Thank you Peter for your interest in our continued safety. When can we expect copies of receipts for your payment for the audit? We’ll need copies all other associated documentation as well.
We do have to make sure everything is above board when there are legal requirements at play, and it’s important that we keep our lease – it would be unfortunate if we ended up in GCTS’s situation.
Since HSBNE is a do-ocracy, your volunteering to champion fixing the laser cutter is welcomed. We look forward to the day we can proudly say that Peter Caelli successfully repaired the space’s biggest investment.
In the interests of transparency, I thought I’d post a list of recent works performed on the laser cutter. This is a demonstration of just how much time and effort is constantly going into this machine, so hopefully general membership doesn’t think that we’re resting on our laurels delaying it or anything like that. Progress is happening pretty darned quickly, there’s just a lot of stuff that needs to be done!
Without further ado, here’s all the stuff that’s happened in the last two weeks since my previous post:
-Rubber sealing has been ordered to seal up the large gaps on the lid and access panels, to properly contain the beam in the event of catastrophic errors. Tracking says it’s left China and is due to arrive in the next few weeks.
-The fan on the electronics panel had its orientation reversed to pull air in, not push it out. The power control circuitry now gets a direct breeze onto it, which also passes through a dust filter before it enters the machine. This should increase the machine’s output consistency and reliability.
-Test filter material was ordered for the air intake. It’ll take a month to get here. No idea if it will be suitable, but the price was amazingly low so it’s worth a shot.
-All interior surfaces of the laser cutter have been thoroughly cleaned, excepting the honeycomb bed which needs a deep soak. We don’t think this has ever been done before because of the amount of crud we removed. This is one of the key things that will need to be done hourly as a maintenance aspect for this machine. Failure to do this will quickly dirty the optics and compromise the system’s output.
-The water cooler was opened up and cleaned of all dust. It wasn’t that bad inside here; we can get by doing this job less regularly than others. Note that this is not a chiller unit which means unfortunately that if it’s above 30’C that day, the laser cutter cannot be used. Just a fact of not having a full chiller. The cooler can be modified by changing its flow through the radiators to improve its cooling performance though, and we may need to investigate doing this during summer to keep things from overheating if people want to start cutting during a hot summer’s day.
-The air assist pump was opened up and had the dust filter cleaned. It also wasn’t that bad, so I think we can get away with doing this job less regularly.
-The Smoothieboard’s power source has been rerouted to use the machine’s on-board power supply and the USB cable has had its +5V line severed. This means the Smoothieboard is only powered when the laser cutter is on, and you can cycle the Smoothieboard by just turning the laser cutter off and on again. No more messing around with plugging and unplugging the USB cable, yaaay! Note that if you’re resetting the Smoothieboard, you need to count to 10 with the power off. Yes, really. I’m srs about waiting that long. It holds its state a surprisingly long time.
-We noticed the Y-axis belt was a bit loose and this was re-tensioned. We’ll check the rest next week.
As always, check the Trello for a list of jobs that need doing if you want to start helping out. There are many small tasks that don’t need a lot of skill, they just need doing!
@Wrexx, you’re more than welcome to inspect the machine any time. I’m not sure what value there is to doing a full audit right now because there’s so many places the machine can (and will be) modified & improved before it’s considered operational, and there’s so much that still needs doing that hasn’t even advanced enough to the point it’s hit Trello yet (like the induction material etc). I’d suggest something quick at this point in time, just so you don’t waste much of your time.
Is the +5V line severed in the cable? If so, it might be wise to mark the cable / socket mentioning that using cables with working +5V lines could damage the USB voltage regulator in either the laser cutter or someone’s laptop (or the kiosk).
The 5V line is severed in the cable that connects from the Smoothieboard to the external socket on the laser cutter itself. This cable isn’t a simple male -> female USB extension cable, it has screw-in mounts and is physically screwed into the machine. The external USB cable between the laser cutter and the host computer is still a normal cable and that’s fully interchangeable with any other USB cable.
This panel is not considered end-user-accessible and will be locked before the machine is “operational” so having this modified cable in there shouldn’t be an issue. Worst-case though, if you did bypass this severed connector and connected a full USB cable to the Smoothieboard, it wouldn’t cause any issues. The +5V system is isolated to just the Smoothieboard so it can’t feed back anywhere else, and it’s a pretty precise +5V so it’s doubtful much would feed back into your laptop/computer to cause an issue. To put it another way, this is how it used to be wired up a few months ago - power came from USB and the onboard system simultaneously, and you had to disconnect both to shut it down. That was a pain, so the USB one was severed.
It’s been a while since I last updated this thread. It’s honestly really disheartening to be standing beside the laser cutter as I work on it, and hearing someone on a tour describe the machine as something that “is always completely broken” or “will never work again” (both actual overheard quotes). So to try and dispel these rumours yet again, here’s another status update.
-Rubber door weather sealing strips arrived and have been re-purposed to act as a seal around a lot of the edges of the machine. This has greatly improved the machine’s safety to help contain the beam. It turned out really professional looking and I’m quite happy with it!
-Borrowing more weather sealing materials, adhesive foam strips were bought and have been applied to all edges that don’t have rubber strips on them. This guarantees the laser isn’t going to be escaping from any gap or crack no matter what. It also means that when the extraction fan is running, aside from the front air intake this machine is now pretty much airtight. The room with the laser cutter no longer gets smelly when you’re cutting stuff, yay! This is also important because the front air intake is the only area we’ll be adding an air filter… And it’s important that all air entering the machine is filtered, to cut down on how much cleaning we’ll have to do!
-Speaking of, yet more cleaning of the interior. This time there were surfaces that I know we already cleaned but were already dirty again. Ugh, cleaning is going to be the bane of this machine.
-Ten micro-switches have been acquired, courtesy of @nogthree. We need to fit at least two of these to the front/rear lower hatches, as these will be left unlocked so members can clean up once they’re done using this machine. This means we need to automatically disable the laser in case someone does something silly and opens these hatches while cutting.
-We recently performed some more test cuts, cutting through whole-bed plywood, MDF and etching glass cups on a curved surface. All tests went beautifully and we’re amassing a list of known-good cutting speeds/powers. Also it seems the machine is far more powerful than we expected and what it used to be because we accidentally melted and cut some of the glass. Um, oops. Note this machine is not to be used to cut glass in the future because that’s a bad idea, but engraving glass is on the cards.
-The window into the laser tube was removed and thoroughly cleaned (there was still a lot of residue on it from when it was bought; this is another area that had never been properly cleaned since we bought the machine). You can now clearly see the tube light up when it fires. It’s kinda neat to watch.
-A “panel open” Indicator light for when the laser is disabled because a panel is open has arrived and needs fitting. Having this will help a lot during troubleshooting when the laser stops firing and you don’t know why.
-The black paint has been touched up in a few areas where it was damaged by vigorous over-cleaning to remove some sticky residue. The machine looks all nice and shiny everywhere again!
-Filter material for the front air filter has arrived. Now we just need to knock together some brackets and a cover to mount it. I didn’t know how much resistance it would offer until I bought it, and thankfully it seems pretty low-resistance so it’ll work just fine.
-It turns out that sealing up all those massive gaps and cracks meant the system didn’t have enough air intake. So, the front air grill was enlarged by just cutting it all out. This gives much better airflow inside the machine too, since the air has to flow over the material when before it just came in from all directions. The exposed metal edge was protected by painting it with red spray paint that’s close but not a perfect colour match to the rest of the machine, but it doesn’t matter too much because that area will be covered up by the new air filter anyway.
See how the machine has a big worrying hole in the front? It’s now literally just as unsafe as it was before. Yeah. Bit more obvious just how unsafe it is, right? That’s why we’re trying to make it better! We’re almost there with the safety and the reliability issues, but not quite there yet. As always, you can see our current to-do list on the HSBNE Trello, and we’d love some help.
Thanks for all the time and effort you and your team are putting in to it. I see the laser as a hell of a selling point to membership even more so after i got quotes to got some jobs done.
Will be up on friday and will have a good think about how best to organize the cutting stock shelving
I’m not yet added to the Trello group but thought it would be worth mentioning that I can bring some filter-foam that I found in one of these foam and rubber shops up the coast. I’m using it for a dust-extractor for my cnc. Have some leftover that could prove useful for the air intake? I’ll bring it on Tuesday.
We’re well overdue for another update on this machine. I had to step back for a bit due to other commitments, but I’m back to working on it. Here’s what’s happened since last time I posted.
All locks, hinges, latches etc have been greased (many came with zero grease from the factory so they squeaked under use)
Keys to the laser cutter have been found and all are accounted for. There’s two types - one for the interlock and one for the access panels. I have both types, @devians has both types, Wrexx has just an access panel key, and there’s just an access panel key in the server room on the Vending Machine keyring. We need to cut a few more of the interlock one for redundancies.
Safety shutoff microswitches were added to the front and back lower access panels were added by @niklas. This means that if you have one of those doors open then the laser will automatically stop firing so you can’t hurt yourself. Yay! Thanks Niklas!
To help manage the fact that there’s now three safety shutoff microswitches that can trigger, a “lid open” indicator light has been added to the front panel by @niklas. This helps diagnose problems quite a bit!
A flashing shroud was installed on the front air intake by Wrexx, redirecting stray laser beams to the floor. (btw, due to legal/safety reasons, enclosed footwear and long pants will be a requirement of using this machine once it’s open to general use)
Filter material was added to the front air intake by myself and @crofty. Now all air going into the laser cutter is dust-free which will help extend the time between major cleanings. This filter material is easily removable and washable, so future servicing for this task shouldn’t be hard.
The large gaping hole on the laser tube access panel has been sealed up. Now we just need to physically screw this panel shut to prevent access by untrained persons.
All belts were retensioned and set correctly. Should eliminate some bounce we were getting.
@pelrun has joined in to help sort out our software stuff, because the current software workflow is pretty poor. Right now he’s changed Inkscape to use an up-to-date version of Python, fixed all the crashing, and fixed the way that you need to manually set the document units to “px” too. Longer-term he’s looking at ripping it out and using Visicut instead, which has a much nicer UI and will give us more features. Thanks James!
The new filter on the front air intake. It just sits there and these bolts prevent it from falling out, but you can undo them with your hands to remove & wash the filter.
An example of the amount of crud that gets deposited inside the machine every time it’s used. Yes really, this was the residue from a single cutting job! This is why it’s so important to clean it aftereveryuse, so we can keep this stuff off the lens & mirrors.
If you walked into the laser cutter room on Tuesday, you might’ve noticed that it was a flurry of activity. The whole room has been reorganised now, to a much better layout! We’re still sorting and organising things in here, but here’s some shots of what the room looked like at the end of Tuesday. And a reminder that because this room has to be kept neat & tidy as part of the Digifab cause, nothing should be placed in this room without clearing it first with the cause leader (even donations). At bare minimum we gotta dust it and find it a proper home rather than have it left lying messily on the table’s prime working space.
I was planning to name people to thank for their help with this task, but I’m so bad at remembering things I know I’ll forget half of you and I’d hate to do that. So, thank you to everyone who assisted with this room reshuffling!
Methinks it’s time for another update! I’m positive I’m going to be missing some stuff in this update, but I thought I should say something before I forget even more stuff.
@pelrun has done an amazing job by completely ripping out our old, clunky software workflow and replacing it with Visicut. This new software is awesome and SO much better at generating and sending Gcode - Pronterface is officially only used by us during debugging now! It will also let us do all sorts of new and exciting things down the track, which I’ll only mention once they’re fully working.
Speaking of new and exciting things it can do… We can now do Raster engraving. Yes really. The feature still needs a LOT of testing and calibrating, but we have confirmed it works with some test cuts.
A bunch of settings have been optimised and calibrated on the Smoothieboard. The tickle power has been increased from 10% to 15% to give better cutting of short fine details, but more importantly the maximum speed has been greatly increased. It’s now capable of zooming around the bed so quickly that it rocked the entire machine, so we had to dial it back a bit. It’s that fast. This makes it super quick to move around the bed in between cutting lines though!
Because the machine is so fast and so powerful now and it uses new software, this means we need to redo all our material defaults again. It also turns out that it can’t be used to raster engrave paper. I spent around eight hours trying to do that on Tuesday and I just couldn’t get it to work right. It can be vector engraved though, and vector cut. I’ll try to see if it can be done another day, but paper may just be too thin and delicate to successfully engrave with this powerful machine.
Two belts in the machine have been re-tensioned, a third is due to be re-tensioned but I ran out of time on Tuesday to do this.
Cleaning fluid for the honeycomb bed has been acquired, and a wooden tray has been built from scratch by @crofty for cleaning the bed. We still need to acquire some plastic to line the tray, then we can let it soak for a few hours.
Spares of both the access key and the interlock key have been cut and distributed. I have one set, @crofty the Digifab cause leader has one set, @devians has one set I think, and one set is on the master keyring in the server room. Now that we have enough keys, for safety reasons this machine will be kept locked when not in use. Gotta make sure we meet all the safety requirements we’re supposed to meet with something so dangerous. Note that this key system is basically temporary and it will be replaced with the swipe interlock system whenever that’s fully working.
Efforts have begun to start documenting stuff for inducting other people to use this machine. Yes really, work on induction materials has begun! There’s a LOT to do before it’s done, but an official start is worth mentioning on this list.
One of the very first tests of raster engraving - a HSBNE coaster made out of MDF. The engraving is solid and pretty deep, and the fire burn marks on the flaming chip kinda works too.
Necklace design by @Stevie, made out of two layers of different coloured Acrylic, used to test layering effects.
As always, if you’d like to help with some of the work still needing to be done to make this machine available to all members, come see either myself or @crofty.
I’ve built a stepper drive mechanism for the Z table in the cutter; we just need to figure out how best to mount it and wire it up.(and maybe replace the untested tiny stepper I’ve got in there right now). I’d give you a photo but Discourse doesn’t want to let me upload any. Can’t even upload dickbutt in protest.
Edit: well, it decided to work again after that ascii art I originally posted here.
(that is after I vacuumed up all the crap that had accumulated in there: it’s amazing how much stuff got under the belt cover plate…)
My original mounting plan turned out to be a bad idea, so I needed to print a new bottom plate with slots to take the original idler mounting bolts. It sort of has to be assembled around the belt, but it’s rock solid even before the final bolts are added.
Unfortunately, it’s still not completely satisfactory. The bushings I used don’t really like the side loading from the tensioned belt, especially since I’m using a threaded bolt and not a nice ground shaft as an axle. I’m definitely going to have to do yet another top/bottom plate and put in proper bearings instead.
Also, I still have no idea whether that stepper motor is any good.
I reluctantly redesigned and re-printed the top and bottom plates of the z drive mechanism to include skate bearings… which I could find none of in my stuff or floating around the space. So I stole them from the extruder in the space’s defunct sells mendel. Sorry!
Reassembled the drive and it’s just as rough as before. So all that work was mostly pointless, but it did imply that the problem is simply that I’ve over-tensioned the belt. Instead of trying to adjust the position of the drive mechanism, which is fiddly and inaccurate, the correct method is probably to move the plate in the centre of the bed with the two pulleys on it instead. I’ll do that when I next stick my head in there. Just wish it wasn’t a pain to remove the belt cover plates…
I’ve also been trying to figure out a way to use a visible laser to re-align the beam path. Unfortunately there really isn’t enough room after the laser tube to do this easily (32mm!). My next idea is to run the visible beam backwards - take the nozzle and collimating lens off the head assembly, and replace with a jig that holds a laser diode perfectly centred and firing vertically. The task then changes from getting the beam to the nozzle to getting the beam to the laser tube aperture. I think that’s going to be much, much easier.
I’ve also drawn a few diagrams to figure out which mirror movements are logically identical to adjusting the position of the laser tube, as it’s far easier and safer to move the mirrors than the tube. That said, it still might be necessary to replace the tube mounts with some finely adjustable ones - good thing someone’s already designed a printable set…
I suggest fundraising for a beam combiner + mount (Pelrun knows this, we have discussed and I am more enthusiastic than him, I’m just sharing) however it’s expensive enough that it’s a non-trivial investment.
It’d be great to have a beam combiner, but cost aside, there currently isn’t any room for another optic mount in the beam path. It’s pretty darn cramped in there. I think the money right now is better spent getting replacement/spare optics, as we’re completely dependent on the set we’re currently using.
Spent tuesday night going through hell fixing a belt tension issue on the right-hand Y belt; many thanks to Aaron for his help getting this sorted. It looks like it solved a major problem with the cutter binding and losing steps partway through jobs.
I got a new laser module for alignment purposes, and then fried it. That’s two green laser modules down - getting annoyed now!
It’s my own fault. I can actually get one working module out of the remains of the two broken ones (cracked the IR pump laser diode on one, damaged the lens over the DPSS crystal on the other… so I can put the crystal assembly from one on the pump diode from the other and get it successfully lasing. Mounting everything is another story.
Really though a super cheap <1mW red module would be equally workable for the alignment task, so I’ll stop fooling around with the green ones, at least for the moment.
That’s why we’re trying to make it better! We’re almost there with the safety and the reliability issues, but not quite there yet. As always, you can see our current to-do list on the HSBNE Trello, and we’d love some help.
