place. Use care and you'll be fine.
- Do a Z-datum at the middle of the table, and do another alignment check. The focus should be good across the entire table.
- These tell the laser that its reached the uppermost or rightmost edge of the work area. The laser checks these when it's powered on.
- If one of these doesn't work, the laser will reach the right edge or upper right corner, then make a buzzing/grinding sound as it tries to continue seeking.
- Hit the small, square "stop" button on the control panel to stop it.
- Do not hit the Big Red Button. You don't need to, for this.
- Show them how to reset the BRB if it does get pressed (twist it).
- If the switch is working poorly but not broken, the temporary fix (as documented on the panel) is to use the cursor keys to back it off, click the switch a couple of times with your hand, then hit the "reset" button on the panel to have it try again.
- If there's a highlight bar on the control panel, it means you're in a menu (and the cursor keys won't respond). Hit "Esc" to get out of the menu.
- If seeking failed, it'll be in "don't damage itself" mode. Press and hold a cursor key; the first four seconds will be slow, then it speeds up.
- The limit switches are a standard type, and are attached with screws. There are usually spares in the laser supplies drawer; if not, post to the list and more can be ordered. Make sure to connect the wires when reinstalling one.
- For a 3-terminal switch, pay attention to "normally open" vs "normally closed" configuration. One terminal will always be connected, the other terminal selects NO vs NC; there are markings on the switch to indicate which is which.
- The limit switches are wired as "normally open".
- The chiller must always be on, and should always read a relatively cool temperature (usually 23 degrees C).
- If the chiller is off, or is reading warmer than 30 degrees, turn everything off, sign the laser as out of service, and post to the list. If the laser tube gets too hot, it'll destroy itself, and it's quite expensive to replace.
- There's at least one interlock switch on the case that turns the CO2 laser off if the lid is open.
- If the laser is never cutting, this switch failing might be why.
- It's a magnetic switch on the left side of the enclosure near the front.
- If people store magnets here, it can mess up detection.
- If the magnetic bead it interacts with is out of position, that can mess up detection too.
- This is almost never a problem, but is worth checking if the CO2 laser isn't turning on.
- Without the students watching (and with the laser powered off), the air hose gets kinked, a wire from one of the limit switches is disconnected, and a foreign object is placed under one corner of the grille. Each student has to identify and fix these problems, with different variations per student.
- Optional extras: Unplug the extraction fan, remove the shim from under the air compressor, put a magnet near the interlock switch.
- Note: I wasn't able to reliably get the interlock to mis-register using a magnet. Omit that test until a reliable way is found.
- Do not mess with the screws. Anything I did that caused rattling also threw alignment completely off. Less-extreme options didn't reproduce the cutting problem (cuts came out fine). If the problem occurs again in the wild, make detailed notes about which screws were loose.
- Students can watch each other; they just can't know where the problem spots are when it's their turn (so that they have to diagnose it).
- Two systems of note: the dichroic (which combines the red laser beam and the IR beam), and the laser tube.
- Not covering: Laser power supply, laser controller board, wiring. Show where these are, instruct not to mess with them.
- The "dichroic" is the beam combiner, merging the IR and red laser beams.
To test red laser alignment:
- Place a card in the path of the beam, and push the "laser" button on the control panel. Observe the position of the hole or burn mark, and the position of the red dot. If the red dot goes through the hole, the beams are aligned at that XY location.
- If the laser doesn't mark the card, check settings in the control panel laser menu. 100% power for 100 ms worked for me.
- Use a piece of tape to hold the paper in place, so that it doesn't move during the test or while you're checking the spot position. Opening and closing the lid will cause enough air motion to move the paper.
- Do not just use tape; the glue will boil and coat nearby optics.
- Don't use lens paper or other tissue; it'll burst into flame.
- I normally move the lens assembly to the lower right of the work area, then check the red laser at the front mirror (at the lower left) and at the optics mount (at the lower right).
- If the red laser is misaligned, you'll need to check at least two points along the beam path to figure out whether position, direction, or both need to be changed.
- The goal is consistency, not centering. As long as the IR beam overlaps some part of the red spot, and it's always the same part, it's fine.
To inspect and fix red laser (and dichroic) alignment:
- The dichroic mount holds two parts: A coin-shaped optical element (the dichroic itself) that is transparent to IR but reflects (some) visible light, and a red diode laser that shines on to the dichroic, reflecting in the direction of the machine's IR beam.
- The red laser has to be realigned whenever this is adjusted or moved, and realigning the red laser is an incredible pain in the tail. Avoid adjusting the dichroic mount if possible. Avoid going anywhere near it if at all possible.
- This can be accidentally jostled when aligning the back mirror, and can fall out of alignment if the cover for the back area slams.
- The dichroic itself should be lined up with the IR beam. It can be positioned by loosening the nut holding it in place, moving it, and tightening the nut again.
- The IR beam should always be very close to the center of the dichroic. If it isn't, it might hit the mount instead, which will send a high-power cutting beam reflecting around the back chamber. This is very bad.
- The dichroic mount is held on by friction and wishful thinking. Jostling the machine can throw it out of alignment, resulting in 30-60 minutes of swearing while I or another maintainer tries to realign it. Avoid jostling the machine.
- The red laser's position and pointing direction can both be changed. Both will need to be changed to line it up properly. The laser has a soft plastic shell, and there are six screws that hold the shell and the laser in place. Adjusting these screws changes where the laser starts from and where the laser points.
- At least one of these screws is stripped.
- Because the shell deforms, and not all screws are in contact with the shell at any given time, there is no obvious relationship between where the screws are and where the laser starts and is pointing. If you have to adjust this, good luck.
- Moving the dichroic mount side to side (slightly), up and down (slightly), and bending it forwards and back (gently) will also adjust position, in more predictable ways. You'll still need to mess with the screws, though.
- This is a gas discharge tube. A high-voltage discharge makes the gas glow, mirrors bounce the light back and forth so that lasing can happen, and a water jacket keeps everything cool.
- This is a sealed system. We can't repair or clean it. We can't even clean the output mirror, so cover that opening if doing anything with the cover open that might make dust.
- Make sure the power is off before messing with the tube. The high voltage supply has exactly the right traits to kill you dead: enough current to stop the heart, not enough to restart it, and enough voltage to deliver that current.
- The tube is on adjustable support posts. Do not adjust these.
- They can only be coarsely positioned, and it's easy to point the tube in a direction that makes it impossible to align the laser properly.
- Moving the tube will also require realigning the red laser, which is an incredible pain in the tail.
- Per previous section, the IR laser must always pass through the middle of the dichroic (or close to it). Hitting the dichroic mount instead is very bad.
- The tube will eventually burn out. The main symptom is that it loses power. Alignment will be perfect and it still won't cut.
- This has only happened once, over many years. Usually the problem is something else (like the power supply, an interlock switch acting up, or some similar problem).
- Have each student do a card test to measure the alignment of the red laser beam with respect to the IR beam.
- If a large hole is burned in the card, reduce the pulse duration using the control panel's menu.
- Do not misalign the red laser or have them fiddle with it. The alignment setup is atrociously bad, and it'll take 30-60 minutes to get it back in alignment, mostly by sheer fluke rather than skilled manipulation.
- If we ever replace the red laser and dichroic mounts, then alignment can be added to the practicum.
- Class members are now authorized to diagnose and fix problems with the laser on their own initiative.
- Everyone gets a fob or card to show that they've passed the course and are authorized to do this.
- Everyone is also authorized to practice on their own.
- Keep lens cleaning to a minimum, though, because even a careful cleaning shortens the life of the lens.