ProDarwin's adventures in 3D Printing

*flashback time*

The year was ~2010ish.  Hobby-level 3D printers were just starting to gain some traction, but not widespread by any means.  I think Makerbot was just starting up, and Reprap based stuff was the most popular.  I had a lot of fun with a Dimension 1200ST at work.  It printed PC-ABS with dissolvable supports, and 90% of the time make cool prints with excellent material properties.  I had this wonderful delusion that if I got my own home level printer I could make cool stuff, make money off it, and it would generally be awesome.

Narrator:  It wasn't awesome.

I picked up a BFB3000 Printer.  This was around $3500 IIRC.  This was effectively a pro-grade build of a Reprap.  It. Sucked.  I fiddled with it during my free time for months and never really got any good parts out of it.  Getting the bed level was nearly impossible.  The slicer software available at the time was pure garbage.  So after a few months of frustration, I sold it off at a mild loss and moved on.

*flash to 2022*

At work I've been using Objet/Polyjet printers, Fortus, Carbon, etc.  All 6 figure machines.  Some are better than others, but several are work horses that run nearly 24/7 and crank out excellent parts with zero thought put into how to make them. 

Individuals on forums I am part of have been picking up hobby level 3D Printers for absolutely silly cheap prices and getting good results from them.  Some of my coworkers have as well, and I'm talking about people who are not technical.  I've been watching from a safe distance, admittedly gunshy after my first experience, but recently decided I am comfortable trying again.

So with the help of some people here I convinced myself to go buy one.

 

The current state of printers is ridiculous.  The Ender, which is Reprap based, is basically a commodity.  You can buy any number of upgrades off Amazon and have them at your door the next day.  Its super well understood.

I ended up going with an Ender 3, that included a tempered glass bed.  A roll of filament, a tubing/feeder upgrade, and a cover for heated builds.  For less than 10% of what my first printer cost.  This is the order sheet:

My goal is/was to be able to run ABS, but first I want to get good at handling PLA since it is considered easier to work with.

Assembly went smoothly, but I skipped the upgrades at first.  For my first print I chose a 3D Benchy.  It failed only a few layers in.  Apparently it was an extrusion failure.  Upon close inspection, the drive cog the moves the filament was pushing, but not gripping the filament.  I manually advanced it and was able to get it working again.

At this time I switched to the glass bed, but ended up chasing my tail for quite a while trying to get it good and level.  The binder-clips that hold on the "stock" bed wont work on the glass one, so new clips were needed.  In addition to that I followed the recommendation to switch to heavier duty leveling springs and adjusters.

After a couple amazon orders I was finally able to get a properly adjusted level bed and print the benchy:

After printing the Benchy I decided to do a few more quick test prints:

During the Benchy print above I was able to get a feel for the current state of "slicer" software.  Holy. berkeley.  The change from 12 years ago is even more staggering than the advancement in 3D printer mechanicals.  Its super straightforward, there are tons of presets, good control over almost every aspect of the process, nesting, accurate time estimation, etc.

That said, its not all rainbows and unicorn farts.  A hobby level FDM won't print exactly what I design.  I needed to take some intermediate steps. 

On my first print that actually serves a purpose, I intended to put some threaded brass heat-set inserts into to the part.  This meant thickening the wall locally around each hole where they would be located.  It took me a few tries in Prusa Slicer to master the art of using a "modifier" to achieve such a thing.  Even now having figured it out, its a really clumsy process if your part isn't designed in metric.  And because I engineer all day in freedom units, its just easier for me to remain in that world.

Here you can (barely) see the locally thickened walls around the bosses.  In retrospect I should have thickened them more.  But also I am not well versed in using heat-set inserts with PLA and I didn't find much information on it with a first pass.  In my experience with Dimension a long time ago, I could design for them exactly as if I was doing an injection molded part.  That doesn't seem to be the case here, but I plan to do some experiments to understand what the best practice should be.

Anyway, I was able to properly insert the inserts.  I did redesign the part relative to the model I found on Thingiverse.  The original design had the inserts on the top side of the bottom piece, meaning load forcing the blocks apart would cause them to pull out.  I repositioned them to the bottom side of the block which is a much much stronger way of using them.  Unfortunately the modeling required using SubD modeling to do it because the author didn't upload using a solid format and uploaded a mesh instead.  Much more of a pain to edit :(  I'm not sure why people posting content to free sharing sites do this.

I did some clean out on the hole afterward and was able to assemble my parts.  I think these all together took about 20 hours of printing.  These are replacement brackets to strength the way the pedal assembly mates to a playseat challenge.  Such a huge improvement.

One more thing I printed:  A spacer to inside the torque limiting engine mount of my Kia Soul.  The factory mount has a huge void in and is a really poor fit for a manual transmission (even a brand new OEM one).  I tried a polyurethane mount there, but its too solid and rattly (although it does drive much better).  I ended up keeping the poly in one side of the mount, putting the rubber back in, and putting this spacer under the rubber to fill the void:

Yesterday/Today's work:

I made myself a bracket to mount my massage gun to my weightlifting rack so I can actually reach my back muscles.  I currently have a tight spot in my right lat, causing a pinched nerve, and its no fun.

This was my first print ever with supports made of the same material as the main print.  They were... not fun to remove.  Also the control over them is super difficult.  The hinge side of the bracket has gaps between one part and the next for clearance.  I was expecting the slicer to just put a small amount of support material in those gaps, but it really wants to fill in all around them with a huge mess that seems totally unnecessary.
 

I have a lot of experimenting to do.  

Regarding the actual designed part, it worked out ok.  I need to strengthen the next revision of it, probably with slightly more offset, higher infill %, and hexagonal instead of rectilinear infill.  Its going to double the print time, but it should be worth it.

I also need to not be stupid and put heatset inserts in an orientation where I can access easily with a soldering iron.  Doh.

Update:

This thing has been great at chewing through PLA.  I've printed various things and gone through about 3kg of material with no failures. I did, however run into some warpage on some non critical parts (board spacers for re surfacing my deck).  The bed-side of the parts seems to warp up a noticeable amount compared to the top surface of the part.  I'm guessing because it is cold out now and the printer is in an unheated room? (~52F when I checked today)

Anyway, since I have a tent for it for doing ABS I decided to set that up today to see if it would help with temperature stability.  When printing PETG I measured a solid 70F inside the tent, so I guess that's better?  The tent itself is a pain in the butt though.  You have to practically dismantle it in place to start filament from a new roll.  Not a great design.  I'm immdediately re-thinking how I will approach this.  I'm thinking of just boxing in the section of shelf its located on (~4'x2') as it would give me room to get in and poke around.  I'd just need to fab up some sort of door/window.

Today was my first attempt at PETG.  Part #1 was a benchy obviously.  It looks great, but it actually adhered to the bed too much - as in more than it adhered to the layer of the part above it.  I'm not sure what would cause this/how to correct.  Print temp was 240, bed temp was 70c, room temp was 70F in the enclosure.  I'm inclined to think that a lower bed temp would help, however the spool of material actually recommends 80-90.  I used the default Prusa Slicer "generic PETG" settings.

How are you embedding the images in your posts? A lot of them aren't showing up 

Ugh.   This is super frustrating.  I have a post about it somewhere as others have had the same issue.

Usually copy paste, but sometimes drag and drop.

If I browse this thread on my phone, there are only 5 images. 

If I browse from my personal computer there are 14.  

If I browse from my work computer there are also 14.

My thread about it:

https://grassrootsmotorsports.com/forum/off-topic-discussion/pictures-appearing-on-pc-but-not-on-phone/198428/page1/

ok.  All the images are fixed :)

So I have spent a decent amount of time trying to dial in PETG settings.  Recommendations are all over the place. Its pretty amazing.  PLA is practically plug and play.

So I made myself a little test part (cut out a piece of a part I was having trouble with), and decided to run it using a variety of settings.

Round1:  Nozzle and Bed temp.  I changed settings and re-exported the G code with Nozzle temps of 220, 230, 240 and with bed temps of 60, 70, and 80.  I didn't try them all, but the higher temps gave me better prints.  I landed on 240/80 for now.  I recognize this is in the ball park and I will likely have to come back for finer tuning.  This is with first layer speed set to 50%, and small perimeters set to 50% as well.   First layer is 0.2mm, all others are 0.24mm

This had really good adhesion in the middle, but definitely screwed up a tad on layer 1 on a few letters, and clearly stringing is a real issue. 

Currently running a few tests all at 240/80 with a couple more changes:  1) Layer 1 is 25% speed (10mm/s).  Probably too low, but I can bring it back up later and 2) a few different Z offsets from 0 to -0.05mm as I read pushing the first layer of PETG into the plate can bet helpful to get that first layer to stick.  One thing I may try on this round also as I am now reading about it is increasing first layer height .3 or even higher as I have read doing a beefy first layer height on PETG makes things easier.

I didn't update this, but what I printed was a bubble bath dispenser.  Turns out PLA was just fine for it anyway, so all of that PETG nonsense wasn't really needed.  Still learned a few things.

Some other stuff I have printed recently:

Failed this print about 90% through:

Printed the last 10%

Brought out the ol' plastic welder (soldering iron) and they are as good as new.  These are for the hydroponics tower I have and nobody can see in there anyway.

This finally cleared up the extra dyson attachements that have been sitting on top of my dryer for 3 years.  There are a lot of variations on thingiverse of these but none I liked.  They all hung off the original mount and most didn't have room for the motorized head, so I made my own that mounts to the stud just above the "dock" and its very solid.

I finally got around to snagging one of these off Amazon the other day for $40

My board was a 4.2.2 so this isn't a huge upgrade, BUT the quiet stepper motor drivers do make a huge difference.  Now the noise of my printer seems to be about 90% fans.  Trying not to go down the rabbit hole chasing sound levels.  The sound fans make wont penetrate through a door, but the stepper motor squeal could be heard several rooms away, through walls and floors.  Would not be conducive to apartment life.

Speaking of rabbit holes... my bed platform (not build plate) is not perfectly level.  The glass build plate is kind of pulling it flat.  I want to go to a PEI build sheet, but I can't really do that without a flat bed... thats another $60 to get a quality flat bed platform (unless anyone knows any tricks?).  

Sooner or later I'm going to have Bambu levels of $ invested in this, which will be just silly.  We picked up a few Bambu Labs X1 at work, and they are very impressive for consumer grade harware for <$1500 all in.  I just wish they would make one double the size (in x/y at least).

Prusa XL might be worth looking into it you want something that's twice the size of the Bambu (maybe more) but has way-better-than-Ender levels of quality and dependability. It's also about twice the price of the Bambu. 

I'd imagine PEI plates can deal with a little bit of bowing in the factory bed mount (bed sled?). Lord knows they pretty much have to, I doubt there's a perfectly flat (or near it) Ender plate in existence, aftermarket or no. 

Prusa XL is about 50% larger.  I need 100% (for work stuff, not home stuff).  Its a weird hole in the 3d printer space.  Ideally, 24" x 36" x say 16"Z  - very few manufacturers make anything of that size in the consumer or prosumer space, even though stuff like Prusa XL and Bambu should scale to that size quite easily.

Anyway, RE: PEI plates, my concern isn't that they can't deal with warp, its that I wont be able to get a flat bed.  The only reason my bed is flat now is because the glass plate is pulling the bed below it flat - a PEI plate isn't nearly rigid enough to do that.

That's interesting. Maybe a Voron? I don't know how much freedom you have setting printing volume when you build one. Also, you have to build it yourself, to include printing parts and wiring. Also, cost ranges from pretty reasonable for what you get to utterly ridiculous but extremely capable. 420 mm/sec. Sub-6 minute Benchy, anyone? https://youtu.be/JdVZZ4i2dS8?si=-nqHkomUBnvk4Upo 

Maybe one of the conveyor-style printers? I don't think the CR-30 gets you to all the dimensions you want, except for one. The iFactory One gets you close to all three.

Ah, I see what you mean. Why PEI? I do everything with glass and glue stick now (very rarely I'll use hair spray). I usually separate prints by hand or by putting them in the freezer for maybe 10 minutes. Not a lot of issues with adhesion, whether it's working for me or against me. I've only needed a chisel once. I like friend's printers that have PEI beds (Voron, Prusa) but the Ender might just be a bad candidate.  

I like that with PEI you can just bend the sheet, pop off the print, wipe it, and get started again.  With the glass plate if it doesnt pop off super easily, I have to take off the clips, remove the plate, do that all then reinstall the plate with the clips.  Its time consuming, but also I think it contributes to it going out of level more frequently than a PEI sheet which is just a simple magnet.

I havent had adhesion issues with the glass and PLA really.  Not once I got it figured out anyway.

Yeah, that's understandable. I've wondered at just how much of an impact removing the glass plate has on level. I'm sure it's not zero. This also assumes I manage to resist the urge to try and pop the print off while the bed is still clipped in (I can about 75% of the time now).

I've definitely hit the point where I don't want to upgrade my Ender anymore. It's great for what it is but I'd rather save for something that's got good features and isn't a tinkerer's delight.  

Turbo_Rev said:

I've definitely hit the point where I don't want to upgrade my Ender anymore. It's great for what it is but I'd rather save for something that's got good features and isn't a tinkerer's delight.  

Yeah, this is exactly my concern.  At some point it doesn't make any sense to put any more $ into it.  A creality K1 isn't that much more and has almost all of these issues solved right out of the box.  Same with a P1P, or a X1.

Recent prints update:

Shaver holder (right pocket is where my fitbit charges when I'm in the shower)

Bathroom Caddy (this is oddly gamechanging when it comes to keeping the bathroom clean):

Lampshade.  This was more of a grasshopper experiment, but it makes a cool accent light.  Next I'll do some vertical versions of it, hopefully with a triple pendant.

Not sure if anyone reads this, but having a big problem with large prints right now.  Its only making it an hour or two into the print before the extruder chews through the material.

Looking at a failed print, there is either stringing or poor adhesion.  

Reading a few posts online, I think this is whats happening:  The hot end fan is failing.  It works fine on shorter prints, or prints with more open area where the extrusion isn't as constant (?), but on a long constant extrusion print, the fan isn't quite getting the job done.  Hot end overheats -> filament melts inside the heatsink instead of just in the nozzle which results in a clog -> extruder gear chews through the filament.

Thoughts?  If thats the problem, recommendations for a hot end fan?

Failed print:

Anytime I have issues like that with my Ender, I change the nozzle and Bowden tube. Usually fixes it.

Do you have an all-metal hot end?

No?  I'm not sure what exactly an all metal hot end is.

i have the metal extruder parts and Capricorn tube shown in the first post.  I just put on a fresh nozzle.

The thing that makes me agree that it's heat build up is how a shorter print like a Benchy is totally fine.  

You probably already checked this but is that removable rubber insulation on your heating element intact? Mine had maybe 10% missing a while back and it could absolutely not keep the filament hot enough at normal printing temps. I replaced it and hey presto. 

In reply to ProDarwin :

From the factory, there's plastic tube all the way to the nozzle. My Ender used to heat and deform that tube on higher temp prints, which would cause a clog. A hot-end with metal through the heat sink fixed it. 

Any recommendations for a hot end?  FWIW, this is PLA, so it's not a "high temp" print