Model MF - GB CLOSED

[matt3o]

Printed samples of the side spacers, but really think we need a pattern on the side to break up the "3d printed" look. I threw this together, and will give them a go on the printer later today. Keep in mind the blue cutaways will be much less pronounced in real life. I realize this is silly stuff, but want to do what we can to make this has polished a product as possible.

I would probably use a regular honeycomb pattern. can you share 3D models, I've got a 3d printer myself I'd love to play with it

[lot_lizard]

Reduce the infill , you will be amazed how little you need for strength.

I tried a 50% honeycomb, and the I could visibly see it on the walls. I could tell it wasn't as rigid when bending around, but certainly didn't warp, crease, or crack in any way with some serious stress (this co-poly really is a special material). I should probably try going back to 50% on the honeycomb infill, and try bumping the walls up to 1-1.2mm to see if it eliminates the appearance issue (it was a nit, but annoyed me).

Bump up the print speed on the infill , it does not matter how accurate that is.

I'm currently at 40mm/s on the wall, and 60 (default was 50) on the infill. Let me know if you think I should try even a little more. I just can't have mess-ups, or I really do start running out of time, but really like the idea of trimming down the timing where I can

They look far enough apart to print one at a time not all at once , this will allow you to bump up the parameter speed and acceleration without stringing, it will also reduce the need for retraction.

They are currently set to print one at a time (might not have been clear before). I like that better as well, plus if we do have a screw-up on one cycle, hopefully we can salvage some of the parts

I would probably use a regular honeycomb pattern. can you share 3D models, I've got a 3d printer myself I'd love to play with it

Happy to ... The right spacer is identical for both form factors. https://github.com/lot-lizard/Model-MF/ ... ris-Shield

Appreciate the thoughts... I am not TOO concerned about timing unless we start experiencing some issues along the way. All of the retainer clips are printed, along with the nylon stabilizer inserts (both turned out nice). Each will need to be tested against the finished product though to confirm. I did raise the C/micro/mini USB ports with a smaller lifter insert that you will slip on the retainer pegs to raise the USB controller itself... they were just too close to the bottom edge

[andrewjoy]

Your using PETG right ? Or at least a co poliester ?

The ultimaker , as with any good core XY printer can fly 120mm/s easy on PLA, but PETG like materials need to be printed no faster than 50 - 60mm a second, i would bump the walls up to 3 ( so thats 1.2mm, 3 0.4mm walls, assuming you are using a 0.4mm) and take the infill down to 40% infill. There are good settings for PETG on a solid core xy in this video https://www.youtube.com/watch?v=xFt7vzweUiw I would switch to sli3cr if i was you, cura is good for just throwing stuff at the printer but you get much better advanced options with sli3cr.

What speed do you have the fan set to ? You may be able to push the PETG a little faster with more fan speed 60% or so, whist this will reduce layer adhesion a bit , it will still be vastly superior to the layer adhesion of PLA and ABS.

I plan to convert my i3 wanhao to this printer. i have already ripped it apart and have ordered all the stuff to do it.

I would test a few prints on the UM2+ we have in work to get optimal speeds for you but i don't have any PETG type material to test, all PLA here and ABS for our UP plus2 's ( beacuse if you want to print forever in ABS and have it never fail you need an UP!)

[lot_lizard]

Your using PETG right ? Or at least a co poliester ?

http://colorfabb.com/co-polyesters/colorfabb-xt

Really appreciate the thoughts. Can I get away with .2mm layer height? I haven't even tried that yet. That obviously cuts the time in half, but I am assuming there is a great opportunity for error if the room temperature isn't consistent, etc...

[andrewjoy]

Ahh colourfab XT , nice stuff its a modified PETG just like E3D edge

yes 0.2mm will be fine i would say 0.2 is standard for mechanical parts, no need to go lower.

look at that video i linked , they are printed at 0.2 and some of them look like injection molded parts they are that nice,

is 0.4 the only nozzle size you have ? Could bump it up to 0.5 with a 0.25 layer height

EDIT

Also , bloody hell only 39 euros for a 2.2kg spool of 1.75mm XT , thats bloody cheap! Just look at what we get charged in the UK for a bog standard PETG

31 GBP for a 1kg spool

https://rigid.ink/products/petg-1-75-mm ... 7814523713

!!!!!!!

[lot_lizard]

is 0.4 the only nozzle size you have ? Could bump it up to 0.5 with a 0.25 layer height

.2, .4, .6, .8... I went back to .4 for this. .2 is the only other thing I have tried smurfing with to see the detail I could get out of the printer if I ever wanted it.


One of the issues I was having with these little spacers was getting them to stick to the plate. Tried everything, but they are so long and thin that they just weren't holding down. I eventually settled on putting a 2mm brim on the base, which seems to be working (at least at the moment).

I am going to try the .2 layer with the other adjustments we mentioned (1.2mm wall and 50% infill). Stay tuned

[andrewjoy]

What are you putting on the bed to make it stick ?

[lot_lizard]

What are you putting on the bed to make it stick ?

I've tried everything I could think of. Glue stick, painters tape, kapton tape, the cleaned glass... I haven't tried Jello yet

Maybe I should be really cranking the bed temp up. I have just gone from 60-70C

EDIT: I probably should try lowering the fan speed. They are off initially, and then eventually kick in. These prints aren't very tall, so we could probably get away with little or no fan and never have to worry about sandwiching. The left sides have an overhang, so I will need a fan at some point on those though

[andrewjoy]

I would say 70 - 75 is the sweet spot for co polyester
Slow down your first few layers and turn off the fan for the first few layers.

[matt3o]

I would go as low as 25% infill with a cubic fill pattern

[lot_lizard]

Okay... So I did the following:

• Reversed the direction of the spacers so the heat from the fan is trapped better (half-a55ed draft shield)
• Set the line height to .2mm vs. .1
• Set the infill to 50%
• Bed temp from 60 back up to 70
• Reduced the fan speed by 30%
• Removed the brim (all or nothing on this test)

Now I can't pry the damn things off the plate

They look good. I might kick the fan back up a bit so the screw holes turn out just a touch more rounded (look close to perfect), but I think we are very close to dialed in. Unfortunately these things are so long and small (height and width) that the infill patterns don't have much impact on the overall timing with a 1.2mm shell. What it did noticeably do though is reduce the amount of time the infill takes. The issue all along really has been that the infill (even at 100%) would occasionally leave a slight ridge when retracting (this Cura software elects to use some very random patterns, and I haven't seen a way to alter the algorithm). The ridge would harden enough that it would impact the shell pass, causing a "speed bump" as it passed by. Enough slight bumps in the process, and the silly thing would come off the plate.

The combination of the .2mm and reduced fill time have eliminated the hardening of any "glob", and now we cruise right along with smooth beads throughout. The other MASSIVE upside of the .2mm is we have cut the print time by half. The definition of the print is obviously less, but really only noticeable on the portion that would be covered up anyway. I'll try to get some pics later.

Appreciate your thoughts andrewjoy and matt3o. Also, if anyone decides to take the STLs and apply your own patterns with success, upload that for everyone if you would.

EDIT: photos from the above mentioned prints

IMG_0364.JPG (1.37 MiB) Viewed 5323 times

IMG_0366.JPG (1.22 MiB) Viewed 5323 times

[andrewjoy]

Looks good.

The ridge would harden enough that it would impact the shell pass, causing a "speed bump" as it passed by. Enough slight bumps in the process, and the silly thing would come off the plate.

Increase your retraction speed ( i think this is more an issue with the 3mm ( or 2.85 if you want to be accurate ) filament as more of it melts per mm extruded than 1.75 thus requiring more harsh retraction) and if cura has the option for it reduce the extrusion multiplayer for the infill only by a few % , look for an option to print shell first ( not sure if cura has that) this should help too.

[lot_lizard]

All of the white right-side spacers (applies to the full-size and SSK) have been produced, and onto the left-side for the SSK. Should be done with the whites altogether by the weekend. I mistakenly thought I have co-poly in black already, but the black is in nylon only (using for the stabilizer clips which are next). I have ordered a couple of black Colorfabb XT reels and a clear so I can play a little with indirect LEDs in the future. The following images show the progress to date on the spacers.

Also, I spoke with our box friends, and we are going to start production on those soon. They are going to make one more change for me in prototype form. I need the width and height of the box reduced by 1/4 inch each way to make the palletizing of the Euro shipment work. Otherwise I will be hit with a non-standard freight dimension fee

All right side white spacers for both the SSK and Full-size

IMG_0370.JPG (1.63 MiB) Viewed 5272 times

Right side on top, left side SSK on bottom (inverted)

IMG_0371.JPG (588.75 KiB) Viewed 5272 times

Left side SSK

IMG_0372.JPG (1.32 MiB) Viewed 5272 times

[POTV]

Just thought I would show an interesting keyboard made by a 3D printer - looks pretty advanced...

https://3dprint.com/120299/3d-printed-dactyl-keyboard/

[Ir0n]

The spacers look like ribs... Lmao.
spoooooky!

All this 3D printer talk is making me want one more.

[andrewjoy]

The spacers look like ribs... Lmao.
spoooooky!

All this 3D printer talk is making me want one more.

Build this one with me . It will be fun!

http://www.thingiverse.com/thing:1752766

I am learning so much putting this thing together.

[lot_lizard]

White spacers are done (weren't too many). The black filament delivered today, so anxious to try that this evening. On the Dactyl bit, that is a really fun looking project. It's nice that it prints a single seam case for each side. We would need something in a minimum of 4 pieces as well merged into one vs two cases.

IMG_0491.JPG (1.66 MiB) Viewed 5126 times

[lot_lizard]

Build this one with me . It will be fun!

http://www.thingiverse.com/thing:1752766

I am learning so much putting this thing together.

I would LOVE something that is 500mm wide. That would really change my world for these projects. How easy would it be to adapt this to that size?

[andrewjoy]

very easy , its built to do that , just get longer rails and aluminium extrusions for the axis is question.

[lot_lizard]

very easy , its built to do that , just get longer rails and aluminium extrusions for the axis is question.

Was assuming, but thought there might be a devil in the details. That would be the golden child for me... 500mm x by 250+mm y heated bed. I could likely get away with a 250mm z-axis, so anything higher would just be gravy.

How many hours will you have in on it you think? Before you multiply that number by 1.8x to account for reality (scope creep)

[andrewjoy]

I have done the ramps control board the hot end and the extruder, not long on them a few hours . Main reason for the delay was dry solder joints on the crap Chinese ramps board i ordered because i am cheap.

All steppers stripped down cleaned and tested as they where from my old printer.

Still waiting for the t slot nuts and brackets to fit the frame together.

The firmware is fun to play around in , you get to lean how the printers work and process g-code and how the axis move , helps when looking for issues on pre-built printers to know how the controller is put togther

This is my BOM

https://docs.google.com/spreadsheets/d/ ... edit#gid=0

Once its up and running i am going to switch the control logic to something a bit better.

http://smoothieware.org/

In this setup the printer is being held back by the slow logic of the arduino as it has to calculate the core xy kinematics, it can just about handle it. Its funny because you see people building delta printers or selling them on ebay and the little 8 bit controller does not have a hope in hell of calculating the kinematics for a delta fast enough.


EDIT

Oh and the PSU has had some work done



40mm fan swapped out
voltage regulators fitted to the heatsink properly no longer flapping in the breeze
additional heatsink added to regulators

I am confident it can now pump out the amount of power it claims on the label

remember if its from china divide the safe power rating by 2

[lot_lizard]

The blacks are off and running. Will be several days before we plow through them all, but I'm a huge fan of how the first couple turned out (don't photograph well without natural lighting)

IMG_0505.JPG (583.12 KiB) Viewed 5148 times

[ohaimark]

^ Terrible out of context.

[lot_lizard]

Haha... Good thing I didn't mention the part to andrewjoy that they seem to prefer warmer temperatures as well for some reason

[DMA]

^ Terrible out of context.

andrewjoy wouldn't the frame be wobblier if you just make everytghing longer keeping cross-sections the same?
This can affect head stability - it's pretty heavy after all.
Or you should start caring for that after a meter or so?

[andrewjoy]

^ Terrible out of context.

andrewjoy wouldn't the frame be wobblier if you just make everytghing longer keeping cross-sections the same?
This can affect head stability - it's pretty heavy after all.
Or you should start caring for that after a meter or so?

The aluminium extrusions are bloody strong until you get to around 800mm in one direction i would not worry about it.

the thing i would worry about with a larger bed is the bed sagging. but that is easy to fix , just mirror the stepper lead screw and rails on the back of the bed on the front. You would have to have a second cross member on the bed and make the bed long enough to reach the rods.

I would also at this point switch the x y gantry back to steel rods over alu tubes , at the size we are talking for the default printer they are totally fine . but i think over about 400-500mm they would start to flex some.

I would even consider altering that axis all together at about 800mm and replace the rods with an aluminium extrusion ( 2020 like the frame) and use a hiwin linear rail for that axis.

The blacks are off and running. Will be several days before we plow through them all, but I'm a huge fan of how the first couple turned out (don't photograph well without natural lighting)

That so needs to go on the front page

[lot_lizard]

The following are what we ended up going with for the backplate labels (for serial numbers, dates, etc.)

Spoiler:

Screen Shot 2017-01-24 at 8.41.06 AM.png (1 MiB) Viewed 4994 times

http://www.avery.com/avery/en_us/Produc ... 05523.htm?

EDIT: wrapped in a spoiler because the image is obnoxious

[lot_lizard]

The final box quote is in with the modified dimensions and looks like it will palette well. I am checking on the pricing differences between #200 vs #275 stock. If it isn't much, the #275 is certainly interesting. Given we aren't very big in 2 of our 3 dimensions, either would certainly be sufficient. We have budgeted for another $.80 per box though, so why not splurge. We'll see...

Either way, another prototype will be made. I will likely order extras of these to keep on hand for future endeavors and reduce future costs. We will own the cutting die btw (one of the few deliverables that has future value)

[livingspeedbump]

so why not splurge.

Sounds like a good plan. Live a little.

[E TwentyNine]

so why not splurge.

Sounds like a good plan. Live a little.

I think he's just going to build a fort out of the extra boxes.