RoStockFAZZ

[mikefazz]

Hi all,

So after plenty of time with my two delta printers (RostockMAX v1 and SpiderBot) I felt the need for a bigger one. Having designed my own replacements parts for most of the machine (other than the frame) I decided to go for my own design from scratch because its more fun that way. My goals were big, rigid, accurate with a chamber, and not too crazy expensive, or heavy on custom manufacturing. I like 8020 but didn't know of good options for 60 degree corners until I found 40-4096... At this size I don't trust plastic vertices.

So after many evenings playing around with a design I have something close to what I will eventually build... looking for input here...

v1.0.pdf

(1.71 MiB) Downloaded 309 times

(3d pdf)

So here are the specs:
- 400mm round build surface
~ 870mm build hight
- Water cooled triple hot ends
- Chamber using polycarbonate multi-wall

Components of interest/concern:
- Heated build surface, current plan is to put power resistors on a 0.25" aluminum plate, didn't find any good/cheap options at this size
- V-rail 1500mm length, never used
- Common GT2 belt (5mm wide?) should I go bigger for such long lengths?
- My triple water cooled hot end (http://forum.seemecnc.com/viewtopic.php ... 454#p56454)
- igus ball ends, I haven't used them before
- Chamber walls fit in 8020, using polycarbonate multi-wall from HomeDepot, I wanna avoid getting parts laser cut
- Megatronics v3 (I like thermocouples)
- Updated version of my extruder (https://www.youmagine.com/designs/nema2 ... e-extruder)

And of course how much will it cost? $2270 I hope to get the price down by ordering the 80/20 from a distributor and not buying it on their ebay site.

BOM.xlsx

(15.66 KiB) Downloaded 366 times

Still a few details left but feel I am getting close

[brent]

That looks incredible!

[teoman]

This is going to be interesting.

I am also dreaming of doing something similar in my next lifetime.

One thing I noticed is you have selected an atmel based board. There has been a lot of debate and it would seem that foe almost the same price you can get a much much much more powerful card.

Have a look at the replicape that goes on to the beaglebone black. With linux cnc installed you will not need to have the raspberry pi either as it can serve both functions.

[Durandal]

As for the question regarding the belt, my Max has a height of 815mm, and the 5mm belt is fine with it, produces very accurate dimensions even after some 700 hours printing. I think it would be highly reliant on the mass of your end effector, that triple water cooled hot end can't be light is my thinking.

[jdurand]

As for the processor, many of these printers use processors based on Arduino boards. That means almost anything is more powerful.

I'm not an Arduino fan after having to design products that my client needed to work with Arduino boards.

[teoman]

I have mixed feelings about the arduino.

I have about 100 of them (Great fan of automation). They are perfect for single task small applications. Prices are also good you can get an UNO clone with an LCD for 11 usd, and the nano's i recetly purchased about 25 for 2.5 usd each, You cannot beat that price. And for quick and single task projects that one would want to deploy they are perfect. If you were building a kit printer i would say go for it. For something like the orion or the rostock max that one will not fondle too much they are at their limit but adequate. However for an ongoing project (which i think your printer will be, as you will be wanting to constantly improve) i would recommend something a bit more powerful and capable of expansion without the pain of swapping boards and having to redo all the wiring.

[mikefazz]

Hi all thanks for the input,

Durandal: Thanks for confirming about the belts, I haven't had any issues with mine but my SpiderBot has really wide belts and is smaller than the MAX. It is much more rigid than the MAX so I am wondering if wider belts would remove a little slop. While everyone sees my water block as heavy it really isn't that bad...

Teoman, idurand: There are certainly lots of electronics options and on the arduino side I have messed with many (cupcake, RAMPS, RUMBA, Rambo, Megatronics 2 & 3) so I am up for trying others but haven't done the research on ARM or other 'CPU' based ones yet. I haven't seen the need to move away from arduino but just because it ain't broke doesn't mean I can't fix it;)

Holding off on building this for now... hopefully it isn't out of date by the time I get to it later in the year.

Mike

[626Pilot]

If you use a Smoothieboard (or Azteeg X5 or other clone) you can run my auto-calibration firmware. It uses a machine learning algorithm to figure out the true dimensions of the printer (arm length, endstops, delta radius, tower radius and angle offsets, and surface normal), producing improved dimensional accuracy on all axes. If enough slop is left over to impede 1st layer adhesion, the firmware can also depth-map the bed and use this to continuously adjust Z as a function of the bed surface.

This requires a Z-probing apparatus. If an effector-mounted probe is to be used, it must be mounted dead center. Bed leveling and geometry adjustment data is stored to the SD card, so you don't have to re-run the calibration unless the printer is moved.

[mikefazz]

Well its been 9 months and I have finally gotten through the majority of the build.

I made a few improvements:
- Re-did the bottom to be simpler and have the bed leveling screws in line with the vertical axes
- Extended the chamber to above the filament to hopefully help keep it more dry
- Adjusted the location of the axes to get more build space

The cost with 95% of the parts included is at $2300 (but involves a decent amount of labor to make some of the special parts)

It is coming along pretty well and looks good (if you like that 80's robocop look;)

[stonewater]

your vertical supports... what are they called and where did you get them? looks like pairing them on each corner has given you the angles you need.

Tom C

[mikefazz]

They are 80-20 4096 extrusions, I think they are fairly new at least when I started the design. They are only in the metric 40mm size but I found the 40 size is close enough to the 15 series (1.5") which I had some of. they are also nice to use since they have a number of internal channels which help with wire management and electrical shielding from each other, though I haven't' gotten to wiring it all up yet.

Here is the current BOM which shows those parts.

BOM_v1.1.xlsx

(24.03 KiB) Downloaded 307 times

With the exception of the openbuilds v-rails the structure is all 8020 and so far seems very rigid more event then the much smaller spiderbot I have.

I am still trying to figure out which electronics board to go with. I have a few messed up megatronics boards which I don't want to deal with anymore. I am looking for a 32bit+ board that supports 3 extruders and 3 thermocouples... so far the alligator board seems like a possibility. Now if only there was a board that could handle this 400W heated bed... 34A is a bit out there.

[bvandiepenbos]

Great looking machine. I like the 8020 extrusions, nice stuff.
I have been thinking about the 4096 profile for quite a while for a frame, but had not figured out how to use it.

[mikefazz]

Thanks... not quite as simple as the MAX METAL but does work well for building around having an enclosure. If 8020 ever releases a '2096' I may consider designing a smaller SeeMeCNC size version... or just buy your upgrade;)

[mikefazz]

So I am almost done with the build, lots of little details that took quite some time, and wiring has proved to be rather tedious, I wish I had routed wires down the extrusions earlier...

Anyways I am currently stuck on one electronics issue.

I am running a RADDS on arduinio DUE and I keep blowing the 10A fuse (non-heat bed fuse).

After looking at my loads I am wondering how my other printers which use megatronics and apparently have a 7A capacity, don't also blow their fuses.

So here are my loads, all on the 10A fuse:
3 - 40W heaters (~3A each) 9A total!
3 - NEMA17 axis stepper motors (from ancient MB cupcake) measured ~6 Ohms/coil (~2A max) or 6A total
3 - NEMA23 extruder steppers measured ~2 Ohms/coil (6A max) or 18A total

Now I know the steppers don't pull that much as they are driven by DRV8825 pololu drivers which can supply 1.5-2A each.

So during testing if I hit home so the axis steppers are on: ~4.5A then turn on all the hot ends at the same time (+9A) the fuse blows 13.5A>10A

So 2 questions:
1. Is the 10A fuse too low and how safe would it be to put a 15A fuse in instead? Would I fry some traces?
2. How has this not happened on my other machines (with megatronics boards)? I have definitely had the axes steppers and all extruder heaters on at the same time and no fuse blown.

I don't have part numbers for the little NEMA17 motors but from an online forum they supposedly have 35 Ohms/coil yet I measure much lower resistance.

[mikefazz]

So for anyone following along the issue was the little NEMA 17 motors. The ones I have: moons type 17hd4063-01n I can't find any documentation that gives its operating voltage but whatever is going on after replacing them I haven't blown a fuse yet. I still don't get how the apparent undersized fuse works but at this point if it works I don't really care. Now onto calibration...

[mikefazz]

So I am on my first print!

Some new features I haven't mentioned yet:
- RADDS yes I decided to go for a new 32bit controller
- Flying Extruders mostly worries about long bowden tubes, and thought I would give it a try

I did a quick overview video feel free to check it out here:
https://youtu.be/uwi8UBFYH8s

Still have to try out the other extruders and it could use a bit of work getting the wiring neater.