LGX Lite fully assembled is $70
Sherpa Mini kit is one open source deals where your going to have all kinds of pricing. LDO is a good brand with a kit at $60. Probably less if you print the body which why wouldn’t you with 8 printers. Put a box over one and print with ASA.
A CHT nozzle would cost about $20 if you need to increase your flow rate. I improved mine by switching by about 8 cubic mm per second which is dang good.
Here’s a speed test with a LGX Lite on my V0.1.
I really appreciate all the info. I am seriously considering it. I like filastruder, I never saw that kit there…hmmmm.
No problem. I don’t know if there’s any advantage to getting the SLS printed parts since they are FDM designs but one less thing you have to worry about.
What are the non bed slingers that were in your picture? Your design? Let me know. Thanks!
Yup second corexy design for me. I actually just revised all the parts a tiny bit for slightly better looks and fit on a few parts. MP3DP v4 - V1 Engineering Documentation I have been taking some notes so I can whip up a BOM to add to the CAD, and maybe a few instructions.
I should have less moving mass than the one you showed but at speeds that high and those accelerations the extra rigidity yours has with the dual gantry rails might have the advantage.
I’m running mine at about 12^3mm/s flow rate, cooling really starts to be an issue at that point. So minimum layer times factor in. I have an idea for a better print fan angle but I have not had time to hash it all the way out yet.
And the extra motors on X and Y! Yeah it’s stupid…
I really like your design. Using the lone linear rail for the X gantry instead of strapping it to an extrusion removes a ton of mass I bet. Got a friend looking to get into the hobby. Perfect op to build one!
That chonky Hemera would limit your speeds and accels huh? Maybe the lightweight extruder could free up weight and space for dual 5015s? Or even 4010s?
Ryan, if you got rid of some of those, and added water cooling, you could go a lot bigger:
Oh wow. That is awesome.
First of all, this is an extremely impressive machine. Well done! I’d love to build this thing…
Second, whenever I see machines that can handle these speeds, the examples typically show them printing larger parts with a bunch of rounded corners and geometries. How does your machine perform on smaller parts with finer details or squared edges? Do you need to slow it down a bit in your slicer settings to account for rapid direction changes at the corners? Do you get good bed adhesion at the corners when moving that fast?
I’d be interested to see a Benchy print on your machine. I know a Benchy is pretty rounded in most cases but it does have some smaller and more drastic movements towards the top (boat cabin and exhaust stack).
The video linked does show a bed full of parts. I am sure printing the first layer slow for adhesion is good and I think as long as you can cool the previous layer just enough not to curl fast printing probably make very nice parts as the whole bed of parts has less time to cool between layers.
I have definitely worried about that, though. I print at 55mm/s with a larger nozzle and that is so much heat I have to set a high minimum layer time to allow for things to harden up enough to print on.
So I think the simple answer is single prints will look just like a regular printer, they have to print slow to allow for cooling.
You’d be surprised. Once I get back in town I’ll print an example.
I’m not a mechanical engineer but with the crossXY gantry and the super rigid and constrained carriage it can handle corners very well. The hotend has a thin metal plate attached right above the heater block that is tied into the carriage which keeps it from shifting on fast corners. Square corner velocity (jerk) is 15mm/s right now but that is the minimum for the machine. Once I’ve got it tuned for PA and IS it should be able take corners faster.
Yup. Cooling is definitely the bottleneck on the machine. They have a new cooler in beta called the Frostbite they are working that is supposed to remedy that. Not on my machine though.
I’d say it is not a PLA friendly machine since it needs so much cooling. ABS, ASA, PC, PA are the types of material that this is designed for.
Where did you here about that? I am not finding much other than the store and their reddit, which does not have much info.
They have a Git for all their projects. Community is on discord.
The only reason I knew about them was because they came up with the “speed Benchy” challenge that was a popular thing on Youtube for a while.
And CHT knockoffs are now available on Ali for a couple bucks as well. If you run input shaper etc then honestly weight on toolhead is not a big issue unless its extreme.
The K3 is one heck of a machine.
Annex panel clups are amazing as well i have them in my Vorons and love them
The clips are very nice now that you mention it. There’s lots of things they designed on this that you don’t realize how clever they are until you install them.
My favorite general purpose part they designed so far is the 3 part raceways used for cable management in the rear box.
Everything goes together with a satisfying “click” !
Someone order a boat?
ABS
0.4 nozzle
0.2 mm layer height
3 perimeters, 3 top solid, 3 bottom solid
5 sec min layer time
As @vicious1 pointed out, min layer times tanked the overall speed on this single, smaller model. Had to have some time for the ABS to cool or it’d curl right up.
Also had to knock the print speed down to 250mm/s. Was getting small, random layer shifts. Could be my motor current…I hope.
Over 25min of print time but we can only get faster from here. @MattMed it’s hard to tell but the tight edges are pretty crisp. Just the ghosting afterwards is pretty heavy. Most if not all the print defects seem to be related to resonance and the inconsistent speed.
Wow, it really is an impressive machine. Everything from the print speed to the end of print nozzle cleaning process. Really nice. There’s not a whole lot you can do about min layer times, the plastic has to cool. But I am sure you can knock another couple minutes off playing around with it.
Interesting finding.
After getting the machine running better (higher X/Y current, deracking gantry) I ran some test parts to figure out the ringing problem. The left part is the print at 14k acceleration about where I was printing at before. Really ringy. Middle part was at 22k. Ringing is substantially reduced? Right is 22k with input shaper calibrated. Ringing almost completely eliminated. All parts printed with the same gcode at 250mm/s.
I would have never thought increasing acceleration would reduce ringing. Must be the 14-16k accel range produced some extra vibrations?
Resonance is super fascinating, and difficult to plan for!
Amazing project you’ve got there👌
Thanks! Yeah it’s very interesting. Challenged my assumption that slower speeds equal better prints. Not quite that simple it seems.
