Does (bed) size matter?

Hello everyone, new guy here. Currently going through the process of "designing" my 3D printer. Or... rather still deciding on what I want from it...

Anyway, I would like to hear some opinions on what bed size / build volume do you find most useful?

I realize this is ultimately my decision to make, and I should know what I need... but honestly, does any of you know what you want?
Regarding the usage of the printer - I don't plan on doing actual "work" on it, I see this more like a learning experience, and I appreciate the challenge of building one myself.
While I would love to print some 1:1 scale props from video games and such, I am under the impression that I could just join them up from different pieces.

Now the reason why I am asking this, is because if I end up not using 75% of the bed for most of my prints, it will be just wasted space, wasted money, and wasted speed (which I could have gained by using lighter materials)...
But then again, if I ever need more space than I have... I'll just feel terrible for not making everything bigger...
My current design revolves around a CoreXY system, using supported linear rods for Y, 2 regular 12mm smooth rods for X, and a belt driven triple leadscrew Z. Using 500mm rods, I would get about 400x400x400mm of build volume. But the more I think about it, I start to realize this will add a bit of mass to the whole moving assembly and might make everything quite a bit slower than I would like. Besides, I'm not even sure 12mm is enough (or too much) for this kind of distance...
Looking at commercial printers, I often see build volumes of 200-250mm, which seems to be the standart... but that might be just to cut down on costs with no additional benefits... so THAT's why I'm asking for help

Attached is a photo of a quick mock-up of my current design, and a SketchUp file which I try to keep close to the real dimensions...

12mm rods will be pretty heavy, and with the linear rod bearings you will get a lot of ringing. At those sizes above 200mm, there's more engineering required to actually scale a printer up. about 8" cubed, its pretty simple, but above that you have longer canttilevers and weights that you must deal with.

12" cubed is pretty good, any larger is extra bonus. My printer is 14" diameter and 18" tall in build volume which is just about right for me. Also at those sizes, unless you want to do some more fluid flow simulations, you will be stuck with low Tg materials such as PLA. You won't be able to print PC or ABS at 400mm^3 without a chamber.

Then factoring in building a chamber, whats the extra cost to you going to be? The extra engineering, the requirements of hotend to not clog under those temperatures, the parts you've purchased (can it handle the heat?). How much belt stretch will you get at elevated temperatures because in your current design they are exposed to the elements and if so the heat as well. Is weight a concern? etc.

That's quite similar to what I've been thinking recently - building a smaller printer is just easier...

Quote
3DRapidClone
My printer is 14" diameter and 18" tall in build volume which is just about right for me.

I take it you have a delta printer then. These seem to be easier to scale, right?
About the build volume... May I ask how often do you need to print something close to 12"? (and what kind of things are these big prints if that's not a secret?)

Quote
3DRapidClone
12mm rods will be pretty heavy, and with the linear rod bearings you will get a lot of ringing.

I'm not too concerned about noise, as I will be using the printer in my workshop (detached garage), but the weight might be an issue, though sagging under load might be a bigger one

As far as heated chamber goes, I've always thought I'll be encasing the printer with something, but never actually considered what impact that might have on the components... The hotend would probably need water cooling, or at least a flow of fresh air from the outside, and the belts... well, I have no idea how would they react...

For a delta with heated chamber, I'd put more consideration into the electronics and steppers than to the belt.
I'd move electronics and steppers on top of the frame and keep them away from the heated bed and outside the chamber.
The printer gets top heavy then, so I'd use double horizontal extrusions on both ends and fix the bottom to a wooden plate.
-Olaf

Based on my experience with a few different machines, I suspect the vast majority of hobby grade 3D printers out there can't reliably print over more than just the center few square inches of the bed, regardless of actual bed size. The hordes of hobby printer kit buyers seem unwilling to spend more than $300 for a printer, so what else can you expect? A bigger bed is a powerful draw, but people won't pay for one done right, so they get crap because the kit maker wants to sell as many kits as possible.

If you want to be able to use the entire bed, no matter its size, you need something like cast aluminum tooling plate, a properly designed support/leveling system, and an adequately powered heater.

If you build a cubic frame, the enclosure is almost free- all you need to do is cut some insulation panels to fit the sides and stick them on with printed parts, magnets, or even duct tape. A plastic garbage bag slipped over the top of the machine will do in a pinch. A delta is almost as easy to enclose as a cubic frame cartesian machine. When you are designing the machine, design it so that the electronics is outside the print chamber.

I print ABS in an enclosure every day with the temperature at 45-50C, and no warp or delamination. At that temperature there are no special requirements on the hot-end- just leave its fan on all the time. If you decide to print PLA, leave the enclosure door open. In my machine, the 450W bed heater keeps the rather large enclosure toasty without any extra heat input.

I was using 1/2" end supported guide rails in my 12 x 12 x 11 build volume machine and ultimately swapped them out for fully supported linear guides. The machine is so stable now that I recently adjusted the level/zero for the first time in 6 months. If I were designing another printer today, I would not consider using anything else. I bought mine via ebay- you can get some real bargains if you are willing to scour the 10+ pages of listings that come up when you search for "linear guides". I recently picked up an unused, 21" long, IKO linear guide with a bearing block for $34 shipped. That part will be a perfect X axis in a cartesion machine. The first time you handle a linear guide you will fall in love. They are so perfect you will wonder why anyone would use anything else. 3D printers are essentially zero load on linear guides (my recent purchase is rated for 3700N static load) so they will last forever in a printer. Don't worry about what size to buy- any size is fine (as long as it is long enough) because they are all spec'd so far beyond what they need to do in a 3D printer that even small ones work fine. You can cut them to needed length with a cut off wheel on a grinder.

---

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

It's not the size that matters, but how you use it. (Sorry. Couldn't let that setup go.)

Since this is your first printer, I would suggest keeping the bed size small to start with. Get a feel for how it all works before trying something larger. As @3DRapidClone mentions, as the build gets bigger, there are more engineering challenges that have to be addressed. Keep the bed under 200mm x 200mm for your first build. That is the bed size for my modified Printrbot Simple Metal and I find there is very little I want to print that I can't with that size.

Quote
Dalius98
That's quite similar to what I've been thinking recently - building a smaller printer is just easier...

They are much easier to build.

Quote
3DRapidClone
My printer is 14" diameter and 18" tall in build volume which is just about right for me.

I take it you have a delta printer then. These seem to be easier to scale, right?
About the build volume... May I ask how often do you need to print something close to 12"? (and what kind of things are these big prints if that's not a secret?)

It is in fact a delta printer. They are easier to scale in the Z as you only have to replace the 3 linear rails and belts. For X and Y you need to replace 9 extrusions and the 6 arms.

For a cartesian, you need to replace 4 vertical extrusions for Z, use a longer leadscrew and replace at least 2 longer linear rods.
For X and Y, you need to replace the belts, and 4 linear rods and 8 extrusions, then you must go to a triple leadscrew for any decent size as the cantilever style bed will not work, so 2 more lead screws, and bottom belt which has to be a closed loop belt and 3 pulleys, with a bearing tensioner below as well.


Also, I print a lot of things in batches, so the space is really nice and I print 1:1 scale cosplay items. You'd be surprised how much money they can go for especially for cosplayers for a very intricate pieces.

Quote
3DRapidClone
12mm rods will be pretty heavy, and with the linear rod bearings you will get a lot of ringing.

I'm not too concerned about noise, as I will be using the printer in my workshop (detached garage), but the weight might be an issue, though sagging under load might be a bigger one

As far as heated chamber goes, I've always thought I'll be encasing the printer with something, but never actually considered what impact that might have on the components... The hotend would probably need water cooling, or at least a flow of fresh air from the outside, and the belts... well, I have no idea how would they react...

Ringing doesn't refer to noise, but rather the memory that polymeric materials have where your machines vibrations and momentum due to high acceleration leave ringing ripples in your print, mostly when you print holes in the Z direction.

Quote
3DRapidClone
Ringing doesn't refer to noise, but rather the memory that polymeric materials have where your machines vibrations and momentum due to high acceleration leave ringing ripples in your print, mostly when you print holes in the Z direction.

ah, I remember seeing this problem discussed before, didn't realize it's called ringing... Unstable or wobbling Z axis on a CoreXY would cause that too, right?
If I keep the bed size and X carriage weight to a minimum I should be fine with smooth rods and linear bearings?

Quote
the_digital_dentist
I was using 1/2" end supported guide rails in my 12 x 12 x 11 build volume machine and ultimately swapped them out for fully supported linear guides. The machine is so stable now that I recently adjusted the level/zero for the first time in 6 months. If I were designing another printer today, I would not consider using anything else.

A "regular" cartesian, or a corexy cartesian?
Well, I suppose I could invest a bit more into a real linear rail system like Hiwin and it's clones, if I decide to go big. Especially since they are not THAT much more expensive than quality rods... Also, these ones are better than the Chinese knockoffs since they are using 4 rows of ball bearings like the real Hiwin does... I'm just a bit worried if one linear carriage has enough support to not twist under load, and if I had to use two per row... that would almost double the cost..

The one thing I want to avoid is having to buy everything all over again when I decide to upgrade for some reason. Having two linear slides for Y and one for X, would also allow to re-use them in a Delta style printer if I decide to go that way (though they might be a bit too short for that)... but I just hear it's a neverending journey of tinkering with it, and when something goes wrong you have no idea why... Though the posts stating that were at least 2 years old - did something change here?

The ultimate goal of course would bit to use a double extruder for support printing, but that's certainly not for a first time build... and I hear it wouldn't really work out with a delta or coreXY since it adds more weight...

Quote
ElmoC
It's not the size that matters, but how you use it.

See? this guy has the right idea



Oh, and by the way... how powerful motor do you actually need to power a direct(or geared) extruder? I've just disassembled a desktop laser printer and it had a nice little stepper motor (can't find any specs, but it looks like one of those permanent magnet types)

An unstable Z would cause your layers to overlap, and quality to drop and could clog your nozzle. Smaller the better for less ringing.

As for the delta stuff, autocalibration is a big thing now and as for tinkerering, delta kinematics is not as easy as Cartesian and I find that statement about not knowing what's wrong incredibly false, but most of my experience is,with deltas.

You can do dual extruder with either, Bowden is recommended. I run 4 extruders on my delta.
Extruder should be as torquey as possible. I wouldn't use the one from an inkjet

My machine is a "regular" cartesian- see the link in my sig, below.

I used a ball screw for the Y axis and two linear guides, but I think that a single linear guide with two bearing blocks could work fine for the Y axis since there are no rolling forces applied to the bed. I can't think of any way to describe the single linear guide and bearing block in the X axis except to say they are perfect. It is compact, easy to mount, doesn't sag, free of unwanted motion, and there's no need to try to keep two rails parallel.

I would not worry about whether they can handle the loads presented by a 3D printer- linear guides are usually rated for huge loads compared to those presented by a 3D printer. My recently purchased IKO linear guide and bearing block ($34 shipped via ebay) are rated for 3700 N static load (>800 lbf!). The linear guide will be working like new long after the machine has become a museum piece.

Edited 1 time(s). Last edit at 11/04/2015 03:57PM by the_digital_dentist.

---

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

Quote
the_digital_dentist
My machine is a "regular" cartesian- see the link in my sig, below.

Ah, stupid me, somehow didn't see that... tried pretty hard to find it by looking at your recent posts
Your build was one of the first few I saw on the instructables when researching. Super nice!

Anyway, big thanks to everyone for helping me out and answering the questions that have been answered a million times already (probably). I just feel a lot better hearing everything firsthand.
Honestly, it's probably the first time a newbie doesn't get bashed for asking stupid questions. Guess there's only creative people here on reprap forums

Now I guess I'll be off to think whether I need the massive build volume and the troubles that come with it, or not... (and the usual CoreXY vs Delta debate)

Quote
ElmoC
Have you looked at the designs on openbuilds.com?

yes I did, I also looked at some designs over on thingiverse too, one in particular was something close to what I had in mind, but honestly, most of them seem to be as much of a "shot in the dark" as my current "design". Pretty much everyone says they are happy with the print quality, or they just need to do some calibrating and it will be ok... At least doing something my way, I'll have to live with my mistakes, not someone else's
Not saying they are all bad of course, they work, while mine doesn't even exist, and there are a lot of creative solutions to some problems which might take forever to figure out on my own... When a person says he has owned (or built) more than a few printers, and that some things didn't work out, why they didn't work out, and how to overcome them - I respect that, and I am pretty confident in in replicating that... but the most builds I've seen have no reasoning behind them, or even worse - the reason is "it was cheaper"...

Edited 2 time(s). Last edit at 11/04/2015 05:09PM by Dalius98.

Have you looked at the designs on openbuilds.com?