Model Train Forum banner
41 - 59 of 59 Posts

·
Registered
Joined
·
975 Posts
Discussion Starter · #41 ·
Yes, but 8 figures would take roughly 8 times longer than 1


Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
11,096 Posts
Yes, but 8 figures would take roughly 8 times longer than 1


Sent from my SM-G781U using Tapatalk
On a filament model yes. Not on a resin printer.
 

·
Registered
Joined
·
2,691 Posts
Seems like its 8x resin or filament. resin may laser a little faster, but it will still take 8x to do the movement. Filament printers can also print different colors
 

·
Registered
Joined
·
11,096 Posts
Seems like its 8x resin or filament. resin may laser a little faster, but it will still take 8x to do the movement. Filament printers can also print different colors
A resin printer doesn't laser anything -- unless you're referring to very high end industrial ones. The ones in our price range use a UV LED the size of the build plate to print an entire "slice" (layer) 0.1mm thick at once. It doesn't matter whether that is one little part 2mm on diameter, one huge piece the exact size of the build plate, or 500 little unconnected pieces scattered all around the area. One layer, one dip of the build plate, 40 seconds. 10 layers = 400 seconds, 100 layers = 4000 seconds (more than an hour, and that's only 10mm high or less than 1/2"). There is no motion other than the up and down dipping of the build plate, and that has to be slow enough to allow a new layer of viscous resin to completely fill underneath the build plate and the partially printed model. Print duration is determined by 3 things (only): speed of movement on the build plate Z axis, cure time set for each layer (it's variable for different resins and operating conditions), with the first 20 or so layers set for a longer exposure time; and the number of layers to be printed, which is determined solely by the height of the printed item. The number or volume of pieces printed in each layer is irrelevant. Think of old fashioned acetate overlays for maps, anatomy drawings, or whatever.

Color may be important to some, but since I paint everything, it's irrelevant to me. The much finer resolution / detail achievable with a resin printer (in the current state of technology) is also a deciding factor for me..
 

·
Registered
Joined
·
2,691 Posts
So the printer uses an array of UV LED's the size of the build plate with the resolution based on the number of LED's in the array? Or like an LCD monitor facing the build plate which used a UV LED's for illumination. That would eliminate the xy motion.
 

·
Registered
Joined
·
11,096 Posts
Correct. There is no X-Y motion in the printer (setting the correct orientation on the X-Y axes is part of zeroing the printer). It moves in the Z axis only. The "monitor" as you term it, is composed of an extremely fine array of LEDs that emit most of their radiation in the UV spectrum. The scale of the individual LEDs is too small to see with the naked eye. The print is built upside down from the bottom up, with the layer currently being printed at the bottom of the resin vat and therefore exposed to the UV light. When completed, the build plate moves to "Max Z" position. The build plate with the print attached can be moved to a solvent cleaning bath as a unit, or the print removed and cleaned by itself.

The post processing cleaning can be done by hand. Anycubic sells a post-printing unit that cleans prints. I typically run one cleaning cycle with the print still attached to the build plate, then remove the print from the build plate, clip off the supports, and run the print through the cleaner again by itself. The build plate goes back to the printer... hopefully having been handled gently enough that it does not need to be re-zeroed before the next print. I generally re-zero every 10 prints or so anyway, just to be sure (and of course, a failed print is a good indicator that you messed up the zeroing).

After cleaning, a finished print needs to be cured under a UV light for approximately 30 minutes to endure complete hardening of the resin. The post-printing unit mentioned above also has a UV light for this purpose.
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #47 ·
@CTValleyRR am I correct in my assumption that resin printers are less "fidely" than FDM? I'm finding lots of little variables with fdm. Large prints are prone to warping, small print prone to vibrations causing errors, poor adhesion,etc. Nothing insurmountable, but it's a learning curve and some trial and error. Seems like resin eliminates a lot of those variable (but I'm sure adds others)?

Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
3,711 Posts
actually no, fdm printers are less fiddly [i have both] ... the resin printer needs more cleanup, both the printer and the image .
it's really pretty simple, if you want rain water barrels as an example, you would pick a printer that closely matches on size, fdm for HO and larger... resin fr N and smaller ..
nice and simple ....
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #49 ·
Cleanup I can deal with, as long as I get viable prints. Failed prints are a bit frustrating.

While I will certainly print several N scale items, it will get used for HO (I have both) and general use as well. I'm hoping to print a 47 Plymouth and a White COE for my dad. Not scale specific, but penalty something around 1:50. Plus general use, replacing broken doodads or making a better thingymagig. I like to tinker in lots of stuff, so there will be lots of that, possibly even items for one of the 3 classic cars in the family for which parts are obsolete.

The possibility exists for both, I'll keep this loaner Ender 3V2 as long as I can. But considering the Photon Mono that's on sale for a few more days. So the question is, is there a lot of settings to adjust with each print in order to get a viable result, or just cleanup of the print and machine one it's done. Nozzle and bed temp seem to vary with different prints on the fdm and can cause issues (warping is the one in trying to figure out right now).

Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
11,096 Posts
actually no, fdm printers are less fiddly [i have both] ... the resin printer needs more cleanup, both the printer and the image .
it's really pretty simple, if you want rain water barrels as an example, you would pick a printer that closely matches on size, fdm for HO and larger... resin fr N and smaller ..
nice and simple ....
I partially agree. I'm not sure about the zeroing process. The resin printers are pretty easy to set up and level.
Scale. Post print processing, though, is definitely more work with resin than FDM (although a modicum of care taken while removing th finished print will avoid any significant cleaning of the printer). I have to say, though that it's a fair trade for the detail achieved -- I wouldn't base my choice of which to buy on that.

I'd disagree about scales, though. The fine detail is worth it in HO, too. Maybe you meant that you can't print tiny N scale objects with an FDM printer, which I'd be inclined to agree with. But if you can only get one, get the one with the resolution that meets your needs, regardless of scale.
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #51 ·
This is about the best I can do with the ender. It's workable, a little sanding and some putty will clean it up I think, or just cut off the brake wheel and replace with commercial available one. Point is, ya, it's a challenge for N scale and remain affordable. Maybe a high end fdm could do better, but I'm keeping it to about $200.


Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
11,096 Posts
I don't know if that's typical of FDM printers or not, but there seem to be a lot of thin strands of plastic to clean up.

Notice also the distinct layer lines visible on the the ends. That definitely IS typical of FDM printers. You don't see that on resin prints.
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #53 ·
I think the stands is typical. Comes off easily, just need a light sanding or file. The lines are as well for sure, sanding would help, but very difficult in this scale. Paint might help as well, but again, hard to sand this. So primers are less effective if you can't sand off the high points. Hopefully I'll be able to work on finishing it this amend and see how it goes.

I think there is room for both for me. The larger build volume of affordable FDM opens up a lot of options for kids toys, part prototyping, functional items etc. While the fine detail of resin is preferable in modeling.

Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
2,691 Posts
The stringy stuff is more a function the filament material. PLA is not stringy while PETG makes fine strings all over the place. As prints go, it looks great to me. I don't see the individual layers or any real artifacts from his FDM printer. The break wheel is probably beyond FDM, but one usually have a sack full in the junk box!
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #55 ·
This was PLA, but it is old. And I think the temp was a little low on that one. She on the brake wheel, that's kind of the plan. Cut it of and use a spare from something else. I've printed the three of these cars now. One lifted and warped and is useless, another one the roof failed to print correctly, but the other piece are good. So I've got two usable cars.

Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
2,480 Posts
The stringing can happen if the temp is too high or too low. There are also at least three main retraction settings that control your stringing. For the overall quality, you can gain a lot by cutting your print speed in half (default should be 60mm/s, drop that to 30). Just changing that should at least give you a better idea of what the printer can do.

At that small of scale the brake wheel will never print vertically on FDM, but you could print it separately and horizontally, then glue it on. The rails at the ends should have printed fine though... Going from one support to the next is called 'bridging' and it's not uncommon to see bridges of 1-2" completely unsupported. The stringing may have destroyed them during printing, and the retraction settings are also responsible for the ability to bridge, so these two issues will be closely related.

Are you printing with a skirt or a brim? I ask because it looks like you've got elephant's foot on your first layer. This either indicates that your nozzle is too close to the bed or your first layer settings in your slicer are wrong. You might also check your E setting (when you tell the printer to extrude 100mm of filament, do you get exactly 100mm?) as this might explain why the layer lines are so visible and the stringing is so bad.

What nozzle size and layer height are you using? Even with the stock 0.4mm nozzle I would go with 0.08mm layer height on this, but your first layer is going to be 0.2mm regardless so you may want to go with a raft for bed adhesion so all layers of the car itself are only printed at the smaller layer height. Either way, you've got to nail that first layer or everything after it will be distorted.

I would say at this point ALL 3D printers are still pretty fiddly, plus there's that steep learning curve because there are just so many different factors you have to master to really get the fine details, but once you get everything figured out your prints will show the difference. Keep poking at it, you'll get there...
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #57 ·
So, I understood some of what you said, haha. I'm pretty sure it was English.

Honestly, I don't know on most of that stuff, I ran the file through the slicer, put it on the card, into the printer and hit go. Played with temps a bit, but that's it. But anything larger than this I have struggled with. Warping, loss of adhesion, shifting, etc.

The nozzle is stock (.4mm?) I'll look for those other settings and play around a bit. Maybe I can get better use from it. While not perfect, I still think I can use 2 of them. If I can get them a little cleaner I'd be thrilled. And id love to print some larger prints, stuff that requires supports.

Sent from my SM-G781U using Tapatalk
 

·
Registered
Joined
·
2,480 Posts
That's why I offered to help you out before... ;) Yep there's a lot to learn if you want to get good control over your results. There's a ton of info below, so just work through it one step at a time.

First thing to do is check your belt tension. The X axis belt should make a tone, like strumming a guitar. Not terribly tight because you don't want to strip out the screws in the aluminum frame, but do what you can. The Y axis belt is longer, I find it doesn't need to be as tight but you should still get a tone out of it. Tight belts mean less slop as the head moves around, so the sides of your print will be cleaner.

The next thing to tackle is bed leveling, and this is not only one of the most important setup steps, but it is also the one nearly everyone struggles with. You have leveling knobs underneath the bed, and the idea is to adjust them so that the nozzle is the proper distance from each corner. Take a look at the springs underneath -- are they almost completely closed? If not, the leveling needs to be done from scratch (loose springs means the whole bed will go out of adjustment from the vibration every time you make a print. Do it right and you won't have to touch it for a year). Also helps to know if your springs are silver or yellow.

There's more than can be done to the hardware, but this will be enough to get a start on quality control.

After that it comes down to the slicer software. I forget which one you said you were using? I've been using Cura, it's a decent program and it's free. Regardless of the slicer you use you should have very similar settings, they just won't be organized the same way.

The first thing to check is your bed adhesion method. A 'skirt' means it lays down a round or two of filament around the perimeter of your print, but this doesn't actually touch the print. It's just a way to get the filament flowing before starting on the good stuff. A 'brim' will lay down several rows around the print, actually touching it. This helps prevent warping at the edges but you also have to remove it and clean up the edge later. Very handy for really small parts though! And finally, a 'raft' means that it lays down several layers on the bed first, then your print is created on top of this. This one can be tricky because you get some advantages and it can eliminate warping, but if the settings aren't right then your print will fuse to the raft.

Somewhere in your settings should be a way to set the layer height. There will be a separate adjustment for the first layer height -- NEVER change this value until you know what you're doing! That first layer is critical to the success of your print. However the general layer height (which is probably also set to 0.2mm) can be adjusted. With the 0.4mm nozzle you can get decent prints down to 0.04mm (note the extra zero) but I find 0.08mm makes for excellent quality. I use 0.2mm or more for large prints or when prototyping a new model. Of course using a smaller layer height means your print will take that much longer to complete.

I should also mention, for the software you're using, did you find a profile for your printer? If you're using some default setup then the quality is just going to naturally suck. Every type of printer is different so you want to start out with settings that were tuned for your device and go from there.

For the stringing and bridging I mentioned this is related to your retraction settings. In Cura these are under the 'material' settings. For this printer the best starting point is a retraction distance of 6mm, and a retraction speed of 25mm/s. If your values are different, take note of what they're set to and try mine instead.

And finally, your temperature settings. This is going to vary quite a bit depending on the filament you have. I don't just mean whether you have PLA or ABS or whatever... I mean different colors from the same manufacturer can use different temperatures. When I find a good temp for a roll I'll write it on the side of the spool. For PLA I generally start around 205. If it doesn't stick to the bed it needs more heat. If it starts warping it's probably too hot. Adjust in 5-degree increments until you find what works well. In your slicer settings, the first-layer temperature should be five degrees hotter than your print temp. For the bed I've found 65 for the first layer and 60 after that is good, but I have a glass bed which changes things a bit. You may do OK starting out at 60/60.

OK now that you have some initial setup in place you want to test it out. Look on thingiverse for 'calicat' (there's a ton of other calibration animals you can use as well). The original size is 20mm square. I like to print mine at 50% (10mm). This is like a small torture test and there's a lot of fine detail you want to look for. Does the first layer stick? If not, increase the bed temp (and make sure you cleaned the bed before you started printing). If it completes, look it over. The tail is difficult to get perfect, especially the underside, but dialing in for this will give you nice bridging and overhangs. Can you see the eyes, nose, and whiskers clearly? This indicates the ability to print fine details. Other than a single strand where the print finished, is there any other stringing around the ears? Get this nailed and you'll print the railings on the car without any trouble. And of course check the sides to confirm good alignment of the layers. The corners will be rounded because of the nozzle size, but should still be crisp. I keep my final calicats in a row on my desk as a reminder of what color filaments I have (15 so far!) and to see what kind of detail I was able to get from each. Some filaments are much harder to use than others, and the cheap stuff has additives that create inconsistent results.

That should be enough for you to absorb today. :D Steep learning curve, but it does get easier.
 

·
Registered
Joined
·
975 Posts
Discussion Starter · #59 ·
That you did! And I forgot about your offer, thanks for stepping in! I now have some solid places to start from. And I'll try those this weekend hopefully. Sending aPM to you now.

Sent from my SM-G781U using Tapatalk
 
41 - 59 of 59 Posts
Top