Sheet Metal Brake and 3d Printer.

[S. Heslop]

I've been thinking about dual extrusion but yeah it just seems like a hassle. What i'd want it for is having ABS inserts on the surface of a larger PETG model, so I can glue to the ABS and hopefully reduce the shrinkage/ warping problems of an entire ABS structure. As far as I know the two materials don't adhere so it'd need to be sort of dovetailed in. But then I was thinking I might just be able to have inserts made separately and screwed/ snapped in. Or maybe just try insulating the hell out of the chamber (with water cooling for the steppers etc). Lots of ways to approach problems with this stuff! Water cooling might not be too bad to set up. Aquarium pump and hex brass drilled through with hose barbs machined on the ends, brazed to a larger plate that screws onto the rear of the steppers.


I'm starting to wonder if you could forgo the aluminium plate in between. It's used more as a heat spreader as far as I can tell. But if you painted the bottom of a big tempered glass plate black (with that oven curing glass paint) and set the nicrome wires with a bit of distance, then radiation/ convection would help spread the heat out... Maybe thread the wires around a mica sheet like on a toaster and stick some fiberglass roving underneath. And fiberglass rope to seal around the edges.



Can't find a single mica board big enough but can easily find it in half-size. I'd assume tempered glass won't explode since you'd only be heating it relatively slowly. Thing is how would you monitor the temperature of the plate...


Maybe the way you did it with tape and low voltages is the most sensible solution.

[S. Heslop]

Alright those plates arrived. I started to assemble stuff then figured I should probably paint the plates first.

The tolerance on the plates is real incredible. The bearing is a slight press fit into the hole. So paint is going to muck that up. I tried masking the holes with hot glue but it didn't really get into the corners. I still think the idea has legs though since most hot glues don't stick great to metals.


After my usual excessive amount of paint leading to excessive amounts of drips and streaks, it started to rain.


I like black and gold colour schemes, but this gold paint looks kinda crappy.

Still not dry after several hours so I bought them in so they hopefully won't get covered in condensation overnight. They actually don't look so bad in these photos, and really that's all that matters!


Also drilled all the holes in the extrusions.

Actually while writing this post I thought I might as well put the frame together. I was considering that id' need some big washers so the screws arent trying to 'open' the extrusions when pushing into the V groove. But I doubt i'd be putting that much torque on them.



When making stuff in CAD it's been my habit to get surprised by how big it is. But this time it seems a bit smaller than I imagined. Although it'll be larger with an enclosure.

Also it rocks on the desk if its sitting one way, but moved 90 degrees it doesn't rock at all. To me that suggests both my desk and the part aren't very flat. I'll need to investigate it, but part of why I like weldingrod's design so much is that it should be fairly forgiving of things not being entirely parallel. But I'm not so sure about what'd happen if the beams are being bowed.

[S. Heslop]

It's like Ikea furniture from hell...



Had to wait a day since the 2 hour dry hammerite spray paint turned out to be more like... well it's still not anywhere near fully cured but it's not so sticky now.

But without any adjustments except rotating the eccentric sleeves on the wheels it moves very freely.

I also made a coupler for the Z axis. I seem to be missing alot of drills from my set in the range of 4.5-5mm so with what I had the hole is fairly oversized, but I think some brass shim might sort that out. I need to make another but right now I also need to take a break. After that it's just the whole Y axis rod, the build plate, and then routing cables.

[WeldingRod]

Its thrilling to see another brainchild growing up!
I agree, it's hard to wait on paint!  I didn't paint any plates till version... I don't know... 6ish?

Sent from my SAMSUNG-SM-G891A using Tapatalk

[S. Heslop]

Its thrilling to see another brainchild growing up!
I agree, it's hard to wait on paint!  I didn't paint any plates till version... I don't know... 6ish?

Sent from my SAMSUNG-SM-G891A using Tapatalk

I was really in two minds about paint. I kinda like the look of bare metal, but i've had enough stuff rust now to consider paint.

This took me longer than it should've.

Got 2 of the 3 belts installed as well as the Y rod. Had to tilt the idler pulley plate up a bit to make the belt more level. Not sure what went wrong there but i'll need to think about it, but only if it causes problems.


Not my best machining work. I kept making mistakes. At one point a hot chip bounced right up my nostril. Quite unpleasant!


They're little jacks with a locking screw. Leather washers made from some welding gloves I bought that are so shoddily made internally that I cant get my hand inside. I figure a soft washer of some sort would be important to let the table tilt around the holes a bit.


The plate is from an old oven top. I figured the rolled edge might help support that long span better. I don't think it's entirely flat but i'm considering how to rest the glass on top. Blobs of silicone perhaps. But yeah it turns out the grid pattern on the plate is a quite precise 5mm spacing, which made laying out the holes a doddle!  All sheet stock should come with a grid printed on it. Just wish I took better care of the plate - did alot of welding and cutting on top of it.


Drove to the coast yesterday to get some filament. That's a lie, I went to get chips and because I sort of vaguely remembered I had a reason to go to the coast. Was hoping i'd remember on the way but I just couldn't till I got back. Aside from that I think I just need make a bracket fo the extruder then figure the wiring out and it should be operational. Who am I kidding something's bound to go terribly wrong!

[WeldingRod]

Looking awesome!

The Y belt pulley and idler are different dimensions; that's why it tilts.  You want the bottom side of the belt straight and the top side angled.  That way there isn't a tiny little angular error in your motion and you move back and forth.  I used belt tensioner springs on mine, BTW.  I think you can use the spring for a clothespin to do that.

[S. Heslop]

I'm not sure I understand.

This is how i've got it. With the top of the belt, where it attaches to the carriage, flat. So that the belts don't change angle as it moves.



Clothes peg springs are a smart idea though.

[S. Heslop]

Oh couple other questions. What's the plate that hangs off the bottom of the X carriage for? It's not installed in that photo as I don't have screws long enough. But aside from possibly stiffening the wheels a bit I'm not sure if it serves a purpose. Maybe for a depth probe?

Also what is that laser cut... shape in the Z plates on your machine. I figure it's a logo of some sort but I can't tell what it is.

[WeldingRod]

D'oh!  I forgot the attachments are on top!  Yeah, you want the top level!

The bottom plate is for the z zero prox.  I'll take a pic tonight.

Mine has been cranking on a print for two days straight, after the first 36 hour print finished...  full scale, here I come!

Sent from my SAMSUNG-SM-G891A using Tapatalk

[WeldingRod]

Oh, my first initial in script: R.

Sent from my SAMSUNG-SM-G891A using Tapatalk

[S. Heslop]

D'oh!  I forgot the attachments are on top!  Yeah, you want the top level!

The bottom plate is for the z zero prox.  I'll take a pic tonight.

Mine has been cranking on a print for two days straight, after the first 36 hour print finished...  full scale, here I come!

Sent from my SAMSUNG-SM-G891A using Tapatalk

I'm really dreading those multi-day prints. I was dreading them already but reading about electrical fires from all the strange high amperage Chinese boards made it worse!


Got two of them drag chains in. Of course they gotta come back out for me to have any hope of getting the cables through. I'll probably tie them to a thin steel rod and drop that through.



One to go. Just figured out where i'm gonna squeeze it. The machine is oriented as is for easier access to the interesting parts, and I figured i'd have it rotated 90 degrees in the end since that Y bar gets in the way. But I neglected to notice till now that the belts also get in the way. So as it is is probably the best orientation for easy access to the parts.

It's also interesting, to me at least, that the end pieces of this cheap drag chain is more what I anticipated - but the chain itself is made from a very tough glass filled plastic.


Finally, I need to consider how i'm going to house the electronics. I could slide a board into the upper groove on the base extrusions, and have a compartment isolated from a potential heated chamber in the future. But that won't give easy access to the electronics. And speaking of fires...

Maybe I could put a board in that groove, but also stand the whole thing on some 50mm tall stilts, then mount all the electronics to a tray that can drop and slide out.


Anyways either way, hopefully tomorrow i'll have it wired up and ready to blow some stepper drivers!

[WeldingRod]

Here's the bottom plate of the print carriage

Sent from my SAMSUNG-SM-G891A using Tapatalk

[WeldingRod]

Once your bot is running properly, overnight prints aren't a big deal.  You might want to add a filament run-out sensor; my last epic print took about 48 hours and spanned three rolls of filament.  If it runs out and the bed hasn't cooled, you can just stop it, pull the SD card, lower the head till it just touches the last extrusion, and delete everything from the gcode file before that layer height, then tell it to print.  That way you keep the zero point!

Cable chains are CRUCIAL to long term printing; glad to see you fitting them!  You have to be sure those dratted wires don't get caught!  Yes, straight chains are MUCH easier to thread.  Use fine stranded wire in them, BTW!

I run mine in the same orientation you are using.

I highly reccomend having a spring loaded build plate.  That way if you crash into it it can move away and maybe not booger up stuff too bad.  Ask me how I know...   https://madmodder.net/Smileys/default/confused0068.gif

My brain and power supply lives in the left corner of the base frame.  You can slip a piece of fiberglass in on top of the base cross bar and screw it down to create a cavity that is partially isolated from the hot cavity.

Mine has feet screwed into the bottom of the Z rails that raise it a bit.

I think I've posted pictures of the Z before, but here's a fresh one.  I had the X and Z cable chains as one continous chain, but connected backwards at the X/Z junction so they bent the other way.  The flip happens at the one with the "X" on the side.  The end of the Z chain is a leftover plate that I bent to shape.  I might have had a few laying around ;-)

[S. Heslop]

I was considering a spring loaded build plate just to make adjustments easier, since I wouldn't need to trap the plate between soft washers. I decided it wasn't a good idea since I might need some hefty springs if I decide I want to produce a solid block to the machine's maximum volume, but I didn't consider the idea of the head crashing.

Got the Z axis chain installed now but not much else. I thought i'd do a quick temporary repair of the steel cold water storage tank by sticking a glob of epoxy putty on it. I figured it might be low pressure enough that i'd get away with it, but I ended up spending all day trying to drain and dry the system out enough to get some on. Might still not work but this tank needs replacing anyways. I'd be happy if it lasts a week.

[S. Heslop]

Was dreading this since I kinda knew it'd be the fiddliest part of this. What I neglected to consider though is that loaded with wires those chains weren't going to flex as easily. Especially since I got shielded cable! Had to fiddle with the Y axis chain to drop it down a bit, but now I need to add a couple more segments to get the full Y travel. Also lost about 20mm of Z height.

Still gotta extend the Z axis stepper motor cables and add a few end connectors. Should hopefully have it running tomorrow. I think I said that a couple days ago...

[WeldingRod]

Useful trick on tanks: hook a vacuum cleaner to the top port and pull some vacuum!  I've pulled a 1/2" NPT fitting out of a full hydraulic tank that way.  It was under about 2 feet of oil!  It went "glug glug" when I pulled it, hardly anything came out.

Sent from my SAMSUNG-SM-G891A using Tapatalk

[S. Heslop]

Got it sort of wired but the X motor is... trembling. It moves but in a very loud and stuttery manner. Wonder what the cause of that could be. I guess I aught to check continuity on the cables.

[S. Heslop]

Continuity seems okay. I switched out the stepper driver and that didn't do anything. My best guess is that either the stepper motor itself is damaged or the approximately 3 meter long cable is too much for it. I'll try exchanging the motor for the extruder motor I suppose.

Edit: Motor is fine, tried hooking the Y motor to the X output and that runs fine. The problem must be with the cable. I guess I could try measuring resistances but I hope the issue isn't just the length.

[awemawson]

3 meters shouldn't be too much Simon, I have well over that on some runs on my CNC Plasma Table. Are the motors 6 wire that you have to interlink, and maybe the linking has gone astray?

[S. Heslop]

I wasn't sure what you meant but I did switch two of the pins inside the plug around after the first issues. I... didn't consider to check the other leads but it turns out the way it was with the middle two wires flipped was correct. I'm not sure what the first problem was. Maybe something stupid like I had the connector set over a pin.

Everything seems to work now but the Z axis. Sharing the one controller it doesn't have the torque to lift the Z axis. Is there any way to configure marlin to make use of that spare 2nd extruder controller?

Edit: Forget that last one, just found the right combo of things to google search. Also thanks for the help Andrew.

[S. Heslop]

Oops. Probing about trying to adjust the currents on the steppers. Went and Touched the wrong thing and now the display shows all the pixels on, except briefly returning to normal as it's powering off. Seems it's permanently broken but I wonder what part...

Edit: I can just make out the display at an extreme angle and turn the heater on, but none of the steppers move. I heard these things were fragile but yowch. Glad I didn't buy the stupid £160 board!

Edit again: Gonna go with the MKS Sbase 1.3 for about £50. I can feel the sunk cost fallacy starting to sink in! But it's more that setting the currents with a tiny potentiometer and calculating the whatever with the vref isn't so appealing, especially after finding out the Legends are true and those ramps boards are very fragile.

Trouble is the Sbase uses JST connectors instead of the dupont connectors on the ramps board. So I might need to crimp on new connectors or make adapters. But i've been assured by Some Guy that the dupont connectors fit anyways, but i'm not entirely convinced. Seems you wouldn't get full contact.

[WeldingRod]

I wired my z steppers in series and was happy with the result!  I think that's in the manual....

Sent from my SAMSUNG-SM-G891A using Tapatalk

[S. Heslop]

I wired my z steppers in series and was happy with the result!  I think that's in the manual....

Sent from my SAMSUNG-SM-G891A using Tapatalk

Honestly it's more the cable i'm worried about than the driver.

Alright after alot of stupid crap I finally got it working.


And alot of fiddling with the print settings. Started out with the head deciding to move up about 5mm before starting each print, but changing the initialization G code a bit stopped that. I was then massively over-extruding, then under extruding. And now i'm sort of in the ballpark with it.




This is where i'm at now.


(usual disclaimers. looks alot worse in the photos. still not ideal)

20x20x10mm cube. I've yet to measure it. I'll grab the calipers next time I go downstairs. I think the slight wobbling is from not having a cooling fan, and the fill layers pulling the walls in a bit. Maybe getting the extrusion setting just right would further reduce that too.

I was very fortunate to have just recently met a guy online who's into this stuff. He was able to guide me through alot of the problems I was having, and I wouldn't have figured most of them out on my own. A big one was needing to install the smoothieware drivers and then, through the Windows device manager, install the drivers again (by looking for them on the list of installed drivers) on the MKS SBASE board. So it'd show up as a COM port.

I'm running Marlin 2.0 on the board. It's still in development which made it a real chore to look information up, since fixed problems got broken again, and broken stuff gets fixed on a weekly basis. So most searches show up outdated information and fixes that no longer apply. Required installing a bunch of IDEs and editors to compile the firmware. Open source guys are weird.

Also they weren't kidding about PLA smelling great. It's like golden syrup.

Edit: Dimensions are about 19.9x19.9x10.1. That's better than I expected without any adjustments, but who knows how it holds over a larger piece. I'll do some more tests tomorrow, like the classic retraction pillars and just a big flat square or grid. I also need to measure skew on parts.

[WeldingRod]

 
Congratulations!  I'm totally amazed that you got a good print with that few ugly ones!

[chipenter]

I agree about Marlin settings I am running Marlin 16-12 for my mill , and some of the setting don't make sense like turning off the end stops except for homing !