What are the considerations should be cared for when building a large 3d printer

I am currently in the research period to build my first large size 3d printer and I would like to know what components to select an what considerations to care for on designing such a machine
I need a building Space of (at least) 500x500x750 mm with considerably high resolution (0.050 layer hight) and the extruder should be fully metallic to endure a temperature of 600C, aluminum V slots will be used for chassis I guess with tempered glass for housing or a 4mm acrylic sheets maybe to control the atmosphere of no more than 100C of temperature
Where shall I start?? What should I choose for an extruder, transmission system, firmware, steppers, and controller
I will be in need for a guestimation of the BOM to such project
Thanks in advance!!

If you are starting from scratch, without any experience with making a 3D printer, you'll need to realize the task ahead. Design, fabrication, learning firmware, then slicing and control software.
Plenty of people do this and you can too. It's definitely a fun journey.

What resources do you have? Can you use a real 3D CAD package? Do you have a well equipped workshop? Are you a completely new dreamer that's drawn to this hobby?

I'd clearly state what you expect this printer to do. People in here have heard it all, and I doubt you have a propriety idea.
Those specifications you want are RED FLAGS, and many in here won't even reply to you. There are so many 1 post dreamers that have come and gone over the years.
Also, those specs you want are pushing the boundaries of temperature/resolution/size, and you don't have the basic stuff learned yet.


600C/1100F for a hotend is smoking hot, but not impossible. Why that high? Aluminum melting point is 1200F so what are you doing?
To create a reliable and robust hotend will be a big challenge for you. It has to run continuously for days and days, not just minutes or hours.

The printer size you want is no problem, the 100C internal temp will require a heater of some kind. The build plate flatness will need good design and be very flat.

you are likely going to need 220V power, and hopefully you already know this and won't kill yourself accidentally.

The resolution is something to discuss. You want 0.002"/0.05mm height layers??

What exactly is your expectation for prints? With tiny resolution and large size, the odds of you getting a successful print are very poor.
The time to print will be days and not hours.

If you need advice: Just buy a used 3D printer and have fun.

Read:

My coreXY design is progressing to the build stage

[forums.reprap.org]

An Almost Reliable, High Precision, 3D Printer: Son of MegaMax

[www.instructables.com]

and
Mark Rehorst's Tech Topics

[drmrehorst.blogspot.com]

Then read them again take notes and heed

Can't thank you enough mate for such care and kindness!
I've been familiar with 1 post dreamers in other fields actually and probably I guess I know what are you talking about...
Let me tell you what can I do and you tell me how shall I start and where to go next...
I am a mechanical engineer, have been working with graduation projects for 4 years by now, which happens to need a bit of SolidWorks design plus the know-how of overcoming design-related manufacturing issues, thus let's assume that the experience needed for the first two steps (Design & Fabrication) are -almost- covered ...
The point now is being totally new to the 3d printing world with some knowledge of Simplify3D and some failed prints on the disk , I currently started working in a firm with 11 FFF machines while the country here isn't flexible with importing printers (the same problem doesn't exist for spare parts luckily) , so the thought of buying such one will be a little bit tricky here...
Assume a well-equipped workshop is already available as the business owner will invest anything to build this one (Will make him on top of the industry here) , we need such amount of heat to print peek and ultem
I am not intending to invent the wheel, All I need is the knowledge of how to select a proper bed, extruder, steppers, ..etc from well-known brands working in the industry and what considerations should be cared for while thinking of such a project
As for me, I would love to hear from you some tips or even a task list to do to get quickly inside this industry, where to start and what should I do to be of good experience so I may help some newbie one day like you do now

Thanks in Advance

Edited 1 time(s). Last edit at 09/27/2017 05:59AM by Hsm_Qady.

Quote
cozmicray
Read:

My coreXY design is progressing to the build stage

[forums.reprap.org]

An Almost Reliable, High Precision, 3D Printer: Son of MegaMax

[www.instructables.com]

and
Mark Rehorst's Tech Topics

[drmrehorst.blogspot.com]

Then read them again take notes and heed

Thanks Mate that was helpful!

A 100c chamber will severely limit what you can keep in the hot zone. I don't know of any hobby builds that completely separate all electromechanical components from the build chamber. I think you will need to start borrowing from Stratasys designs.

Holding a 2'x2'x3' volume of air at 100c for days at a time will dump a ridiculous amount of heat into the surrounding environment unless its well insulated. The temperatures you are looking to achieve are a much greater challenge than the build volume.

So, you are more than the typical "big hat/no cattle" dreamer.
It seems you want make a single machine, and then make/sell parts.
I ended up writing a manifesto, sorry in advance. I'm not busy evidently.

It sounds like you have some budget, which means you get servos and machined/bespoke parts.
This opens up creativity and solutions. My $.02 is my own, and might catch some grief, but it is free advice.
I'd ask your money guy to set aside a USD $40k starting budget, for components and development plastic materials.
I'd also assume a year before this contraption can be commissioned/ready.

Below is how the Fortus 900 works in general, and that is great place for you to start studying.

I'd assume that Ultem and PEEK need large diameter nozzles. Neither material will make tiny parts or large ones with fine details.
Just buy the Stratasys T16 and T20 tips for the Fortus line of printers. I wouldn't fuss with making them. plus you get to use the tip shields as well.
You can buy replacements as they wear out after several hundred hours.
Plan on using support material in a second print head.

Pinch the tip-tube assembly in a steel block pair and embed 4 electric nichrome heaters. for those temps, I'd guess 200 watts (4x50w)
Heat zone is nearly the full length of the tube, and leave a 1/4" sticking up.
You need to bring compressed fresh cold air into the head and focus air at the filament tube entrance throat.
I've done it like this without fresh air and it just bulges at the entrance.
Study the Stratasys patents on their print heads and see the theory behind the piston plug inside the tube length.
They detail what's happening pretty clearly.
For driving the filament, I'd use pinch roll pairs with a NEMA 23 servo. Maybe 2 pairs per filament, to limit strip out odds.
don't worry about the weight of these components.

You should assume a layer thickness minimum at .015 inch, road widths around .020 or more depending on settings.
Your cost of material is $8/cubic inch after shipping. Long prints will give you a major headache for usage and spool changing.
Ultem comes in 92 and 184 cubic inch spools, so plan accordingly. the 92/184 spools are $690/$1300 each.
My friend with a Fortus 900 has 4 day print jobs fail for various reasons, mainly geometry/part/support/warp/build plate separation related.
He scraps $4-5K parts without batting an eye. Get your head around failed prints and expensive oops.
It is guaranteed 100% to happen.
Also, for support material, the material costs the same, so if you have a failed $1000 part, you will throw away about $300 in support material.

As mentioned, an 80/20 frame work can easily create an envelope and cartesian mechanism.
Select big sections, like 8080 cross sections for the major members and 4040 and 4080 for others.
Large cross section frame gives good thickness for insulation.
You might even design a welded steel framework.

Use linear bearing blocks/rails for guiding your moving frameworks,
12mm or 15mm size is plenty, 20mm is too much.

For all motion, use the strongest Clearpath NEMA 23 servos and their 75V supply. $375 each and you'll need 6. They accept step/direction easily.
You can use the Azteeg X3 pro, and it gives plenty of inputs and outputs including 2 thermocouples.

I'd just do a simple X-Y moving gantry in the top area, and a dropping platform going down for the Z.

I would create a separate ATTIC zone, and put all motors in the top area, separated by bellows or sliding metal panels.
Your machine will grow in size so plan on a machine like 4x4x6 to account for lost travel due to ATTIC mechanics and bulky gantry.
Remember typical door frame sizes and you might have to create this machine in the final resting place.
The Attic will need fresh air flushing.

Ball screw driven moving gantry will use 2 screws/1 motor/XL belt to link them. 1605 or 1610 is great.
Go with a dual x-carriage head arrangement, using 2 more ball screws and 2 motors. Marlin firmware has dual x-carriage.
Parking the heads will give a drip free and ready to use heads.

For Z, drive with dual screws vertically 1 per side of a platform frame. 1 motor and belting again. This motor will need an electric brake on it to avoid Z dropping on power off.
guide the plate platform with 4 linear bearings of some kind, either round or block style. you should put the motor/belting in a segregated basement to reduce temps on that stuff.

For the build plate, 1/2 inch blanchard ground steel, electroless nickel plated is enough. It will be heavy, but a nema 23 with ball screws can move hundreds of pounds quickly.
If needed, you can also get reduction with your belting and pulley ratios. You can reduce weigh with pocketing underneath if you want.

For build plate flatness, you need to mount the plate with a 9 or 16 screw matrix from underneath. then when hot, you simply adjust each screw under the head for flatness and gap.
no heater needed on the plate, as it will normalize to the chamber temp. Study the patent from Stratasys for the Fortus 900.
Clearly outlined. Adjustments will need to happen with an open door and quickly so the cool incoming air is minimized.
you might also have an inductive sensor to give the final auto leveling, but use the screws for a basic flatness.

for chamber heating, resistive heating elements and circulating fans will be fine. You'll need a whole circuit breaker for this system, and that division might save your job if that circuit trips out.
wash the print elevation with fresh hot air, and be sure to turn over the chamber from top to bottom continuously and vigorously.
temperature gradients within chamber or if the platform starts hotter and lowers in temperature at deep Z, will cause big dimensional changes. I would create the heating and temperature monitoring in a separate system and not involve the printing electronics.
You will want to have the chamber on and heating hours before the print job, so that the thermal shape and warping will settle. Then tune the build platform quickly.
I'm not sure what Ultem sticks to, but I'd probably just buy the Stratasys build sheet and glue it down firmly to the steel plate. They use vacuum, but that is not a perfect system and peels up sometimes.

Honestly, i'd just buy Ultem parts from a service bureau and get busy selling and profiting on parts.

Why
500x500x750 mm with considerably high resolution (0.050 layer hight)

You can have big or high resolution but NOT both without big bucks!
Is there a readily available 500x500 bed heater?

extruder should be fully metallic to endure a temperature of 600C

Most FFF filament flows below 250C?
What filament needs 600C

May need your own power station to power this thing?

Perhaps you should really engineer this project!
What do you want to print, what purpose, what stakeholders.

Requirements
----- Requirement review --- what is most important/practical
----- Unique design? Traditional design? experimental design
----- buy / make trade studies
----- print style, FDM, SLA, Resin (resolution requirement may drive this)
----- Cost
----- Schedule
----- electronics, power, firmware, software
Requirements Validation

Design / development
----- lots of CAD drawings, cardboard models etc.
----- Design review
----- questions , comments, changes
Requirement verification

Build

Integration and test

Operation

i am just reading the digital dentist build of his UMMD and love it.
great design on all of it. Great print results.

Thanks! If there is anything in the write up that isn't clear, or if you think something is missing, let me know. I'll be posting more images of prints in the near future.

In the meantime, try this: [vimeo.com]

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Looks like I will need weeks to study your comments guys
that's exactly what I needed, Thanks for your care I will keep you updated
I guess I will be trying a 300x300x300 first with overdesign considerations to make it upgradable, don't know actually if this is the right way to go

ok, i think my advice is a bit over the top.
my cnc router last couple of years has me ball screw crazy.

you can use belts without issue. servos are always great. steppers work too.

my current project is near the size you want and it is conventional and not crazy. no attic, but i don't want 100c air temp.
no crazy ultem wishes, just regular volcano capable. i have 1 servo moving the gantry, and the rest are steppers.


DD: . started in your signature, to the blog, watched videos, then found the forum.
I can't believe you did a whole printer, in the public eye, and kept your sanity. (trolls are rare in here for sure) kudos. I just can't find time computer for writeups.

I wouldn't change a thing. plus you gave the data, although i'm not sure it's needed. Pictures are great. Love the Z.
when i did a cable driven core xy, i swore off that style. it was cool but slippage killed it.

I have that same 12x12 line level pad and I'm worried about my 1/4" mic6 plate warping because it's like 20x17 or something. I haven't cut it down yet.
I think the center hot spot will make the plate a bowl shape. I'll do your ssr power safety idea, but since there is a thermistor embedded into the orange matting, I wonder if it can fail.
i'm also thinking of just doing chamber heating and use normalization time to get it all up to 60C or so. my results in the past with ABS in the 40's had issues, and I think more temp is better.

i starting to getting my head back into 3d printing, back from routing. I think i've said that twice in the last year........

300x300x300 is something you can do, following a bunch of ideas. You can buy that size from several people, airwolf, taz maybe, some others.
this forum is obviously a make your own, share your data, use group knowledge when you want. developer, maker, innovator, pioneer, etc.

read the Digital Dentist UMMD forum... That 80/20 frame and mechanics is a great foundation, and you can construct it quickly, and sort the final details.
there's a lot to learn in firmware, slicing, printing. It can be rebuilt/expanded as you wish easily, and you can transplant the guts.

the ultem wishes is the biggest cost driver for part cost. i have some parts made with it, and it's super nice and incredibly strong.

I wonder if ABS or a blend or something would be enough for your needs. The newish PET stuff is what I'm doing first with my new build.

You still haven't said what you want this machine to make.
Can you share the end part needs?

Quote
davew_tx
DD: . started in your signature, to the blog, watched videos, then found the forum.
I can't believe you did a whole printer, in the public eye, and kept your sanity. (trolls are rare in here for sure) kudos. I just can't find time computer for writeups.

I wouldn't change a thing. plus you gave the data, although i'm not sure it's needed. Pictures are great. Love the Z.
when i did a cable driven core xy, i swore off that style. it was cool but slippage killed it.

I have that same 12x12 line level pad and I'm worried about my 1/4" mic6 plate warping because it's like 20x17 or something. I haven't cut it down yet.
I think the center hot spot will make the plate a bowl shape. I'll do your ssr power safety idea, but since there is a thermistor embedded into the orange matting, I wonder if it can fail.
i'm also thinking of just doing chamber heating and use normalization time to get it all up to 60C or so. my results in the past with ABS in the 40's had issues, and I think more temp is better.

i starting to getting my head back into 3d printing, back from routing. I think i've said that twice in the last year........

Data is hard for intelligent people to argue with (though you may not get that impression from politicians, these days, but intelligence was never a qualification for office in the US, and lately seems to be the only disqualification). Besides, I'm always curious and measuring things and examining them under a microscope scratches an itch.

In my printer's bed the temperature difference between the edge and the center is so small that the warp of the plate will be inconsequential. If you put a 300x300 mm heater in the middle of a 20" x 17" plate, all bets are off. Having the thermistor in the heater itself protects the heater from burning itself up (as long as you get it stuck to the plate without trapping any big air bubbles).

Edited 1 time(s). Last edit at 09/28/2017 11:01AM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Quote
davew_tx
You still haven't said what you want this machine to make.
Can you share the end part needs?

We usually do large architectural models and they are usually of very high detail and it isn't easy to print them in parts, we had to deal with such hassle when the architectural stuff were the worst fears...
lately, we were asked to do some prototyping for house appliances (Ovens, refrigerators and etc...) it would be crazy to make such large models in parts (otherwise I will have to make 12 prints for example to build a product)
That's why we need something to serve both purposes ...

I'd suggest that Ultem be removed from your needs.

You can move quickly to a large format printer.
80/20, linear bearings, belt driven with steppers, get a real extruder (volcano is my choice).
You can have break away same material support. one head is plenty to take on for a first try. You can add a dual head later.

I made my first printer to make vacuum cleaner prototype parts (Soniclean was designed by me and an ID guy!).
I made a few parts, but the big ones needed Stratasys quality/accuracy.
We spent $1000 each for the top cover and base part.
Made a ton of parts on my Uprint, glued together worked great.
Made a few on a machine I made.

you can definitely do this!

you can also buy a large printer.
there are a few , but I can't remember them at the moment.


you can budget $10k and get started making protos, selling protos, and focus on what makes you money.

brute force I can make a machine first: sometimes becomes a time-sink rabbit hole that keeps going down.

search on 3ders.org. that site highlights all the printers over the years.

They posted a article on my kickstarter too! x-truder.......... fail.......

and consider a cnc router for large panel manufacturing.
you can use modeling foam, cut your shapes, there are coatings to make them smooth, colored, etc.

Edited 1 time(s). Last edit at 09/28/2017 11:31AM by davew_tx.

Quote
davew_tx
and consider a cnc router for large panel manufacturing.
you can use modeling foam, cut your shapes, there are coatings to make them smooth, colored, etc.

LOVED the Idea!

GIGABOT

[re3d.org]

ErectorBot EB 2076 LX

[www.erectorbot.com]


BigRep ONE

[bigrep.com]

3D PRINTING FOR ARCHITECTS

[www.lpfrg.com]