The Monocle frame: full-size pieces in place

I started work piecing together the frame. I started to regret making it only 1 meter high, though 1.5 meters would be absurdly high. I compromised by making 3D-printed plugs that fit the bottom of the 4040 extrusion and add 20mm. They also hold a metal socket for locking 60mm casters that will add another roughly 66mm (with stem) to the height, so the overall lift will be about 86mm to the total height. Then I realized that I wanted some sort of cap on the top to hide the sharp edges, and if I made those caps 10mm thick I could slide all the horizontal extrusion up 10mm without any real change in overall rigidity, which ultimately moves the bed up another 10mm — almost 100mm total height increase, so that’s an improvement.

I designed a set of tap alignment jigs to make it easier to tap and used polished spiral-flute taps to power tap with my drill set to slow and alcohol as a lubricant. Worked great; never worried that I was going to break a tap, and it was fast. Then the caps on top are attached to the posts with M5x16 flat-head screws. I think it’s a nice look. The top caps and bottom plugs are printed in PETG.

Ultimately, the bed will still be a little lower than I want, but not enough for me to buy 1.5 meter 4040 and then cut it down to size. If I ever change my mind, it’s just changing out a bunch of screws, one corner at a time, and I’ll be set. And it’s tall enough to fit the Q800 refrigerated cooler underneath with room to spare.

Here it is with a tensioning strap I used to hold it together tightly while fastening the middle c-beam:

I had the 1.5 meter c-beam sitting around. In retrospect it wasn’t a bad idea to add that stiffness but if I were doing it again from scratch it would make more sense to use 2060 and add stiffness when I bolted a panel underneath.

I’ve realized that if I add 1mm shims on either ends of all the horizontal parts, I can use 1.5 meter extrusion for the lid and front access door, rather than having to cut 2mm off each of them to fit. 1mm (or 0.04 inch) aluminum sheet costs rather more than I would have expected. Maybe I could 3D print a dozen custom shims with plugs to align the with the ends of pieces; that would save cutting 2mm off 5 pieces of aluminum extrusion. Or maybe that’s a silly idea.

So really, pretty much all the rest of the pieces require cutting, and therefore I really need to commit more precisely to dimensions. There’s only one more piece that doesn’t require cutting (assuming I trim the door pieces), but it mounts to pieces that need to be cut, so I’m to the point of needing to decide and commit.

At least one of the 1.5 meter 2040 extrusion pieces came from OpenBuilds with a big ding near the end, and another one came so badly bent that I can’t get a t-nut into the slot. That was a disappointment, since I thought OpenBuilds was considered premium.

The only piece that actually needs v-slot is the 4040, because I’m running wheels in it; everything else could be t-slot. I think that when this prototype helps me nail down the sizes, the next person will be able to upload a cutlist to misumi and might be happier. Don’t know if that would be cheaper though.

4 Likes

The Lasersaur BOM uses Misumi profiles because at misumi you can order each profile in the exact length that you need. You can even have threads on the ends or holes if you need.

3 Likes

Looking good! Are you considering any sheets as siding, or maybe diagonals? This will obviously stiffen the frame even more, making it less prone to vibrations and defacts in the cutline.

and Claudio, we also bought from a dutch shop who cuts and taps as you ordered. Damen CNC, a great supplier for all serious stuff for CNC milling / turning / lasercutting etc.

2 Likes

Bingo! This is in fact exactly why I switched back to working on rebuilding the CNC router: to get ready to cut side panels from masonite to which I can bond aluminum sheet. That should make it quite stiff.

Rebuilding the CNC router involves drilling a bunch of 2.05mm holes (#46 drill) to tap M2.5 for limit switches and terminal blocks. Managed to break a #1 center drill last night for the first time. Felt silly doing that in aluminum; I was just careless. Left the tip inside, so I had to turn the piece over, re-align, and drill out the hole from the reverse side with a cheap 2mm bit before going back in with the #46 bit and then tapping M2.5 which is basically never fun. I have another 20 of those holes to drill and tap in three more plates before I call it done. I might skip tapping the holes that I’m not sure I’m going to use yet and can get to later if I change my mind, but I’m going to drill them all before I put the plates in place.

That was going really well until I got a bit too comfortable and snapped my M2.5 spiral flute tap. I haven’t been doing enough small work lately and got a little ham-handed. Fortunately, I did this in a hole that is not absolutely critical, and I can drill another hole nearby, but I still feel dumb. And I’ll always have the tip of the tap stuck in my router now as a reminder to take it easy.

Switched to #45 bit for the remaining holes (.03mm wider), discovered that I really wanted that set of holes moved a bit anyway so the broken tap isn’t in the way, and learned about alum for dissolving steel out of aluminum, so I drilled the holes where I really want them and might try my hand at dissolving out the broken tap.

But my new tap doesn’t get here until Monday. Haven’t yet decided whether to risk my straight flute M2.5 tap on the rest of the holes in the mean time. Tapped the remaining holes with a 3-flute straight flute tap. Went carefully, used tap magic, got them all done without disaster.

Decided not to try using alum, because that will delay getting the router functional again, and I have zero need to use the hole that I broke the tap in.

Ordered the aluminum sheet for making the side panels. Should be here within the week.

1 Like

Since my Reci W4 is on the way I urgently need to get enough of the frame put together to test the tube during the 30 days during which Lightobject covers return shipping for warranty.

I decided to reduce the available Z depth by ~40mm to make the C-beam fit comfortably over the Q800 chiller when the casters are installed, and used cutoptimizeronline’s bars support to do some quick cutlists. It turned out that for the frame, with the reduced Z there was very little waste! This was inadvertent, but maybe the dimensions I ended up with mostly by accident were actually particularly good! :slight_smile:

I did rough generous cuts on the bandsaw, and then squared and finished to size on the mill. The bandsaw was more square than I expected, and I probably could have cut all the pieces to final dimension on the bandsaw if I had trusted it. But for most of the pieces, I was able to get precise dimensions by using a metric (12mm) endmill, and then use the DROs to add 12mm to the desired dimension and trim the other end.

Then I found out that I had missed installing some slide-in T-nuts, as well having installed at least one in the wrong channel. So I had to take apart the frame a bit to fix that, which was more trouble than one might have expected.

My father was a mechanical engineer, and he said that good engineering always benefits from the universe’s ability to help an elegant design come together by serendipitous means. I just like to call it “accidental engineering”…lol

3 Likes

Having a handful of drop-in t-nuts laying around after you have everything installed is always a good move in my opinion. The best laid plans, and even making a drawing showing where they all need to go, I seem to still leave a few where they don’t belong, and some empty spots where they would be needed.

2 Likes

Yeah, I have a stash of drop-in t-nuts as well. I’m expecting to use them for attaching the sides because aligning slide-in t-nuts with lots of holes in the sides seems like an exercise in frustration, and I haven’t even designed the sides yet. And if one of those fails, it’s likely to be obvious (rattling side noise) and easy to fix (just re-install the sides; they are meant to remove for easy maintenance anyway) so I’m OK with drop-in t-nuts there. I just like slide-in t-nuts where I can for structural things.

In theory, I should have roll-in t-nuts on hand as well for places where I discover late that I missed a slide-in t-nut, but so far I just roll my eyes at myself and find a way to take things apart just enough to get a slide-in t-nut installed.

2 Likes

I have almost all of the frame rough-fit together. Only four more pieces of extrusion left—and I ran out of corner brackets. The best price I could find was about $25 (price fluctuates by a few dollars) for a set of 20, and already put in 100 of them; I ordered 20 more that arrive today and yesterday I realized that I still won’t have enough to finish the doors. That doesn’t include the $30 I spent on three dozen hidden inside corner brackets; slightly cheaper but also weaker. I’m going to double-check that I need the seventh set of corner brackets before placing another order; if I do, that will be approximately $200 for corner brackets alone.

In case anyone was wondering whether DIY is the inexpensive route…

4 Likes

I have those four pieces of extrusion installed. They are two pieces on each side. The two vertical members have almost a dozen attachments each; the most complex pieces on the entire laser.

And one of them I have installed at least six times now. And I think I need to remove both sections again, because I realized there’s one more change I need to make now that I figured out I could rearrange to fit a servomotor in for X!

For the lid and door, I might just drill through and use screws instead of corner brackets. Corner brackets are flexible, but if I don’t need the flexibility it makes more sense to use screws, and the door dimensions are pretty clear. It’s possible that after I’ve validated the entire design and ironed out wrinkles, I’ll decide to take it all apart, drill and tap, and screw it together more permanently.

Later…

Oh, no. The tweak to fit the servomotor, which also makes for a really nice way to mount the X gantry, has cascading effects that require me to redesign the back frame of the laser chamber. This would be one of those reasons to still be using corner brackets and not drilling holes through the frame extrusion yet! I need to sleep on this.

2 Likes

I have the X and Y rails in place, including the linear rail, and the X servomotor to drive it. I also have the outer bed frame assembled though not yet suspended.

I haven’t tested yet whether this location for the servomotor leaves enough room for me to mount the second mirror, though I should do that soon.

1 Like

I now have the Z lift rod at the top rather than the bottom, and the Y rod and belt are below the Y rail instead of above as in the original plan. I’m still playing around with location. I’m currently planning to wind the Z line directly around the rod, and to make clamps to hold the line to the rod. I’m milling something from aluminum for now but I’m not sure that’s the best plan.

Both rods are sufficiently far from the laser plane.

2 Likes

First try at milling clamps ended up too big and interfered with the V bearing block parts. Trying to decide whether to try to save them with some lathe work and more milling or just start over. But not tonight.

Overall, I’ve been trying to avoid a need for lathe and mill to make custom parts here, and I keep finding ways to solve problems with a 3D printer. I was worried that I couldn’t make screws hold tight enough in a 3D-printed clamp but probably using nuts or inserts would solve the problem.

4 Likes

More at Suspended laser bed? [Spoiler alert: No...] on no longer trying to suspend the laser bed, so the whole Z lift rod went away and was replaced by single-start trapezoidal rod and guide rod in linear bearings.

2 Likes