Tag: 3D Printing

Posted on

A 3DPrinter Only as Good as Its Parts

Good Morning from my Robotics Lab! This is Shadow_8472 with a side project for the week. Let’s get started!

I’ve had a hit/miss relationship with my 3D printer. Some of the time, it works great, but when it’s not, I’m having to scavenge a solution to whatever is ailing it.

Fan


One fateful project, I chipped a couple fan blades off in my attempt to service it while in operation; it seemed like a good idea at the time. Don’t repeat this mistake at home. In the case you do, there exist sets of replacement blades. I was able to print one up on my damaged printer, and were it not for clipping right through one of the posts holding the motor in place, it would have worked beautifully. Instead, it would grind loudly whenever I started the printer up. I would always have to tap it a few times until it managed to find its balance.

The only real fix for this problem was a replacement fan. We had to be careful when ordering one when considering its size, voltage, and pin count. We ended up with a close enough match I was able to install it with few questions – there are tear-down tutorials for in-depth instructions. The cord was way longer than it needed to be, so for my installation, wrapped it around the box behind the printhead a couple times before plugging it in.

Bed

And then there is the printer bed. Metal warps with heat cycles – especially when it’s made on the cheap. Aluminum print beds are no exception. Mine has reached the point of unusability having formed itself into somewhat of a slight bowl or saddle. Note that this is after giving it a gnarly scratch right in the middle.

This too took a replacement part. We ordered one for my specific model of printer (Maker Select Plus) and it arrived at the same time as the fan. Judging by follow-up research, borosilicate glass is preferred as a print surface, though cheaper, more thermally sensitive glass like you’d find in windows can also work as long as the operator is careful. My glass plate is affixed with four small binder clips, which I will need to mind as my favorite test print for leveling the bed sends the extruder right through them.

Takeaway

One key element I was not prepared for before printing a Benchy this week was hair spray. Makers love their Aqua Net hair spray. Another common-sense recommendation was not to use the metal spatula that came with the printer as it could damage the new bed.

Final Question

Have you ever 3D printed on glass? I look forward to hearing about your experience in the comments below or on my Socials.

Posted on

I Broke My 3D Printer Making a Gift

Good morning from my Robotics Lab! This is Shadow_8472 with a side project of the week. Let’s get started!

The Game Quilt

My mother is made a quilt for a kid who will be spending his summer in the hospital this year recovering from a painful procedure. An accompanying game board has Chess/Checkers quilted onto one side and Tic-Tac-Toe on the other. The idea of paper Chess pieces glued inside some bottle caps grossed me out. I demanded to make something a bit nicer. I browsed the available Chess sets on Thingiverse and settled on a short “travel” set [1].

The Black Army

My printer started the week loaded with black filament, so I printed the black army first. I scaled the pieces up to 150% to match the board and made the bases solid to lower pieces’ center of mass. I tested these changes with a queen and the result needed a larger crown topper that wasn’t so sharp and fragile-looking.

With those changes in place, I started printing, but pieces kept peeling/detatching, resulting in some… spectacular failures that often involved massive misalignments. I made it through four usable batches of two pawns each, but a couple were damaged during removal (my sister and I had similar ideas involving channel locks on batches 3 and 4, respectively).

I printed a bishop, knight, rook triplet and the 10% infill I had been using did not support their flat tops. The bishop was usable, but a little warped on the base. Nevertheless, all three had performed decently, so I bumped it up to 15% infill and tried printing the back row pieces minus knights. Print failure after print failure. Small brims curled on the first layer, even with a high temperature first layer trick I tried.

Finally, I programmed some brims so big they would touch, I also tried wiping the bed with some acetone. Another print failure, but this time only because it stopped extruding after a beautiful, no-curl first layer. I took the printhead out and cleared out a clog, only for it to clog at least one more time. Eventually, I traced the source of the crud not from my spool of filament, but to the e-step motor assembly and even fashioned some tools out of my failed prints to navigate the tight quarters in the mechanism to clean it out. The next print came out nicely.

I knew from the beginning I wouldn’t be happy playing with K’s for knights. Where most of the pieces were just the heads of a more traditional set those knights – YUCK! I appreciate the need for a simpler geometry for small prints, but K is for king in Chess notation where N is for knight. I found my way around Thingiverse to a decent knight designed to be printed without supports [2]. It was a slight adventure putting its head onto a base like the other pieces, but a doable challenge nonetheless. I printed two up to complete the army.

Filament Change and the White Army

I stayed up late to get the knights off as soon as possible and get the white army printing so they would be done before my writeup (an estimated 14 hour printjob). I should have slept on it. Mistake 1: using the firmware’s unload feature; the plastic overheated and stretched off inside the printhead, denying the cleaning filament admittance. Mistake 2: disassembling the printhead while it was on, in a shadow – I snapped two fan blades in successive attempts to engage with a bolt. Within ten minutes, I had found a replacement set of fan blades [3] and quadruple checked the counter clockwise variation was the file I needed. I printed it, enduring the much louder fan as it choked out one last print.

In the morning, my father advised we pass on gluing the original blades back on in the hopes of printing a better set of replacement blades. I clipped the remaining blades and used the fan itself to sand the nubs smooth. My father glued the blades on and we had to squish the fan’s housing to correct for Mistake 3: clipping one of the fan’s three spokes. I printed a Benchy tugboat to satisfactory results.

After all that drama, the white army came out usable with one knight warping a little and one pawn nicked during removal. I plan on adding a couple extra queens in each color because I know from experience that they make for more exciting endgames.

Takeaway

Don’t work on machines while tired, or in the dark. The fix I found is only a patch job until I can buy an upgraded replacement. It’s not as balanced as one made to factory standards, but it is good enough for decent prints on a temporary basis.

Final Question

What self-inflicted accidental damage have you caused by a series of dumb mistakes?

I look forward hearing your answers on in the comments below or on my Discord server, where I sometimes share exclusive nuggets that didn’t fit into the main post.

Works Cited

[1] Raukk, “Travel #chess,” thingiverse.com, March 22, 2012. [Online]. Available: https://www.thingiverse.com/thing:19754/comments. [Accessed June 13, 2022].

[2] Zarlor, “OpenSCAD Chess Simple Printing,” thingiverse.com, Jan. 26, 2019. [Online]. Available: https://www.thingiverse.com/thing:3381939/files. [Accessed June 13, 2022].

[3] CreativeTools, “Cooling fan replacement blades,”thingiverse.com, Nov. 19, 2013. [Online]. Available:https://www.thingiverse.com/thing:186979. [Accessed June 13, 2022].

Posted on

I Built a Model Telegraph

Good Morning from my Robotics Lab! This is Shadow_8472, and today I am building a model telegraph. Let’s get started!

The First Telecommunications

The electric telegraph was invented not long after the discovery that electricity and magnetism were two faces of the same fundamental force of nature. While the image of someone tapping out a message in Morse Code may involve the more iconic piece of hardware –the key– the real heart of the technology is located in the sounder: an electromagnet controlled from miles away. This was what powered the first ever near-instantaneous telecommunications networks crossing America and later the world.

Only limited by the speed of an operator and final delivery, a message from the front lines of the American Civil War could reach Abraham Lincoln in hours or minutes instead of potentially days. It put the Pony Express out of business almost before it got started. It remained unchallenged until Alexander Graham Bell invented the telephone by sending enough information over the wire to reconstruct a human voice – and in fact, smartphones and other personal computing devices still contain the humble electromagnet to deliver convert electrically encoded messages into sounds we can understand.

3D Printed Telegraph Key and Sounder

This project was sparked a post by Mattosx on instructables.com[1]. I’ve already spoken about the issues I had printing this project. Between all the totally failed prints, a case of broken filament, and the slightly peeled pieces, I am lucky to have two sets of parts, though some may need reprinting.

Researching a Group

I have always been at least passively curious about learning Morse. I started seriously looking into it this month. Who all still uses dots and dashes to communicate? Ham radio operators, mostly. I neither have the equipment nor the desire to get the license I would need before participating in clubs using Morse over CW (communications waves).

While musing that I might connect my telegraph over the Internet, I came across Morse KOB on GitHub, apparently the only software tool out there for such a purpose. I reached out to Ed Silky, the only developer I could find contact information for. I figured I might as well so I eventually have a point of contact by the time I finish the Instructable. He responded by the next morning.

Ed shared a wealth of information I had only started piecing together. He called my attention to a critique in the Instructable’s comments, that the sounder doesn’t signal the end of a dot/dash clearly as or distinctly from the beginning of one. Ed warned, “Also, be aware of the differences in the alphabet between American (Landline/sounder) and International (Shortwave/tone / Light) code:” and listed eleven characters and all ten digits with their differing encodings; the earlier American Morse features such things as intra-character spaces and extended dashes – making a properly functioning sounder crucial to understanding American Morse in particular [2].

Assembling the Key and Sounder

Mattosx’s Instructable looks great… until you’re actually part way into assembling the thing. The first red flag is a privated video titled “3D Printed Telegraph Sounder Video” with no written instructions outside the parts list.

The Instructable I’m following is crippled, but I’m limping through it. The 3D modeling isn’t the cleanest on a couple parts – and in fact is missing a piece on closer inspection. I had to creatively chop the parts apart in Slic3r to get them to fit on my damaged bed. Without that video on the sounder assembly, we had to guess at the size of magnet wire, what direction the disks were pointing (in or out), even what order to assemble the different parts in. It’s an interesting idea to make the sounder handle its own electromagnet, but that optional part is unapproachable without any experience working with circuits.

My father and I wound the electromagnet over several sessions, with the internals getting progressively less tidy as mistakes compounded upon each other. By the end, I was just trying to fill in low spots. We hooked it up to two D cell batteries by hand and nothing. We tried a multimeter, but that turned up busted, so I’m left with a project to continue some time in the future.

Takeaway

Based on my incomplete attempt to build this thing, here are my recommended modifications (note that I have not actually tried these yet):

  1. Install magnets, using sandpaper to make the holes a little bigger if need be.
  2. Screw in and wire up the key base before installing the key.
  3. For the sounder electromagnet, search out some sort of spool winding tool to print.
  4. Start with the nubs facing outward instead of inward.
  5. Dry fit the spool so the upper plate lines up with the tops of the base’s posts.
  6. Use a power screwdriver to grab the screw right where the upper plate would otherwise land.
  7. Once you have the electromagnet wound, assemble the top of the sounder.
  8. If all the screws are so loose they fall out, use a snip or few of magnet wire to crowd the screw hole (untested, but this might be a good use for any lines of filament extruded as part of bed leveling prints).

This was my first larger monthly project. It didn’t meet my personal deadline of having a working model to post today because I failed to plan. I only had magnets in hand with a week to go, and I didn’t leave myself any time to diagnose problems as they came up. I now have another project to cycle through as time allows.

Final Question

Have you ever made an electromagnet by hand before? I look forward to hearing your answer in either the comments below or on my Discord server.

Works Cited

[1] Mattosx, “3D Printed Telegraph Key & Sounder,” instructables.com, (no later than March 12, 2019 [3]). [Online]. Available: https://www.instructables.com/3D-Printed-Telegraph-Key-Sounder/ [Accessed April 25, 2022].

[2] E. Silky. “Re: MorseKOB,” “Re: Telegraph Build,” Personal emails (April 7, 2022 to April 22, 2022).

[3] T. Nardi, “Old Meets New In 3D Printed Telegraph,” hackaday.com, March 12, 2022. [Online]. Avalable: https://ieee-dataport.org/sites/default/files/analysis/27/IEEE%20Citation%20Guidelines.pdf [Accessed April 25, 2022].

Posted on

3D Printing: Trial of Errors

Good Morning from my Robotics Lab! This is Shadow_8472 with a side project of the week. Let’s get started!

I have been 3D printing parts for this month’s large project, but things haven’t been easy.

Keeping my print bed level is probably chief among my challenges. I keep having adhesion issues from the bed being too low and squishing issues where there’s no room between the print head and the bed to lay down plastic – sometimes on the same piece. I’ve been using a pattern that prints concentric squares a test, but sometimes I follow up a successful test with a production print that only leaves a little residue so thin it doesn’t scrape up properly.

Even when the first layer sticks, prints can still peel part way through the print. One piece came out mostly fine except for one malformed corner that may be fine. I’m re-printing because I’m after a second set anyway.

The most exciting failure was when I had three black pieces successfully on their way, and I noticed when there was already a sizable gap between parts and print head. I aborted the print. The filament was stuck; it wouldn’t go in or come out, even at temperature.

I took the print head apart, I found the chewed up filament almost broken off. It refused to come out, even when I yanked on it with the pliers. I wound up trimming what I could get at and ramming the rest through with with an Allen wrench. I never got a positive identification on the clog, but as the plug gave way all at once, I remembered a fuzzle I spotted riding on the filament, and I thought nothing of at the time. I followed it up with a little cleaning filament for good measure.

I took the opportunity of the disassembled print head to at least check my flow rate. The machine wanted the extruder to be at least 160 degrees C. I advanced my e-steps ten times and got a 100 mm segment, give or take a millimeter. Assuming it was trying to advance a centimeter at a time, my printer is healthy in that regard without me having to do anything.

Final Question

What is your most entertaining manufacturing mishap?

I look forward hear your answers on in the comments below or on my Discord server.

Privacy Policy