Welcome to the SeeMeCNC Community

I have been using something like this in some projects in the far past. I was thinking this would be an ideal mounting method for panels of a heatchamber.

https://www.amazon.co.uk/Magnetic-Craft ... B001DT262C
Steel counterpart tape is also available.

It would make dismantling the panels easy when you do some dismounting of HE or Extruder for service purposes. Also it would prevent things from breaking if you accidentally run fans into the wals I am lacking the design skills however to do something with it. It would need a ridge of about 1.5mm on the mount side.

DeltaCon wrote:I have been using something like this in some projects in the far past. I was thinking this would be an ideal mounting method for panels of a heatchamber.

https://www.amazon.co.uk/Magnetic-Craft ... B001DT262C
Steel counterpart tape is also available.

It would make dismantling the panels easy when you do some dismounting of HE or Extruder for service purposes. Also it would prevent things from breaking if you accidentally run fans into the wals I am lacking the design skills however to do something with it. It would need a ridge of about 1.5mm on the mount side.

Not a bad idea. It's not as strong as the button sized neodymium magnets I was intending, but more can be easily used and I suppose you won't topple your printer when tugging on a panel.

I've also considered use of the foam/presentation board you can get at various shops. It's opaque, but much less expensive than the acrylic and you would really only need the one clear panel. I've used it for a temporary chamber around an old cartesian printer before and it worked well. Not super durable, but easily replaced.

ccavanaugh wrote:Not a bad idea. It's not as strong as the button sized neodymium magnets I was intending, but more can be easily used and I suppose you won't topple your printer when tugging on a panel.

It is surprisingly sturdy, because the contactsurface is much bigger if you put the tape around the whole window, or even just across the top or the bottom. Because the basis is a foam tape you could make (print) vertical joints 120 degrees that are also easily removable. The foam board you mentioned is also a good idea. It is used a lot in signage industry, and can be easily bended with a glow spiral (much like acrylic).

bvandiepenbos wrote: I also added some brackets for Onyx HPB
http://www.tricklaser.com/MAX-METAL-Bra ... PBBRKT.htm

This is the thing to get!!! Very easy to install - notched to fit exactly onto the frame - and the use of the screw-heads to push them down into place is very clever. I also tried the plywood version, but didn't like that the frame had to be disassembled to install it. I do have a PSU I need to mount, so I designed a 3D-printed rail mount for it. I will upload that after I have a chance to print it and make sure it fits without running into the snowflake. I intend to use Taulman BluPrint, which is their most heat-resistant alloy (can withstand immersion in boiling water for 5 hours).

You can also use these to mount endstops, if you don't want to use printed parts for that.

DarrenAnton wrote:Anyone have a good Fan cooler for Max Metal and E3D Hot end? what are we all using?

This is a fan mount I made that fits between gaps on the platform. It works with 2 fans. Feel free to modify it to your needs.

http://www.thingiverse.com/thing:1660419

Hi 626Pilot, been reading through all these threads and found these pics from last year. Any chance the files are available for download, really eager to put my 360mm magnetic arms on my v2 with the trick laser trucks. These look really good. Thank You for all you do here

[img]https://photos.smugmug.com/Other/3D-Pri ... ctor-S.jpg[/img]

[img]https://photos.smugmug.com/Other/3D-Pri ... iage-S.jpg[/img]

I have been working for some months to refine those designs. They look slightly different now, especially the carriages. They have a different/better belt clamping mechanism that's designed to fit the tooth contour of the GT2 belts.

The trouble is that printing them on a 3D printer that doesn't have utterly perfect positional accuracy gets you an effector and carriages that are slightly worse than the ones that are already on the printer. It seems you can't make these parts more accurately than what is already installed - and if your printer is already perfect, then you don't need these in the first place.

I tried using a resin printer to do these parts. The carriages came out very well. However, resin printers have an Achilles' heel: the parts have to be positioned at odd angles for drainage and runoff, and to reduce overhangs; PLUS you have to put supports on a lot of places to help the part survive the peeling process between layers, PLUS there are surfaces that have to be flawless and thus can't have any support structures attached to them. Trying to balance those three requirements made me feel like that guy who has to push boulders around in Hell. Because of specifics of the effector's geometry, it's not possible to get a good print, even on a printer as good as the Form 2. The captive locknut channels all come out different, even though they are identical in the model.

Next step is to get these parts machined from aluminum. I'm working on that now. If they work well enough, I will sell them.

If you don't want to wait for me to figure all that out, you can buy an effector/carriages from Haydn, the guy who sells the arms. They are 3D printed, so they aren't perfect. Since my goal is to have the absolute maximum dimensional accuracy, I decided not to go that route.

why not sls nylon?

Qdeathstar wrote:why not sls nylon?

I spoke to Shapeways about this. I asked them what dimensional accuracy I could expect, and they said 5%. That's okay for printing statuettes, but obviously no good for a precision part. As the part cools, it wants to contract, so you can have places that will shrink a little. The other issue is that with any laser sintering process, you have to deal with powder, the grains of which are randomly distributed. This means that there is unavoidable surface roughness, and that certainly does't help.

Maybe I can get away with a 5% dimensional accuracy variance (extremely doubtful) and maybe the surface roughness isn't an issue, but from an engineering perspective, those are two major pains in the ass. The fewer unnecessary variables in the system, the better. I want something that isn't going to be fiddly, that just does what it's supposed to as close to perfectly as possible. Milling aluminum eliminates both of those problems, and I can tell the manufacturer which surfaces are critical, so they can have the machine spend more time on them.

I'm about halfway through the process of porting the design over to Fusion 360. I tried converting the STLs to STEPs, but it's no good. STLs are pure polygons, but the machinist has to have parametric instructions. A pile of triangles that happen to form something like a 3mm hole won't work, because the machine needs parametric instructions like "drill a 2.8mm hole here, then tap for M3 in the same place". It's crucial that the tapping be done by the CNC machine so that there isn't even the tiniest deflection in the ball joints.

Thanks for the info.

I doubt the surface roughness would be an issue since its mostly uniform, however, the 5 percent variance likely would be an issue :-/

Did you see DC42's PCB carriages and effector? Looks pretty interesting. Basically, it allows for very percise location of the arms but without the need for machining. My only doubt is on the stiffness of the effector...

I haven't seen them. Do you have a link?

Those look pretty neat, and I admire the idea of using PCBs. The PCB effector only supports center-feed, 1-in 1-out hot ends, but I want to be able to support Cyclops and Diamond hot ends, so that won't work for me.

His parts may be lighter than mine, but mine will be made from aircraft aluminum, so they won't be "heavy" either. Mine will also use the same center bore size (the huge hole in the middle) and hole pattern as the SeeMeCNC effectors, and have the same thickness, so it will be compatible with existing accessories, and of course you can have multi-in (and multi-out) hot ends. Holes will be threaded, so you don't have to use locknuts. You also won't have to hold the locknut with pliers in one hand and tighten or loosen the screws with the other, which is a little haphazard.

I may be missing it, but I don't see provisions for part cooling fans on that effector - however, if you built something that shared the screw holes with the hot end, you could do it. His PCB can be designed with harness connectors and circuitry (like a HE280), which is nice. To get the same effect with mine, you would need to get a Yellowjacket PCB mount (which I think still has a small center hole, so running multi-in hot ends might not work).

Looks like the 5-axis idea is out of the question. The site I was using to get a quote has an idiosyncrasy that can make you think you're getting the price for machined 6061 aluminum, when you're actually getting the price for sintered nylon. 5-axis machining is too expensive for this, for the time being. I'll look into other options later this year.

The old design uses captive nylon locknuts to hold the joints in place. I think I mentioned above that this makes it very difficult to print properly on a resin (SLA) printer. Since I had to redesign the effector in Fusion 360, and since Fusion 360 has the ability to cut accurate threads directly into the model, I decided to drop the locknut idea and just thread the holes the ball joints screw into.
I finished printing a prototype on the Form 2 earlier today. The threads printed a little tight, so I had to use an M3 tap to clean out the first few millimeters, but the ball joint screws went in just fine. The tightness of the threading gives the screws something to bite into, just like the locknuts did before. Having the whole length of the holes threaded also makes it easier to calibrate all the ball joints to be at the same elevation.

This is the result. The threads printed beautifully, and I don't have to use locknuts to secure the probe mount anymore.
I mounted the new calibrated effector to my printer, and ran my heuristic calibration routine. Average Z error is 34 microns, or 0.0013 inches! This is the best calibration I've ever seen on this printer.
I'm still using the old carriages, which have nylon locknuts. I might redo them in Fusion 360, giving them the same printed-in threads and ditching the locknuts.

If I can get these good enough, and if they make it through beta testing, I'll sell them.

Looks very nice. Just out of curiousity, how much did they want for the machining? Could you possibility just use a standard cnc and then finish the mounts for the magballs on a mill? I'm a complete novice on this sort of thing, but i've been watching a lot of youtube videos from this guy named AVE and he does a lot of milling so it's got me interested. I want to own a mill one day, even though honestly i have no practical need for it ^_^

They wanted over $400 to do a very basic version with no trim, and over $600 to do the "ornate" version you see here. That's just for bare aluminum. It was still more to get it anodized. Unless I want to sell printer parts exclusively to rich people, it just wouldn't make sense. I still want to do an aluminum version someday, if I can find a way to do it that doesn't cost so much.

Drilling the 45-degree joint block holes would be impossible on a 3-axis without some kind of really, really accurate fixturing, and considering that even my enormous 3-axis has only 5" of Z (some of which is taken up by the length of the router bit), it doesn't seem likely to work. Figuring out (0,0) on a CNC table router to within a few microns is... well, the phrase "good luck" comes to mind. I got pretty close for a big project I had to do, but I don't think what I developed is close enough for this application. The dimensional tolerances have to be as close to perfect as possible. My CNC couldn't thread the holes, either.

Since I already have the Form 2, and since it just produced a part that's given me the best calibration I've ever seen on this printer, it makes more sense to go this route.

Q, if you've got a CNC mill to do the basic work, it's far easier to do it all in the CNC. You'd just need ~3 sets of vices. First would be the roughing, which would drill locating details and threads to retain it to a jig plate for the second operation, which would cut the outer perimeter, and rough the top section, then 3 repeats of finishing, then tapping the ball mounts. You could even get pretty close on a bridgeport with a good set of glass scales, though a probe equipped mill is far better.

I maintain I could do it with 2 fixtures, 3 kurt vices, and 5 "operations" where the part is moved by the operator afterwards, so long as I could skip the runic engraving (Or do it at the angle of the ball mounts). It would probably even be something I could write by hand and design fixtures for. So if you're interested in sharing the design at this point pilot, I could design a set of setups to do it on a standard 3 axis machine.

Xenocrates wrote:So if you're interested in sharing the design at this point pilot, I could design a set of setups to do it on a standard 3 axis machine.

The machine only has 5" of Z, and the part would have to be mounted at a 45-degree angle, so it could be tight. Also, my driver box isn't talking to the computer and I haven't got around to sending it in, so I can't mill anything right now. Nevertheless, I'll keep your offer in mind. Thanks!

Wouldn't making a casting of that and then doing some finish work on it be more efficient/reliable in the long run?

g.

That thought had crossed my mind. IM is an option for a mature product (doesn't need any design changes) with enough market to justify producing 100s or 1000s at a time. Getting the molds cut for something like that would cost somewhere in the five-figure range, if I'm not mistaken, plus whatever the IM shop charges for their time + materials + shipping. With SLA, or Xenocrates' CNC method (assuming my Probotix has enough Z to accommodate drilling the joint blocks), my cost to get set up is far lower, and I can make some design changes without the expense of re-cutting molds.

WIth SLA, I can reliably produce one set (effector, 3 carriages, plus cable clamping hardware for each carriage) per day @ 25 microns. There's also about an hour of labor to carefully remove support materials, put the parts into two alcohol baths, and then install and calibrate the heights of the ball joints. Doubling production would mean buying another Form 2 and some incidentals for it (cart, UV curing lamp, etc.), and I'd have to spend more time removing supports and calibrating the ball joints. I'll get faster at that the more I do it, and curing of multiple sets can be done in tandem.

If I sell enough of these, it will make sense to look into Xenocrates' CNC solution, because the limiting factor will start to be how many hours I have in one day. While it would require significant babysitting, there's no support removal or curing with CNC'd parts. With enough fixturing, my Probotix's huge XY envelope could allow me to mill ten or twenty of these at a time. I'd still have to hand-calibrate all the ball joints, but if the machine's tolerances are good enough, it might take less fiddling.

With either solution, there's also further testing to make sure that the parts can produce a good calibration. During the beta testing phase, the kits will all be mounted to my MAX Metal and run through a heuristic calibration to ensure that they're as good as the prototype I tested yesterday. I'll continue that practice in production to the extent that it's necessary.

Here's what the printed threading looks like up close in the IPA bath.

Well by "casting", I was referring to an aluminum casing that was post-processed on a mill for holes, threads, etc.

g.

Has anyone seen a Duet Wi-Wi with 7" LCD Mount for Max Metal frame? I printed out the Duet Housing stuff from the "Max-Metal-master" files and just found out the Duet Wi-Wi does not work in that case. Been searching for one with no luck

Looking for one that mounts just the the Smoothieboard does.

I just purchased the 5 inch for my D300VS. Ultibots told me that the 7 inch was discontinued for now.

996pilot wrote:I just purchased the 5 inch for my D300VS. Ultibots told me that the 7 inch was discontinued for now.

The 7" version was discontinued by Duet3D due to quality issues with the supplier of TFT panels. But we now have a new 7" version with integrated controller electronics. See https://www.duet3d.com/DuetAddons/IntegratedPanelDue. It's available from Filastruder too.

Already have all the hard ware, from filastruder.com Just looking for case to print like the Smoothieboard one.