Hydrogen Generator

[pyronaught]

The stacked method of batched gaskets looks like it may work, it is just a pain separating the gaskets. The quality of the prints is usable, but a lot of time is spent prying off support material from the parts. I'm looking at the "Separation from Part" settings on support to try and figure out how to loosen the grip of the support material between the parts, but not really sure what to adjust though. I've never fully understood how support material is kept from sticking to the part just as much as any other layer. Anyone know the details of how this works?

[Tim]

Anyone know the details of how this works?

Generally the separation will be the height of a layer, e.g., 0.2mm, so one entire layer is skipped, and the support layer does not get squashed down hard onto the layer below, but just sort of drops onto it, giving it a lot less contact area. If you run your plastic too hot, it will melt into the layer below anyway. You will have a delicate balance between printing the parts hot enough to reduce the "sieve" effect of not having a solid interior, but cool enough not to have the support structures permanently bonded to the part.

[pyronaught]

Anyone know the details of how this works?

Generally the separation will be the height of a layer, e.g., 0.2mm, so one entire layer is skipped, and the support layer does not get squashed down hard onto the layer below, but just sort of drops onto it, giving it a lot less contact area. If you run your plastic too hot, it will melt into the layer below anyway. You will have a delicate balance between printing the parts hot enough to reduce the "sieve" effect of not having a solid interior, but cool enough not to have the support structures permanently bonded to the part.

Increasing the horizontal offset from the part helped. Normally it defaults .3, but upping it to .7 made the support material easier to remove. At .5 it was still hard to remove. I'm running the plastic very hot at 265, so lowering that would no doubt help too. I just really like how solid the plastic is when printing at that temp. I've never tried changing temp mid-print with a new process yet. Since the support is all on a horizontal plane, I could create a process for each layer where the support material is and then lower the temp on that. But on a stack of 30 or more gaskets that would require setting up 60 processes in S3D! No way would I want to go through all that trouble, I was just throwing that idea out there.

[Josh]

You could simply setup one job, with two parts, and support in between; then, step through and find where the second part starts. Cut out everything after the last bit of support, add a "G92 Z0.2" (or whatever is appropriate for the first layer position of the first part), then copy in all the code (sans startup and comments) from before that point. So you'll end up with a .gcode file that (if we call the first part A, the support between parts B, and the second part C) _was_ ABC, but is now AB(reset Z position _value)AB; adding more Z resets and AB sets will give you however many copies you want.

If that description doesn't make sense, let me know - I typed it rather fast X)

[pyronaught]

You could simply setup one job, with two parts, and support in between; then, step through and find where the second part starts. Cut out everything after the last bit of support, add a "G92 Z0.2" (or whatever is appropriate for the first layer position of the first part), then copy in all the code (sans startup and comments) from before that point. So you'll end up with a .gcode file that (if we call the first part A, the support between parts B, and the second part C) _was_ ABC, but is now AB(reset Z position _value)AB; adding more Z resets and AB sets will give you however many copies you want.

If that description doesn't make sense, let me know - I typed it rather fast X)

I know what you're saying, but it's probably still not worth the trouble compared to just lowering the temp and doing a one-process print.

One thing I've found that helps is to increase the "Extra Inflation Distance" parameter of the support settings so that the turn around points where the accordion pattern moves to the next row will occur outside of the part (in mid air in this case). Those turning points are the spots that stick the worst, so by keeping them out from between the parts it makes removing the support material easier.

The upper and lower vertical separation layers settings don't make any sense to me. When you increase those, there is an actual air gap left in the model between the top and bottom of the support material and the parts. Of course this air gap can not be printed, so what happens is that when the first layer of the part begins printing above the upper air gap, the traces just fall into the gap and then the next layer above that doesn't adhere well either and you get a stringy surface on the bottom where all the traces aren't stuck to anything. The higher number you use for separation layers, the more stringy, unbounded layers you get on the part. It's a useless setting for any values higher than 1.

[pyronaught]

It looks like the stacked parts are going to work. The bottom side is a little rougher than the top side, but at least I can peel off the support material without using tools now. It still takes a tool to initially pry the two pieces apart though, it would be nice if they could just be peeled apart by hand. The number of days required to print all the parts will almost be cut in half now since the machine won't be sitting idle when I'm not around and it can be run 24 hours a day. There's also the time saved in not having to do this 120 times between prints: let the platform cool, remove the part, prime platform with ABS slurry, wait for platform to heat back up, start the print, wait for the startup sequence. Even if all that only took 5 minutes (and I think it might take longer than that), doing it 120 times actually adds up to 10 hours of work!

[pyronaught]

Here's some stats on the stack method of multiple part printing: stacks of 15 frames are used so that each print job finishes inside of 24 hours. So each morning a new print job is started, which will take 6 full days to get all the frames printed. The additional cost of the filament used to print all the support material is $16, making all 120 frames cost $85 vs $69 for printing them one at a time. An additional 24 hours of print time is added by having to print the support material, making the total print time 138 hours vs. 114. About 8.5 hours of manual labor is saved, minus however much time it takes to separate all the frames and tear off the support material. I think the actual time saved will wind up being closer to 4 hours after accounting for the separation time. The primary advantage is being able to "set it and forget it" at the start of each day and not having to swap out a part every hour all day for two weeks, and also getting all the parts made in about one week instead of two.

[pyronaught]

I had to bail on this stack printing concept. The process of separating parts and removing the support material was just more annoying and time consuming than doing a separate print for each job and the parts don't turn out as clean.

[pyronaught]

Printing at 10,000 mm/min now and still getting good results. I never thought I'd see myself printing into the red zone in S3D! A frame that used to take about an hour to print is now taking 27 minutes

[insta]

If only you knew somebody who had 5 M2's sitting around