Sizing extrusion for a printer build

I've seen a lot of debate online regarding what size of aluminum extrusion is appropriate when designing a printer, but nothing definitive or backed up with actual numbers. Has anyone run FEA on various diameters of extrusion and measured their deflection? It would be nice to have a table or something on the wiki that listed out suggested minimum extrusion sizes for a given printer design/bed size. It would pair well with a discussion of joining methods, I know 8020 has ratings for various joint types buried on their website but it would be nice to have that information here.

You can check moment of inertia for comparison. You can calculate sagging with those equations.

Equations are not appropriate for real life examples because the load distribution is more complex. Also those equations refer to static load (not dynamic).

I think for reprap it's the best to find good existing designs with similar dimensions and use it as reference.

Numerical methods with dynamic loads are not easy to do, and it should be checked with experiments anyway if you have no experience.

You can find 4040 extrusions in UMMD with 300x300 printable area which seems enough.

I have used 15x15 extrusions for a large format lasercutter COREXY build and closed up the box with 3mm plywood. The plywood adds so much rigidity that the 15x15 beams don't flex at all. So keep in mind that if you close up the box design you can perfectly use 20x20 extrusions. If it's an open frame it becomes a different story.

Also keep in mind that the corner connectors play a big role in stability of the squareness of the frame.

---

http://www.marinusdebeer.nl/

I used 40mm stuff because I got it at a local scrap yard for $2 per lb. IRIC, UMMD's frame was about $60 worth of material at that price. I would use it, even for small printers, because it's impossible to build a frame that's too rigid. I have a pile of 45 mm square material now and may be starting another build using it, soon. Even if you worry about the making the moving parts as low mass as possible so you can print as fast as possible, there's no reason not to make a super rigid frame. Adding a skin of acrylic, PC, plywood, sheet metal, etc., greatly enhances rigidity, and by enclosing the machine (and maybe adding a heater), you'll be able to print ABS reliably.

In my experience, there's no reason at all to print PLA or PETG if you can print ABS. Actually, I don't see any reason to print PLA, ever. It isn't any cheaper than ABS and the prints/filament require such a benign environment that the stuff is all but useless for anything you want to last. It can't tolerate any heat, and filament and prints soak up moisture and become brittle. PETG is a little better in terms of temperature, and is less brittle, but the stuff soaks up moisture like a sponge. Keeping the filament dry enough to print with is a problem.

I see a lot of printers with corner brackets everywhere. Maybe they're needed to hold the frame members square, but I have doubts about them doing much for rigidity in most cases because the frame members, especially small cross section frame members, flex. T-slot material is pretty cheap stuff even if you buy it new. I'd bet that in most cases, the cost of the extra hardware to try to make small cross section t-slot frames rigid exceeds the cost of simply using larger cross section t-slot, and maybe paying for the premium cutting service that guarantees the ends will be cut squarely. If you have square end cuts and matched lengths, all you have to do to assemble the frame is tap the ends of the pieces, drill tool access holes, and put the frame together using 1 screw at each corner joint. No special, expensive hardware is needed. A tap is cheap and you can drill aluminum with any drill. Tool access holes don't have to be accurately located and can be drilled by hand.

I see a lot of frame structures that look like this:



but with corner brackets and T brackets all over the place. Why? Those two inner cross members are going to make that frame as rigid as can be (assuming the frame pieces are matched lengths and squarely cut) without all the extra hardware.

Edited 2 time(s). Last edit at 06/15/2019 08:59PM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
the_digital_dentist
In my experience, there's no reason at all to print PLA or PETG if you can print ABS. Actually, I don't see any reason to print PLA, ever. It isn't any cheaper than ABS and the prints/filament require such a benign environment that the stuff is all but useless for anything you want to last. It can't tolerate any heat, and filament and prints soak up moisture and become brittle. PETG is a little better in terms of temperature, and is less brittle, but the stuff soaks up moisture like a sponge. Keeping the filament dry enough to print with is a problem.

I hold a different view: there is no reason to print ABS when you can print PETG for things that need extra toughness or heat tolerance, and PLA when you don't. This is because of the bad experience I had printing ABS. A warm enclosure is mandatory if you intend to print anything in ABS that is more than a few mm thick. I solved this with 2 plastic bags over the printer. But the worst thing about printing ABS for me was the terrible cough that it gave me. When I read about the particulates that are given off when ABS is extruded, and thought about the damage it would be doing me, I decided not to print ABS again without a fume extraction system and particulate filter fitted to the printer.

I have had problems with PLA becoming brittle overnight when off the spool, requiring me to unload a little filament, cut out and throw away about 1m of filament, then reload it prior to the first print of the day. But this happened only with cheap unbranded PLA. I suspect it was caused by some additive drying out rather than moisture. I have never had it happen since I switched to RigidInk PLA.

Right now I am printing with a reel of RigidInk PETG that has been sitting on the shelf in my office for about 2 years, unwrapped. I also have a Crane Quad that prints with 4 mini reels of PETG, which have been sitting on the printer for 9 months. No observed problems with moisture absorption on either printer yet.

Quote
the_digital_dentist
I see a lot of printers with corner brackets everywhere. Maybe they're needed to hold the frame members square, but I have doubts about them doing much for rigidity in most cases because the frame members, especially small cross section frame members, flex.

Adding corner brackets made a huge improvement to the rigidity of the base of my Scara printer (see my blog), although this wasn't the same situation that you are describing. I agree with you that cross bracing or panels bolted on to the frame adds much more rigidity than small corner brackets can.

Edited 1 time(s). Last edit at 06/15/2019 04:26PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Interesting. My PETG experience was printing a full kg spool over a few days and by the third day, the print surface quality went to hell because of moisture absorbed by the filament that I left on the printer. I also found PETG prints to be pretty hairy- a minor annoyance.

I've read up on the particulate and VOC production from ABS printing, but never had any issues other than the smell. Of course, that doesn't say anything about long term effects. I often print overnight and the printer is enclosed, so by the time I pull the print out everything has cooled and maybe the particulate stuff has settled on the printer walls and bed.

I recently worked on an electrostatic precipitator for the nanoparticles modeled after a device I saw in a research paper on using electrostatic precipitation and soft x-rays to eliminate nanoparticles. Data in the paper showed that a low voltage ESP (around 6-8 kV) and soft x-ray source was about 100% effective, and a higher voltage ESP alone (>10 kV) was equally effective at removing nanoparticles, but also produced ozone. Other papers examined the chemistry of ozone and VOCs and other compounds including things like household carpeting, cleaners, etc., and found that while ozone tends to break down some VOCs (and kills many odors), it seems to create at least as many problems as it eliminates, so I shelved that project. The ozone reacts with almost everything, and apparently some of those reactions produce nanoparticles (I have no idea what is happening physically). Odor or lack thereof is no indication efficacy, or the presence of hazardous compounds. I know PETG doesn't smell like ABS, but that doesn't mean it's any safer.

Here's a look into the precipitator- you can see the blue glow of corona discharge that produces ozone:



I'll have to try some more PETG. Maybe I had a bad spool.

Edited 1 time(s). Last edit at 06/15/2019 09:04PM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

@digital_dentist, perhaps it depends a lot on the quality of the filament. The only ABS I have printed with was cheap unbranded stuff. Maybe I would have had better results with branded filament. I haven't used tried unbranded PETG.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
the_digital_dentist
I used 40mm stuff because I got it at a local scrap yard for $2 per lb. IRIC, UMMD's frame was about $60 worth of material at that price.

I can definitely vouch for scrapyards. Last week I went to the scrapyard for the first time in my life and there were enormous amounts of scrap machines laying about. I'm going to check those places every week now because there might come a time when they rake in some sweet miniature linear rails and extrusions. You only pay the kilo-price. Insane!

---

http://www.marinusdebeer.nl/

I'll throw in my 2 cents here.

The corner brackets don't do much good. I have a 300mm bet Cartesian that is a scratch design by an ME friend of mine. It shook like the devil when I tried to crank the speed up (those direct drive heads WEIGH A TON). I added corner plates everywhere I could reach. It helped a little, but I fear actually did little more than make me feel better. I have done quite a bit of redesign on the belts, pulleys and a lower weight exchangeable print head that has improved the machine, but... My next adds will be 3mm Lexan sheets and mounting it to an OSB board (plywood just isn't all that flat...). I love having a large bed printer (yeah I know, some of you guys have 400 or 500mm ones, but my 300mm seems large to me!)

That large bed build used 1" Al extrusions (American ME's think in Imperial units) and the material I inherited was not all that consistently cut, so butted ends bolted did not work out for me and I had to use the right-angle brackets and mix-n-match. Which is why I added the plates.

I print ABS in an enclosed Delta printer and get decent outcome, but ABS shrinks so much I have to make allowances in my designs for precision designs to line up properly. Not so fun since that shrinkage seems to change not only between brands but also between spools. It could be that eSun isn't that reliable. PLA works well for toy designs as long as they aren't too small because tiny details tend to break off of PLA models.

Lately I have been printing mini-figs using TPU with shore hardness values of around 95. It prints small details surprisingly well and swords do not break off! Care has to be taken for those latter swords so they don't look like soft-serve ice cream. This makes bullet-proof table top models. You can print this easily on deltas too, just at a little lower speed.

My favorite filament is PETG. It is stronger than ABS and far more heat tolerant than PLA. My only problem is that ABS and PLA come in every color under the sky. Not so much PETG. Maybe that will change, I hope so! PETG has stringing issues and buildup can carbonize on the nozzle if you aren't careful, but overall the low shrinkage, high strength and lovely shine sells it for me. With care, I can thermoform it nearly as easily as I can PLA.
I make all my printer pieces in PETG now because I can count on the screw holes lining up more so than I could with ABS and I get no warping.

I keep my filament in sealed tote's with desiccant decanters when not in use and don't leave them on the printers when not because lots of filament get fragile/crunchy when left under torque a long time - some extruders actually rotate the filament while extruding, eventually those break off! I also live in Colorado, near the mountains - humidity rarely gets over 30% here.

A university study found that even PLA spews micro particulates into the air which makes a small room with such a printer in it not a very healthy place to hang out. My solution was to get a room air purifier that is oversized for the room. My choice is the Wagner Switzerland HEPA filter unit that sells on Amazon in 350 and 400 sq ft sizes for under $200. This thing sucks the junk out of the air pretty darn well. I can report no scratchy throat while hanging out there for over an hour during printing sessions. My lab is about 200 sq ft and I use the 350 sq ft model.

Sorry for the long-winded response, this thread seemed to touch on many topics that I have a little experience with...
DLC

---

Kits: Folgertech Kossel 2020 upgraded E3Dv6, Anet A8 upgraded E3Dv6, Tevo Tarantula enhanced parts and dual-head, TronXY X5SA Pro(E3DHemera).
Scratch: Large bed Cartesian, exchangeable heads, Linear slide Delta, Maker-Beam XL Micro Delta, 220x220CoreXY.