3D printer ADDon

I’m looking to convert my journeyman to a large format printer(with a super volcano).

I was hoping to get some advice on how to do it affordable and easily.

I was planning to harvest parts from my CR10 v3(currently has the volcano on it) and 4 monoprice maker select plus(if necessary).

I’ve also looked into a cr10 500 extender kit(buy 2-3 kits plus extra parts) to make a 300x1500x500 setup. Seams pricey, but far more straightforward.

this is a video of someone doing a 1000mm mod, but im warry that 1500mm start introducing unforeseen problems

Good luck.
It’ll be interesting to see how it turns out.

Are you doing it because you need to or because you can.?

1 Like

You have enough Z height for your projects?

1 Like

I need it for a project, I’m printing a ton of sub-assemblies on my cr10 and don’t want to have to reset the printer 12-18 times a day. I’m trying to print directly onto a large cnc’Ed acrylic panel(cut with the journeyman) and make the print a single piece instead of 18.

Trying to save time glueing and make a better end product

Yeah that’ll definitely be an issue, thankfully my current demands would be fine if I just move the z axis up to the max mounting position

I’m actively working on 3d printing with my Woodworker. On the scale of the price of the Onefinity itself, I don’t think “affordable” is as much of a challenge as “easy”.

The Buildbotics controller can push 6A on a single channel, so I technically could parallel the two 2.8A Y-steppers to free up the 4th driver for an extruder, but I figure there’s probably a reason Onefinity didn’t do that to start with and I’d still have to figure out some external controls for the hot-end anyway. You can get stand-alone hot-end control hardware with dials to set the temperature, but that doesn’t work in my “everything is software” workflow.

Instead of fighting with and likely frying my Buildbotics controller, I opted to pick up a BTT Octopus board which can drive 8-axes, 6-pwm outputs (fans, spindles, etc), 4-heaters, 2-always-on loads (fans, lights, etc), and a partridge in a pear tree. I’m running 2209 drivers which peak at 2.8A, same as the motors, and which support sensorless homing. For extruder / hotend, I’ll be running an E3D Hemera. I figure if the Onefinity can handle lugging a router / spindle around, the “it’s so heavy” argument doesn’t hold water in this application and it has a great reputation for reliability and print quality.

As of today, I have the Octopus board configured with Klipper and wired up to my Woodworker’s X and Z and STEPPER_BUZZ works on both. This evening I’m working on dialing in sensorless homing on the 2209s, hoping to have safe and usable X, Y, and Z jogging before I call it a night.

I’m documenting stuff as I go and am posting project videos on YouTube with the #octofinity hash tag. I just published the first video this afternoon.

1 Like

I finished my build with a duet 3 board, I was going to share results but my work is sensitive so I’ve opted against it. If for whatever reason you go with a duet 3 build send me a PM and I can help guide you if you get stuck along the way.

Made some headway the last couple nights – sensoreless homing is working with the 2209s and Klipper and the machine is jogging around in Mainsail as it should. I gotta say I do miss my gamepad though. I may have to write a Klipper extension to get that back.

I’m starting to play with adapter ideas to attach the extruder/hotend to the spindle mount. I was originally planning on just hanging it off the front, but once I got my first draft mounting plate on the machine, I like the idea of tucking everything nice and neat behind the mounting plate where the spindle would normally go. That would of course rule out doing prints w/o removing the spindle, but I don’t think I want to be carting around the weight of the spindle while moving as fast as possible for 3d printing anyway.

I shot a quick video tonight of where things are at. I’ll put a repo together here in the next few days to share the evolution of my Klipper printer.cfg. This will include all the sensorless homing, acceleration, max velocity, etc. values that folks occasionally ask about.

About 80% of the way there now. PoC is working, but I need to redo the cabling so it isn’t super janky and a fire hazard and I need to sort out smoothly controlling the temperature of 12-sqft of glass. I’m thinking probably cutting up two or three big heating pads, sticking a few thermistors to the back of the glass, and seeing how that goes. Most consumer heating pads go up to 80C with their stock controllers, which is well more than I need.

  • Klipper printer.cfg as promised in my last update.
  • Hemera mount for the front of the spindle mount. Super simple and a little flimsy right now, but functional. I’m sure I’ll be making this more robust once I start laying down plastic.
  • Connector tables from the video. This info will eventually lead to high-quality cabling of the proper length with the proper connectors on both ends.

Progress Video:

5 Likes

Got things dialed in enough for a solid first layer tonight – letting it run now to see if it successfully completes the print.

Addition Jul 7 as I’m not allowed to reply to myself more than 3-times:

The print from the other day failed partway through the second layer due to poor adhesion of the first layer. I don’t have the bed heated yet and hairspray isn’t cutting it on its own. I’m playing around a bit with purpose-made 3d-printing bed adhesives to bridge the gap.

I’ve started upgrading cabling yesterday and successfully printed a Mainsail logo standing up, overhangs and all. It’s messy still, I think mostly due to lack of rigidity in the extruder mount. I’m likely going to make the mount much larger and cylindrical such that it fits in the spindle holder as opposed to bolting on up front.

I’m playing around a bit with the bed mesh as well. I ran a mesh calibration on the Kilpper console and it helped a bit, but I’m still seeing signs of the nozzle being both too close and too far away in different sections of large first layers. Mainsail has its own mesh configuration UI which I haven’t played with at all yet. I’ll probably pull the mesh out of Klipper’s config and see what Mainsail can do. If it does work better, I’m curious if I’ll still be able to use that mesh if I send the machine instructions via pygdk instead of the traditional slice, upload, print workflow.

1 Like

Hi Cilynx,

So I’ll reply, so that you can reply back.

This project is an amazing idea, and starts with a lot of good choices regarding board and extruder from my point of view. Although a 1f is probably not the perfect 3D printer, it will help us to understand how solid mechanics influence print quality on a huge build area (and speed? how fast did you print? maybe so fat that a mosquito magnum would be worth trying?)

Since I received my woodworker 3 weeks ago, I constantly think about this topic and am really curious how your story continue with respect to bed heading, adhesion on the huge surface (pei?) and enclosure :wink:

I am looking forward to reading more on this!

Substantial progress this weekend.

I swapped out my 30ft heating coil for 100ft and I can now hold 40C on the bed with 1100W startup power and 375W holding.

I finally made a reasonable spindle mount for the Hemera and the added rigidity has promoted this project from a fun toy experiment to an actual usable machine.

The combination of extra heat and being able to put the nozzle reliably where I want it, adhesion is no longer an issue. If I use one of those glorified 3d-printing glue sticks, it’s actually difficult to remove the print. Aquanet works amazingly well on my Kossel, but I don’t want the overspray getting all over the Onefinity’s rails and screws, so I’m sticking (har) with application mechanisms I have more control over.

As for speed, the Hemera is currently my limiting factor. I have my top speed capped at 5000 mm/min and you can hear in the video below that as it gets close to top speed the Hemera chirps as it starts dropping steps. I’m playing around with different run currents to see how much I can get out of it before I burn up the stepper.

1 Like

The machine is working pretty well now as a “real 3d printer”. I’ve gone back to hairspray for adhesion as it works better than anything else, so my next project is designing and printing up some rail protection to hedge the overspray.

4 Likes

Don’t know why I never saw this post before now, but very impressive! I converted an Ender 5 Plus to all linear rails and a Micro-Swiss hot end, all of which are still working great 3 years in. And I like your choice on the Hemera hotend which is on another of my printers.

Apologies for not seeing if you posted about it earlier, but I’m not seeing any bed leveling, is it even an issue for you?

1 Like

Isn’t the bed leveling accomplished here by surfacing the solving first?

Well, you leveled your CNC bed with the router bit, but it’s important to remember that the flatness of the bed is now relative to the CNC router, not the 3D printer nozzle.

3D printers are generally much more sensitive to variations in bed height compared to CNC routers, and this can lead to the infamous “first layer” issue, where adhesion problems and inaccuracies in the final print can occur. For example, my Ender 5 Plus has a printing area of around 350mm square, and the bed’s level/pitch and surface height can vary significantly from one end to the other. I’m using a 0.4mm nozzle and printing layer lines at 0.2mm in height, so even a variance of 0.05mm can cause issues like poor adhesion, warping, or uneven prints. It’s worth noting that even materials like glass can have variations in surface height in the range of tens of microns.

To combat these issues, I manually adjust each of the 4 corners of the bed with the nozzle parked at each corner, and before every print, I run a 4x3 bed leveling grid pattern using a BLTouch probe to measure the elevation at each point. The machine will automatically use this height map during printing to compensate. There are other types of probes like inductive sensors that offer even higher precision but come with their own trade-offs.

In your case, given the giant scale you’re working with, and if you are using something like a Super Volcano, it has the ability to extrude a significant amount of filament quickly, and might compensate for an uneven surface to some extent. Your first layer might not look perfect, but it should adhere well and minimize warping. I personally run my bed at 75°C, which is quite hot (borderline melting) and helps with adhesion. I also use a PEX surface plate from WhamBam systems, and apply a thin layer of Elmer’s glue stick (the purple one helps me see where I apply it). This combination has proven to be quite reliable. What’s great is that the glue stick can be easily cleaned up with water, whereas hairspray can be messy and may require IPA or Acetone for cleanup.

I watched your videos and didn’t notice any obvious layering issues, but if you want to experiment further, you could try increasing the nozzle size to 0.6mm or even 1.0mm, raise the temperatures and printing speed, and possibly increase the layer height to 0.4mm - 0.6mm to see if this helps with better adhesion and smoother printing.

I look forward to seeing more prints!

1 Like

Leveling has been a fun adventure for sure. Right now the slab of glass is just laying on top of the coil of pipe heating wire, so not exactly a precision setup. The bed also changes shape considerably between room temperature and 40C (as high as I can get it reliably right now). Happily, its shape is fairly consistent at any given temperature, so I have a 6x6 mesh stored in Klipper and it mostly just works. The ugliest part is that the difference between the highest spot and lowest spot is almost 2mm and I can easily flex the glass 2mm in most locations just by pressing on it a bit with my hand.

I’m currently debating between getting 4x 400mm heaters or one giant 800mm heater. Once I sort that out, the next step will be more appropriate support under the bed so I can get it really level. Getting that whole mechanism self contained so I can easily remove and replace the bed depending on if I want to print or carve is going to be a fun project as well.

3 Likes