Kept at the printhead

Today I did a bunch on the printhead (well, in the hour I had to work ;b)

Here you can see the adxl345 mounting, as well as the hole to route wires through

Here's how the wires are gonna go

The lil lever looking thing is to zip tie the wires to so that there's no tension on them

Things are going south

It all started Friday the 14th. I posted a design of my Y axis in the #3dp-circlework channel on the slack, asking for suggestions as to how it looked

boy, did I get help

and note that in no way do I think that the guys who helped me are the cause of all this trouble--without them, I'd be making some really bad mistakes. For instance, my Z axis wasn't super stable. I have to use a 4040 on top and extrude it all the way to either post (see the images below for reference on what it currently is) to make it nice and sturdy. This does bring the cost up, but it's not too bad. Or so I thought.

Anyways, another issue was that the Y axis was really really unstable, and wouldn't survive like any accels. The main reason for this was that it was totally unsecured in the back. Not a very big brain on my part. Next, I had put 3D printed parts in between the aluminum extrusions, in hopes of being able to extend them in lengths under 50mm (the AliExpress standard length multiple) while being able to fit the electronics inside. This is a bad idea, and I should've known. A suggestion I was given was to mount the electronics in an external box, and not in the base of the printer. This would allow me to add a heated enclosure later on, and to save me the trouble of trying to fit electronics into the frame like a puzzle. Another thing was that, by using the Peopoly Lancer Medium Melt Zone, I was limiting my print speed to substantially less than my motion system could handle. So I'm upgrading to the Lancer Long Melt Zone (45mm^3/s -> 60mm^3/s, or, at 0.4 nozzle and 0.2 layers, ~550mm/s -> 750mm/s) in hopes of being able to sustain high printing speed. Finally, I was told that if i wanted to make my input shaping better on the Y axis, I should make sure the printhead's center of mass is directly in front of the linear rail.

And here's where the real issues began. I checked out aluminum extrusion prices on AliExpress, Misumi, and 8020.net. The cheapest for my frame came out to around $125. This was not good. It was then that I came to a decision.

Well technically two decisions

But bear with me

Decision one: no compromises. Never. This is either the best printer or a mid printer. I've made a lot of choices up to this point with cost in mind, limiting myself, budgeting myself. No more. I'm not gonna like blow cash over this thing, but I'm not going to compromise quality for cost. More on the implications later.
Decision two: I'm gonna scrap the anicept vex for parts for the cel. While I don't want to make compromises, I don't have infinite money, and so disassembling the vex to help out would be a wise decision, I think. As I put it in my channel, is it sad? Yes. Will I regret it? No. The Vex only serves two purposes for me: to print 300°C and to have a slightly larger bed size. I don't need the larger bed, and I'm gonna use money saved from the vex parts to buy an all-metal (300°C-capable) hotend for my Kobra 2 Neo.

I'm gonna get quite a few parts from the Vex. The PSU, the Orbiter v2.0 extruder, and two MGN12Hx300mm extrusions are all going towards the cel. All in all, I should be getting close to $90 worth of parts from the Vex. (Well, minus $15 for the all-metal hotend I'd need for the kobra.)

This $90 extra will definitely help with parts. But if/when it still goes over $400: no compromises.

So I got to redesigning. First, I removed all of the electronics from the bottom of the printer, and organized some 4040 extrusions to make a super sturdy frame:

No 3d printed spacers, no weak points, all 4040 extrusions. Except for the x axis but it's supposed to be like that so shut up and leave me alone. I did a ton of research, trying to find the right electronics parts. I think I've landed on the BigTreeTech Octopus Pro, as it supports 48V onboard TMC5160 drivers, of which I need 3 now. One for each Y motor, and one for X. No compromises. I'll need to find a nice +320W 48V PSU, but that shouldn't be too hard.

Research was a rough journey. It freaked me out a bit seeing those super-expensive mainboards, $15 motor drivers, and ludicrous aluminum extrusion shipping costs. No compromises. Next, I worked on the printhead. I moved the linear rail closer to the extruder, making the center of mass much closer to it. This was actually fairly straightforward, surprisingly, as I thought it'd be some giant struggle to move everything away from the linear rail screw holes, etc. The most difficult part was probably moving the fan ducts down, as the lancer long melt zone hotend is 8.5mm longer than the medium melt zone. This also helped to move the center of mass downwards, I guess? Anyways, here's a comparison of before and after:

Before:

After:

I don't have any more images for you guys right now, but as you can see, the 5020 cooling fans are moved slightly away from the extruder, while the MGN12H is moved 20mm closer to it. The belt path will go underneath now. Unfortunately, I believe the printhead was the easy part. The X axis will need to be totally renovated now, because of the change in the belt path and extrusion location. But fortunately, as Blueprint has been extended until January, this extra work shouldn't be an issue.

Well, that seems to be it! If you read this entire journal entry, thank you but please get a life because that could not have been an easy read. If you're skipping to this last paragraph, shame on you. The Anicept Cel needs a lot more work, and it more than likely won't be easy. But there's one thing I'll be keeping in mind through the whole thing:

No compromises.

(Should that be my motto?)

Y Axis + electronics

Today (and yesterday, didn't have time to journal unforunately ;b) I worked pretty much solely on the Y axis. As you can see, I've done a pretty big rework, with a 4040 aluminum extrusion running down the middle and not one, but two MGN12H's on either side! This should afford a ton more stability, while keeping around the same cost, since the two MGN12's are about the same price as a single MGN15.

New bed frame:

Dual MGN12H's:

This little bit is gonna have a nozzle wiper!

Rough idea of how the front Y axis motor's gonna look:

And finally I mounted the buck converter that'll power the raspberry pi!

I asked for some feedback in the 3D printing discord, and got some unfavorable replies about motor-driver compatibility... I'm 80% certain it should work, but even still it's got me a little on edge. Fingers crossed.

New Z axis

Today I redesigned the Z axis. Totally. I've migrated from a belted Z axis to a lead screw in light of new complications with the Y axis. However, I'm very happy with the amount of work I did today, and other than mounting the Z motor, it's complete. All I have left (after mounting the Z motor) is the Y axis (shouldn't be too hard, 2 days?), electronics mounting (will be a bit harder, 3 or 4 days) and some more printhead work (mount ADXL345 and microswitches, route wires better, 2-3 days). I'm unfortunately not going as quickly as I would've hoped, as the middle of november draws near, but I'm making good headway. School should be loosening up as Christmas break is getting closer, so that's good i guess

New Z axis:

I'm sad I had to scrap all of the work on the Z belt tensioner and allat, but I think it's worth it. Also I took the X off of the x tensioner to make it cleaner I think

Color, X axis tensioner, Y axis rethinking

Color!!!!!!!!!!!!!!!!!!!!!!

First off, I'm rethinking the Y axis. As you can see from previous pictures, I'd had the idea to put an MGN15H on a 2020 extrusion. However, I'm not sure how well that's gonna work. Now with the advice of others, I'm thinking I should do dual MGN12H's, like some other high-speed printers do. Cost-wise, it's only a bit extra, but redesigning might be a pain. Here's an image:

The blue here would be a 4040 extrusion to mount the dual MGN12s on, unless I can get away with a 2040 pointing upwards. The green would be the two MGN12s side-mounted, which is where the problems start. They're directly interfering with my Z axis. (the red is the actual bed holder)

The way I see it, I have two options:

1. Use a smaller Z motor that affords more room for the MGN's
2. Move the Z axis down into where my electronics are, and do some finickery to reroute stuff

I'd prefer the former, but depending on how small a motor actually works, I may have to go for the latter. While waiting for advice, I did finish the X tensioner!

Good lookin stuff

CAD - more kinematics, re-research

First, I finished (probably?) the z-axis:

In hindsight, I forgot screw holes to connect the tensioner to the extrusion. no problem, will do soon
Then I did some research, cuz my buddy evan told me that the company I was wanting to get a motor sponsorship from (LDO motors) would probably not grant it. So I'm going to use StepperOnline 17hs19-2004s1's instead! I actually own four of these from a previous 3d printer, so I'm going to use those and replace them with cheaper ones in the first printer. I'm gonna do AWD on the Y-axis and upgrade from a 48V psu to 60V, possibly (according to evan) letting me go as high as 2000mm/s at 250k, which is incredible!

Here's the updated Y axis and PSU: (yes i repositioned the inlet again)

I obviously still have a lot of work, but i'm seeing progress and I like it.

More Z axis CAD

Today I finished attaching the belt to the Z axis, and just about finished making the z tensioner! I'm pretty glad that the z axis is almost done, because it's been pretty difficult and I thought i'd have to redesign like everything when I didn't think I could fit the belt attachy parts :sob:

CAD - X, Y, and Z axes

I'm quite proud of today's work! I've started connecting the Z axis belt to the gantry, and I've all but finished the gantry-to-frame connecty bits. Here's some pictures:

Electronics and Z axis CAD + research

today i was forced with a decision

not an easy decision

a hard decision (duh)

it was

quality or price?

turns out I didn't have to choose!!! For some reason a lot of my parts got cheaper so i have more budget. In short, I made the Z axis way chonkier by using 4040 extrusions instead of 2040 on the z axis, and I'm using MGN12H instead of MGN9H rails. I've positioned them inside the gantry now, cuz apparently it's better? Idk. I'm getting all this info from @evan le himself. Here's some pictures

oh yeah also i added a lil fan

Continued Z axis and electronics housing

I'm pretty proud of my progress on the Z axis today! I've gotten the belt system nicely concealed but still kinda steampunk-ish-ly poking out? Not sure how to describe it so here's a few pictures:

Still pretty unsure about assembling the z axis, particularly the rod and bearings to hold the gear ratio thingy. I'll figure it out. Here's the electronics:

It's compact and I think assembling it will be pretty straightforward. Cya tmrw

CAD - Z axis and electronics housing

btw im home from vacation :D

today i did a bunch of work on the Z-axis, trying to get the belt system working right. I've positioned the motor and pulleys in a spot that I really like, as the big 80T gear is kinda poking out and will be visible outside of the casing (which is cool imo), and the belt path is clear and fairly hidden.

CAD - repositioned parts, figured out frame

Today (and yesterday) I did a lot of repositioning parts and trying to figure out how the frame's gonna look. It's hard keeping in mind PSU wiring throughout all of it, but I think I have a good general idea. I'm using 4040 extrusions for stability, and nice big connecting pieces.

I must have repositioned that stupid power inlet ten thousand times, but I like where it is now. I've decided not to do dual z motors, simply because I didn't realize that I have to do a pretty big belt-gear ratio to keep the axis from falling, and those big gears are expensive.

Here you can see how I think the motors will be placed, but that may change. Another thing that I may change is the custom toolhead PCB. I like the idea, and I like USB-PD, but it'd just be much easier to use the wires I already have, save a few bucks with the PD cord, and just to get a regular gyro instead. I can still route the wires together using a wire cable sleeve.

(also im on vacation, so I have a lot less time to work on this :/ probably won't finish by november, but i guess that's alright.)

Guess what! More cad

Today I added the RPi zero 2 w to the design, repositioned some stuff, added 2040 aluminum extrusions instead of 2020, added the motor, the pulley, the bearing, and semi-redid the belt-attachy-to-printhead bit. I accidentally did all this originally using 20T pulleys, when I'd intended to use 16T, so I had to redo all that :sob:

Anyways here's some pics

It's been fairly slow lately... I really have to step it up to finish before November tariffs kick in. Wish me luck.

More research and CAD

The past few days I've been kinda slacking.... I worked a bit more on CAD and did a bunch of research of different unimportant things. I went down a whole 1.5 hr rabbit hole about 3D printer beds. Originally I'd planned to use the bed from the Bambu Lab A1 Mini, but propriety parts are SO EXPENSIVE AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

anyways, next i researched silicone bed heaters (that didn't turn out), 150mm beds (that didn't work either), and finally landed on a cheap PCB bed heater :D

but images are worth a thousand words so here's 2000 words

CFD #4 and began toolhead PCB

First I did the fourth CFD run:

Then I began to work on the toolhead PCB. I found out that USB-PD can carry up to 240W, which is great! Plenty for the hotend and everything.

Besides that, I did a little bit of CAD on the printhead, smoothing some stuff out and making it look nicer. I'm pretty pleased with how quickly I'm getting this thing done! I only hope that I get the PCBWay sponsorship...

Printhead CAD: Attached Belts, Finished(?) Ducts

Today I was fairly productive, and at this rate I think I can finish the printhead in two or so days! After receiving advice from @Evan Le, I've significantly widened the part cooling ducts, as 5020 fans have very little static pressure.

Additionally, I've added the belt attachy thingies, and I'm fairly confident in them. The belt wraps around a 3D printed part (reinforced by a screw) inside the printhead itself, curves in on itself, and meshes with itself as it comes back out to prevent it from slipping out.

I've also added an area that I could put the toolhead pcb, which I plan to put a gyro on.

Unfortunately I didn't have time to CFD test the new ducts... tomorrow!

More CFD Tests

Today I ran a few more CFD tests, with some CAD in between!

CFD Run #1

Here I tried to get the airflow to flow more directly towards the spot right beneath the nozzle, without limiting it too much.

Here you can see the ducts are directing the air too far upwards, so it's effectively cooling the nozzle and making the hotend work harder instead of cooling the actual print.

CFD Run #2

I've angled the duct outputs downwards slightly in the hopes that it'll deflect the airflow towards beneath the nozzle

Much better

I'm going to leave it at this for now, as I'm fairly satisfied with how it looks. Apart from that, I did more printhead CAD!

I love the logo and I love how compact and nice the whole thing works. I've switched back to a linear rail, at least for the X-axis, as it's cheaper than getting two rods and two LM8LUU bearings. For Y axis I'm still a bit undecided as to whether I should get dual MGN12H's, an MGW9 (a wider linear rod) or two linear rods. I may end up doing dual linear rods just cuz of budget, but we'll see.

I plan to put a toolhead PCB on the back, with a USB-C output like the Bambu A1 Mini has! Yippee

More printhead CAD and first CFD run

Today I cadded a bunch more on the printhead! I'm quite happy with my work, and I managed to get a ton of stuff done with connecting the fans and stuff. I made some quick fan ducts and ran a CFD test to see what they look like (spoiler: not good, but I was expecting that). Tomorrow I'll fix the ducts and run some more CFDs!

More Printhead CAD

Today I did a bunch of CAD! I locked in and was able to get a ton of stuff done. I arranged all my printhead parts in a way I think will work really well, and began to make the 3D printed parts. I've decided to put the ABL probe (CR Touch) in the front, which is fairly unusual, but the idea is that it lets me keep the hotend and extruder as close to the X-axis linear rods as possible to minimize the effects of linear bearing play and to distribute the weight better. In moving the CR Touch, I had to mount the heat creep fan at a 45 degree angle, but I think it looks cool so I like it. Finally, moving the probe gives me more room to put a possible toolhead PCB.

Last but not least, I huddled with Gabriel (@gbtsui) and gave him some tips and tricks on how to make a printer, cuz he's making his first CoreXY! Fun stuff.

Began CAD

Today I started to CAD the toolhead a bit. I did a bit more research, but I think I'm gonna get started on the printhead just to make sure I'm not forgetting anything or something like that. I'm using the ProtoExtruder 2.0, the Peopoly Magneto X Lancer Melt Zone standard hotend, dual 5020 part cooling fans, a 4010 heat creep fan, a CRTouch probe, and possibly linear rails, but I may have to switch to rods for budget problems.

Researched parts

For the past little while, I've been researching parts and exactly what I need to make a truly beginner-friendly yet super fast printer. My ideal with this printer is a bedslinger with Bambu Lab ease-of-use and CoreXY speeds. Of course, a DIY printer can't really get as easy to make as a Bambu Lab, due to the lack of custom parts, etc. But I'm sure going to try.

Essentially, what I'm aiming for is a printer I can sell. Again because I don't have access to bulk manufactured parts from a dedicated factory I'm not going to be able to sell it cheaper than say a Bambu Lab A1 Mini, but if I'm able to pull this off, people will be buying a fully open-source, super fast version of it.

Anyways, I've researched a bunch so far, and for the most part I'm proud of what I've found. As of right now, I'm dreadfully overbudget, but that might be because I put way too much into aluminum extrusions. I hope to get sponsored by LDO and PCBWay, and possibly Polymaker or BigTreeTech to cut down on out-of-pocket costs. I spent way too long trying to research a CPAP cooling system, when I didn't realize that they cost at least $70. That being out of my budget, I think I've decided to use dual 5020 fans instead. We'll see how it turns out!