So I went ahead and ordered most of the parts I need. I was wondering though -- is there any reason springs are used for bed leveling, as opposed to Z probing?

 

I was considering using a proximity probe to get the benefit of the G33 bump mapping in the repetier firmware. I use it with my kossel and it's really nice. I haven't used a glass surface before though -- perhaps it's uniform enough that Z probing doesn't provide any benefit?

 

I run a inductive senors on my C-Bot and my i3. Even with that, I still run springs so I can level manually and then use the Z-Probe to catch any errors or changes in the leveling. @TechGirl runs a very interesting z probe setup you might want to check out as well. Another plus of the springs in the case of a head crash. I know that "never" happens, but when it does, it's nice to know there is a little give to save on parts.

 

You mentioned you have a kossel and that makes a lot more sense for that type than this type. I have used auto bed leveling on my cartesian before and it is really a mixed bag. @Carl Feniak is right, you really don't want your z-axis moving that much. If you use a nice sheet of Pyrex (Borosilicate Glass) it will be thick enough not to bend (~4mm), and won't warp with the heat. Plus the process they use to make it virtually guarantees it cannot have "bumps" or be out of level. It is what is known as "float glass" that is the liquid glass is floated on top of another liquid (typically something really heavy like a metal) and allowed to settle and cool. As you can imagine this creates a super flat, super level surface.

Either way I do not recommend an inductive sensor if you are going to do auto bed leveling, use something like a BLTouch or the servo method, they are far more repeatable. The inductive sensors are designed to detect iron not aluminum and so they need to be much closer to the bed to work, they can vary from time to time as many environment factors can affect them, capacitive sensors are even worse. The best, easiest, most reliable sensor is always going to be a physical one. If you DO decide to go with an inductive sensor make sure that you get one that is rated to 8mm as the 4mm ones will not be able to detect the aluminum bed through pretty much any amount of glass, they need to be within about 1.8mm of an aluminum bed to work.

My .02.

 

AHHHHHHH!!!!! The last of my shipped parts arrived today, just two more to print. I am running out of excuses not to start cutting up this extrusion..........

 

@Carl Feniak
My girlfriend is on the board of directors for STEAM Fest - Maker Boulder and is managing the "STEAM" space for Denver ComicCon this year. I have considered taking my build to one or both, if anyone else is in this area, or is interested in being there please let me know as it is an excellent opportunity to meet other makers and impress some kids. 3D printers are always some of the favorite things there, that and face painting.

 

Those are legitimate criticisms of auto-leveling for this type of printer. I'll spend the money on some good borosilicate glass instead.

Thanks!

 

I've never used auto-leveling on the 3 printers I use (rep1, rep2x, cbot),... never really felt the need to: Once it's dialed in it lasts well for a long time, plus I print with a brim, so I watch that lay down on the first layer and dial it in manually as it goes. That + glass + some good woodglue\water slurry (& hairspray to top it off between prints), and stuff sticks a-ok.

I've read that when you auto-level, it has to store those offsets in memory throughout the print, applying them to all subsequent commands. And this can strain things when stuff starts to speed up when using 8-bit hardware. Hearsay & rumors, but it makes sense to me.

 

@Carl Feniak before I start cutting, my heated bed is 12" x 12" (305x305). Should I enter 320 or 305 to get the correct measurements?

 

12" on the dot

 

was just thinking the EXACT same thing

 

Looking at the pics, it seems like the width between the Z extrusions is somewhat configurable without having to re-cut. Is this the case? And if so, is there any problem to cutting a little larger than you need (e.g. sized for 320 rather than 304) other than some wasted space?

 

The problem will be getting the mounting brackets for the heated bed to line up with the holes in your heated bed. This doesn't matter if you are using the rear, or offset mount locations for the lead screws, but it does matter if you are using the front/rear mount due to the addition of the front-cross bar.

EDIT:
For clarity you are correct, you can move the z-extrusions without having to re-cut. However if you are using the front/rear mounting location then you WILL have to re-cut that piece; otherwise you are correct.

 

I think we should add a few things to the calculator after playing with this a bit:

* We need two measurements for bed size, hole to hole, and edge to edge. Because we also need....
* The ability too specify the amount of "padding" you want on either side of the bed (defaulted to the ~80) that you have in there now.
* An option to specify how much lead screw "slack" you want in the top, and some notes about how the expected hotted length is 62mm, etc..

 

No worries! There is really nothing to be nervous about.

The idea is that mounting the screws in front and back helps support the bed in the front, making it not fully cantilevered, keeping it level and lifting it more evenly. This is also the idea behind using the offset screws, by lifting the bed from closer to the center of gravity you are reducing the axial load on the wheels which in theory just makes everything easier. Does this mean that mounting the screws in the back a bad idea? No, probably not, maybe, yes. Depends on what you are doing. Someone on here was talking about using something super heavy for a build plate, in that case the location of your screws could be important. In general the more weight you plan on support on the build plate, the better an idea it is to lift it from a non-cantilevered position. So if you are building it really tall (taller items can be heavier!), are using a really heavy build plate, are planning on using this as a service elevator when you are not printing on it, or something of that ilk then you should be taking that into consideration when you decide where to put your screws (if you are using it as an elevator you may also want to redesign it slightly so that the z-platform has wheels/is supported on both sides and use 1 start lead screws.). Hey good news is that you can move them super easy! Just have some extra extrusion lying around and if you decide you want to move them cut it to size, and pop it in!

As far as the slaving thing goes.......the short answer is yes, you can run two motors off of one driver in either serial or parallel. Serial being where you connect the two motors inline so that the pulses flow through one motor and into the other, Parallel being where the wires are "spliced" together. Either way will work, there are pros and cons to both ways. You can also run two motors off two steppers with certain versions of certain firmware.

The long answer is that actually "slaving" the motors together 100% is harder than it sounds. Even running two motors off the same stepper you end up with them getting slightly out of sync every now and then (sometimes running them on two steppers is better, sometimes worse). It helps a lot if your motors are exactly the same, and in practical terms its not a HUGE deal because it just takes a small adjustment to get them back in sync again. There are lots of ways to solve this problem. Many people just run one motor and then run the other lead-screw with a belt, Some people are running three lead screws off one motor! Carl thinks you could probably physically slave two motors together with a belt and keep them in sync that way. I am looking into the idea of using a dual shaft stepper motor connected to 90 degree gearboxes; but I haven't found all the parts I need in the size I need them just yet.


What motherboard board are you going with?

 

Thanks for the detailed reply! I'm planning on using the offset mount option, so it sounds like I'll be alright as far as that goes.. The dual shaft stepper sounds interesting! The 3 lead screw option sounds particularly interesting as well.

I just ordered a smoothieboard knockoff and I'm debating whether to use it for this build or to put it in my kossel and use my RAMPS board (with DRV8825 stepper drivers) for the C-Bot. I'm leaning toward putting the new board in the kossel at least at first, since I won't be finished with the C-Bot for at least a couple months, and it sounds like people have been having a good deal of success with the RAMPS option.

I am a bit curious on the effector options.. I notice there's a direct drive extruder option in the STLs. Any idea how well it works? If the speed/stability loss is minimal I'd rather go with the direct drive since I already have a printer that suffers from the bowden retraction problems.. But if the downsides are significant I'd rather just use a bowden extruder to get the speed and accuracy that core xy printers are so famous for.

Edit: here is the specific board I ordered... Hopefully it's not a scam, seemed too good to be true.
3D Printer Board MKS SBASE V1.2 32 BIT ARM Smoothieware + USB,1GB SD,XH2.54,TJC3

 

I doubt that is a scam, but I guess you never know lol.

I have two boards. I have an Azteeg X3 Pro that I plan on using for now, I also have a 32 bit DUET board which I plan on playing with.

Honestly I dont know how well the different extruders work, but based on what I know about this design the XY gantry should be more than capable of throwing around that motor without much effort. I am going to probably start direct too since I also already have a bowden setup and am interested to play with both.

Ultimately I am waiting on my Flex3Drive to show up which looks to be the best of both worlds if it lives up to the hype (and from what I hear...it does).

I will keep you updated on my dual driveshaft idea, its getting close. I just ordered the stepper, and I have decided to just build my own gear boxes using miter gears, all in all I am actually fairly confident it will work and probably won't be terribly expensive to implement.

 

Nice! That Flex3Drive looks a little too pricy for me right now, but I'd be interested to hear how well it works.

Are you going to print your gears?

 

Yes, at least for the prototype. I figure I can print them in Taulman 910 and they should be **** near production quality. I am going to use the CAD models off McMaster tho so that I can replace them with metal ones easily.

 

So it looks like 300x300 borosilicate glass is pretty expensive (at least in the US). Any good alternatives? I'm afraid of using regular glass -- I have small children that I can't completely keep out of my work area, and shattered glass is not something I want to risk. I'm thinking maybe just going down to home depot and getting a 12" granite tile..

 

Excuse me if I am retreading on a dead idea, have not kept up with the thread. I would suggest looking at using an Aluminum plate. The downside is that it's not cheap either. My 12" slab was roughly $40 from McMaster. That said, it's still cheaper than replacing boro glass every few months.

 

Agreed. Just make sure it is MIC-6 or similar tool plate to get the required flatness. I got my 300x200mm plate from www.midweststeelsupply.com and it was around $40.

 

Anyone using a three point mount for the build plate?

If so, how well does it work? How did you connect the mid point mount to the Z chassis?

 

This is how I did mine. There are 90 degree corner brackets on the back side to hold it in place.

 

Hi all,

Firstly, thanks for a fantastic resource, inspired me to start self-building after a self-assembled Chinese printer wet my appetite.

So half way through building a modified c-bot. Trying to get the best of all the versions which is a bit trial and error. Am building in Singapore so nil stores have any parts, everything ordered online except bolts, which are plentiful. Built most of the printer using local 20x40, with v-slot for just the roller parts. Works well except you need to edit the files to reduce the size of the insert that fits the groove by 1mm each side. Or you can scrap it off excess to fit. Cost looking like around US$750 for a size 20x30 bed printer.

A few lingering questions wanted to get some feedback on.

1) the flexible coupling from Aliexpress seems crazily flexible. i can compress it over 1mm just with my vernier. is this normal? seems inconsistent with getting the bed level within 0.1mm? should i order a plum/spider coupling?

2) Am thinking of putting the Z axis screws diagonally opposite each other ie front right and back left. This way should prevent both left/right and front/back wobble/tilt. Anyone tried this already or got any suggestion whether it will work or not?

3) do the printed 7mm and 25mm spacers work, or is it wise to use alu spacers?

4) do you need the special 8mm diameter spacers on each side of the delrin wheels as well as the inside?

5) has anyone successfully linked Z-axes screws with 2 stepper motors using a belt. Does it work or do the stepper motors fight each other?

Thanks