FB2020 CoreXY

Ok, I think it's about time I created my own thread for questions etc.

So, the FB2020, it was started as a Fusebox derivative, however, it is now to the point where there is very little in the way of Fusebox Parts left. The FB2020 uses a 2020 Extrusion frame opposed to the 1515 that the Fusebox does. Hence FB2020.

I started out taking Culain's 2020 Version and working from there, my mail goal was to get rid of all the damn cable ties. Cable tied linear bearings are like Gaffa taping your exhaust to your car. The Carriage was the first piece to go and re-designed from the ground up to be under-slung opposed to being mounted through the carriage. This would allow for a better layer fan and even cooling. The Carriage has gone through so many different iterations, I've lost count and I'm still not happy with it. The Y Carriers needed re-modelling due to alignment issues and it snowballed from there. About the only parts left that have not been modified or re-designed are the Corner brackets and motor mounts. Everything else has been tweaked, modified or re-modelled. I've borrowed a few designs from elsewhere, such as the layer fan and extruder, which, have been modified to work on the FB2020.

The Ethos of the Fusebox was to keep parts pricing low, which at the beginning of this project, was the same aim, however, it is now to improve the FB2020 to a point where it's pretty much it's fool proof and highly reliable. I want it to the point where anyone with a set of Allen keys can put it together. I'm happy with the frame rigidity, yes, it's not as rigid as it could be, but I still want to try to keep costs down where possible, by using printed plastic parts this helps a little.

There are currently two official versions of the FB2020. The Standard, with a build volume of 200x200x170 and the Plus with a Volume of 200x200x275. Further info, files and images can be found on the Github at https://github.com/AxMod3DPrint/FB2020

As an added note im also trying to support and work on converting the models to Solidworks files
Looks like a good model worth backing.

Quote
Floyd
As an added note im also trying to support and work on converting the models to Solidworks files
Looks like a good model worth backing.

Thanks Floyd.

The Printer is constantly being worked on at the moment. I'm trying to push one updated part a night to the Github repo. Tonight's effort is the Z Motor holder. It's had a lot of meat added to it. All bolt lugs have been enlarged to allow for easier use of the recommended Flanged Button Head Hex Bolts (image below). I recommend the use of this type of bolt as it saves having to get washers to relieve pressure on the parts around the bolt holes and prevent them cracking or splitting. This mount shouldn't have much in the way of give compared to the original.

My Main focus at the moment is the Z axis and removing any possible Z wobble that the Cantilever bed could have. The Bed Platform is solid, so it's just moving the Z Rods to 12mm, and making sure all mounts have no give or play, and are easy to fit. The 12mm Conversion is in progress, The Bed carriers and rod holders have been re-designed to use LM12UU's and 12mm rods, they have been print tested and work fine. I'm just waiting on parts atm.

I'll be working on the Build guide this weekend as I have someone who is going to build test it from a point of knowing nothing about 3D Printers. That's about it for the mo, I'll leave you with a few model images:

Bed Holder - 12mm Version



Z Motor Holder

Ill have to figure out that Github thing. Im working out of my drop box at the moment for this project in SW.

Quote
Ax
Ok, I think it's about time I created my own thread for questions etc.

great - look forward to following along, seeing how it goes, ax.

Here is a little peek at the Solidworks progress.
Im having to guess at how some of this goes together because the pictures and all that I can see have different parts on the assembly.
Getting there though.

Quote
Floyd
Here is a little peek at the Solidworks progress.
Im having to guess at how some of this goes together because the pictures and all that I can see have different parts on the assembly.
Getting there though.

Just added an IGES file of the full model of the Plus. Should give you an idea of where parts go. The Plastic bed brace is no longer used due to 3 point levelling needing an extra 147mm extrusion, I've just removed that model now, something I'd forgotten to do.

Edited 1 time(s). Last edit at 05/18/2016 05:17AM by Ax.

Perfect. Thanks for the input. The full IGES will be a huge help figuring out the rest of the placement.
It was hard picking through the tons of pieces in the github to find the latest greatest parts.

No worries. The Z motor mount has been beefed up recently along with the Z endstop.

::Edit::

I'm also going to removing the 8mm Rods from the Z in favour of the 12mm over the next couple of days. The Bearings fit nicely and the rod mounts are solid, I just need to print the production parts and fit them.

Edited 1 time(s). Last edit at 05/18/2016 09:25AM by Ax.

I think the only parts I have left are the carriage. Other than that its all just about done..
Just fitting them into the assembly and making sure everything lines up perfect.

After a conversation with someone I've put a kit together for, I'm going to be working on a redesign of the Carriage and Carriers. The reason for this is that the Belts are a bitch to tighten and the hot end is a pain to remove. Don't get me wrong, they work well, but not as well as I would like.

I'm looking at the idea of flipping the carriage to a vertical one with the rods coming in on the top and bottom of the Y Carriers, which will be in two pieces with the bearings in the middle. The hot end assembly will be secured with 3 bolts to the carriage to allow quick removal. For Belt tensioning, I'm thinking about a sliding clamp that will allow tensioning of both belts at the same time.

I'll keep things updated here when I get it a little more fleshed out.

Sure thing. Keep me posted and I can keep the SW counterpart up to date also.

Quote
Ax
I'm looking at the idea of flipping the carriage to a vertical one

don't do it! seriously, don't do it. i did the same thing, because it was a way to save space. photo:



what has happened is the following (despite using 45mm LM8LUU bearings):

• the y-rod bearings move slightly on the rods - only by about 0.25mm - but the total means that they can actually move, if rotated, by 0.25mm up at one end and 0.25mm down at the other, for a total of 0.5mm.
• the hotend (for obvious reasons) is *not* centred at the exact dead-centre (i.e. dead-level with) those y-rod bearings: it is, in fact, some 50mm *below* the Y-rods.
• the x-rods have a small amount of flex in them (yes this is 8mm diameter rods, but they still have a bit of flex).
• also the x-rods, because they are vertical, and those bearings *also* happen to move a tiny bit, contribute to *additional* rotation of the carriage
• again (for obvious reasons) the hot-end is not centred directly between the middle of 2 vertically-stacked rods.
• so due to the movement of the y-rod bearings, and because the hotend nozzle is offset from the centre of those bearings, rotational movement of the y-rod bearings translates DIRECTLY into rotation of the hotend
• likewise due to the movement (flex) of the x rods and also the bearings *on* the x rods, the hotend nozzle also being offset from the centre of the bearings, you get EVEN MORE rotational movement of the hotend.

the combination of all of these factors means that there is a whopping ONE MILLIMETRE of movement of the hotend when you grab it and *GENTLY* move it backwards and forwards. that's.... that's a staggering amount.

so it has to be said that this has been an extremely costly experiment, both in time and money, to create a vertically-stacked CoreXY 3D printer arrangement. given that i cannot fit in a horizontally-stacked arrangement because of the compact space available, i will *not* be going with a corexy design, i will be converting to that "unnamed" arrangement which was used in a plotter from the 1990s. the diagrams and discussion are here: [forums.reprap.org]

it will be a drop-in replacement for the corexy arrangement that i currently have, with the motors and idlers being in the same places - but bear in mind that i have already moved the motors in-board by 20mm and have idlers sitting on the motor parts already.

Quote

with the rods coming in on the top and bottom of the Y Carriers, which will be in two pieces with the bearings in the middle.

yeah i put the bearings at the top and the bottom,above and below the y-rods. this allowed me to reduce the width to 312mm (6mm clearance either side for the belts, whilst the rods are 300mm). however, that's immaterial if the total rotation due to linear bearings moving around results in a whopping 1mm of hotend variation!

Quote


The hot end assembly will be secured with 3 bolts to the carriage to allow quick removal.

yeah i thought of leaving the sandwich200v2 hotend just secured with 3 bolts - can i suggest having rails as well so that it slides in-and-out?



you can actually see there are 5 (potential) screw-holes. there's a row of 3 on the left where you can just drop 3 bolts in to secure the Flex3Drive (black) to the carriage (purple), and then you can see another 2 at the front, which go through that "rail" (front of purple part) into the blue part (hotend holder on the Flex3Drive). there's a second rail, obviously, on the other side, which the blue part slots into.

amazingly it turns out that i don't actually need to put the screws in at all: friction of the rails and on the end-slot is enough to hold the Flex3Drive in place.

Quote

For Belt tensioning, I'm thinking about a sliding clamp that will allow tensioning of both belts at the same time.

do take a look at what... mmm... who is it.... the_digital_dentist did... he came up with something pretty cool. let me find a link... ah here y'go: [forums.reprap.org] - lots of advantages to what he designed, Ax. if you're looking to do a clamp, don't look up how nophead did his belt-tensioning for the mendel90 x belt. it's a D-shaped piece that has a screw to move it, the belt's wrapped around that. nice thing is, as it's a pulley system it does a 2:1 downgearing of the amount of turns you put on the M3 screw. also you could try what i did, which i outlined on that same thread as dd posted, but i have to say it works but is a leeetle fiddly

ok good luck gotta go.

Edited 1 time(s). Last edit at 05/18/2016 02:28PM by lkcl.

Updated SW with carriage now in there.

Quote
Floyd
Updated SW with carriage now in there.

Looking good. Going by the look of that, it looks like the Plus version?

Also I have the final 12mm Z axis parts on the Printer now. I'll be stripping it down later to add in the upgrade.. Images to follow.

Edited 1 time(s). Last edit at 05/19/2016 04:17PM by Ax.

Yes I went ahead and made the model the plus.
From there I can do whatever needs to be done

Right, I have the 12mm Rods on the Z in, and I'm just running off another set of the parts as my tolerances were way out for printing in PETG, so much so, that the bearings slip out too easily, be aware that if you do print these parts they are designed for printing in PETG, which has very little to no shrinkage, so the there is no tolerance added to the part. The BOM will be updated shortly.

As promised, here are some images:

Quick update. I've decided to start hiding electronics and improve the active cooling of the the stepper drivers. For that I've created an enclosure for the RAMPS which replaces the simple mount.





The above images are of an outdated version which is currently on the printer, the new one is printing. The newer version has a reset button and extra height to allow for fitting of the LCD header, I derped on both cases. Cooling wise, It uses a single 40mm fan mounted inside the enclosure, which by my unscientific test of putting a finger in the enclosure feels like it's cooling things a lot better.

I'm planning on recommending the addition of a fan header to the RAMPS as well to allow for auto fan shut downs for the 30mm fan on the E3D and the 40mm fan on RAMPS box, this will require the RAMPS enclosure to be widened slightly, I'll do that once I get the header in.

Quote
Ax
Quick update. I've decided to start hiding electronics and improve the active cooling of the the stepper drivers. For that I've created an enclosure for the RAMPS which replaces the simple mount.

ax... please consider saving yourself a lot of hassle and get a Duet NG when it comes out in a few weeks. i explain a number of times why RAMPS is completely shit - it's the idea of blatantly ignoring application notes and putting stepper ICs which have a ceramic insulator on the top (so heatsinks are ineffective) that are *specifically* designed to soak heat through the exposed pad onto a large PCB... onto 16x16mm tiny PCBs and then expecting there not to be problems, _that's_ why you're encountering difficulties.

the people who started the home-made reprap movement were good at ideas but not very good engineers. they also ignored things like "momentum" when first developing the firmware, and that area's slowly been fixed, but there are some areas like "modular stepper boards" which are still not getting through people's skulls that the stepper ICs were designed with very specific heat dissipation requirements that are in no way met by these "modular" boards.

you've done an amazing job with the FB2020 - the only weakpoint is continuing to use a board that, unfortunately because of their original popularity, continues to propagate over 5 years worth of design ignorance (root word of ignorance: to IGNORE).

sorry, i didn't realise you were using RAMPS otherwise i would have mentioned this a while ago.

Quote
lkcl

Quote
Ax
Quick update. I've decided to start hiding electronics and improve the active cooling of the the stepper drivers. For that I've created an enclosure for the RAMPS which replaces the simple mount.

ax... please consider saving yourself a lot of hassle and get a Duet NG when it comes out in a few weeks. i explain a number of times why RAMPS is completely shit - it's the idea of blatantly ignoring application notes and putting stepper ICs which have a ceramic insulator on the top (so heatsinks are ineffective) that are *specifically* designed to soak heat through the exposed pad onto a large PCB... onto 16x16mm tiny PCBs and then expecting there not to be problems, _that's_ why you're encountering difficulties.

the people who started the home-made reprap movement were good at ideas but not very good engineers. they also ignored things like "momentum" when first developing the firmware, and that area's slowly been fixed, but there are some areas like "modular stepper boards" which are still not getting through people's skulls that the stepper ICs were designed with very specific heat dissipation requirements that are in no way met by these "modular" boards.

you've done an amazing job with the FB2020 - the only weakpoint is continuing to use a board that, unfortunately because of their original popularity, continues to propagate over 5 years worth of design ignorance (root word of ignorance: to IGNORE).

sorry, i didn't realise you were using RAMPS otherwise i would have mentioned this a while ago.

I'm fully aware that the RAMPS isn't optimal and I'm fully aware of where the printer falls short. For now, the RAMPS running Marlin 1.1-RC6 works well. CoreXY is fully supported and has all the options in there that are needed, so it does. The plan has always been to move to a Smoothieware based board eventually, something along the lines of the Azteeg Mini, AZSMZ or MKS SBASE. That'll happen when the electronics get moved to 24v.

As it is, I wasn't really having issues with the open mount and fan, this is more aesthetics than anything else and has the bonus of being cooler. Having the extra fan isn't really a big issue and I'd still probably actively cool the 32 Bit board to curb any overheat issues. I've planned way ahead on this and there are more pressing things to be sorted, like proper belt tensioning and easier hot end removal before I look at the Electronics again.

Quote
Ax

Quote
lkcl

sorry, i didn't realise you were using RAMPS otherwise i would have mentioned this a while ago.

I'm fully aware that the RAMPS isn't optimal and I'm fully aware of where the printer falls short. For now, the RAMPS running Marlin 1.1-RC6 works well. CoreXY is fully supported and has all the options in there that are needed, so it does. The plan has always been to move to a Smoothieware based board eventually, something along the lines of the Azteeg Mini, AZSMZ or MKS SBASE. That'll happen when the electronics get moved to 24v.

As it is, I wasn't really having issues with the open mount and fan, this is more aesthetics than anything else and has the bonus of being cooler. Having the extra fan isn't really a big issue and I'd still probably actively cool the 32 Bit board to curb any overheat issues. I've planned way ahead on this and there are more pressing things to be sorted, like proper belt tensioning and easier hot end removal before I look at the Electronics again.

cool. literally

Hi Ax, very good job so far.
however, you make it very hard to build this printer since you make big changes.

for example, only two weeks ago i sourced all parts to build the fb2020, this was from your BOM, with all rods at 8mm.
the parts didnt even arrive yet and your already basing your design on 12mm rods and bearings in some place.
the original 8mm parts cant be found, and are probably out-of-date with all other changes.

may i suggest some versioning ?
minor fixes can be added to git immediatly (those whom dont require new hardware)
but big changes that require new hardware should be a major version change, and should not happen so often.

Ttl

Quote
ttl64
Hi Ax, very good job so far.
however, you make it very hard to build this printer since you make big changes.

for example, only two weeks ago i sourced all parts to build the fb2020, this was from your BOM, with all rods at 8mm.
the parts didnt even arrive yet and your already basing your design on 12mm rods and bearings in some place.
the original 8mm parts cant be found, and are probably out-of-date with all other changes.

may i suggest some versioning ?
minor fixes can be added to git immediatly (those whom dont require new hardware)
but big changes that require new hardware should be a major version change, and should not happen so often.

Ttl

It's planned. I'm still learning how Git and GIthub work, so I'll get there. I've not deleted any older version files, they have just been moved to the Old and Test Files/Old Files folder, so everything for previous iterations should be there. If you feel something is missing, let me know, all CAD files are current, older versions of the parts will be in OpenSCAD, newer in FreeCAD

::Edit::
I've just Branched off the 8mm Z Rod Version, if you clone that branch, that will be how the files were 5 days ago, before the conversion to 12mm Rods on the Z.

Edited 2 time(s). Last edit at 05/25/2016 05:36AM by Ax.

I almost have the updates in SW complete.

All SW files are now updated to match the files in the Github!

Quote
Floyd
All SW files are now updated to match the files in the Github!

Nice one, d'ya wanna issue a pull request with the files? I'll merge 'em in.

As an update, I've not actually had much time to do any further work. I'm currently working on Kit pricing and the build guide, so that's the focus at the moment. The Build guide has been started and is on the FB2020 Github Wiki.

Quote
Ax
Nice one, d'ya wanna issue a pull request with the files? I'll merge 'em in.

Sure can, let me remote into the home PC and see if I can figure it out.
Do I just drop them into the Github folder on my PC then hit pull request?

EDIT: Cant log in to the house. Ill get to it tonight then.

Edited 1 time(s). Last edit at 06/09/2016 02:38PM by Floyd.

Quote
Floyd

Quote
Ax
Nice one, d'ya wanna issue a pull request with the files? I'll merge 'em in.

Sure can, let me remote into the home PC and see if I can figure it out.
Do I just drop them into the Github folder on my PC then hit pull request?

EDIT: Cant log in to the house. Ill get to it tonight then.

Quickest way is to create your own fork, add the files then issue a pull request on the AxMod3DPrint/FB2020 repo. If you sling me your Github username, I'll add you in as a collaborator as well.

Few small updates. I've re-jigged the Firmware a little, nothing really has changed settings-wise, just more versioning changes. The default name is now FB2020, which you can change in configuration.h using the Custom Name option.

There are now 2 PSU mounts, one for a 12v 200w PSU and one for a 12v 400w PSU. I'm going to recommend using a 400w as high speed (120mm/s +) printing causes temperature loss with a 200w PSU, due to too much power being pulled from the PSU. I'm yet to test the 400w at 200-300mm/s (which it will do with decent cooling and enough power), but it should make all the difference.

I've started re-printing the parts for the original pink one as the filaments used weren't optimal, parts have cracked and I need to put in the 12mm rods on the Z axis and upgrade the Carriage to the latest version. This will be a complete rebuild of this one which will be used to complete the build guide. I have 3 other Migbot Prusa i3s which 2 of them will be used for conversions to FB2020s, and will be built by others to check the build guide. With all the parts, it's clocking in at about an 8 hour build, so you can pretty much build it in a day and have it printing.

Filament-wise I'm going to recommend using PETG as it's stronger than ABS, but damn easy to print, I would say easier than PLA, and it won't soften like PLA has a tendency to do.

I'm going to be changing the Spool Holder as well, this will incorporate using a 508ZZ bearing. This will allow tight clamping of the spool with a M8 Nut and washer, but still allow the spool to move. The current version works if you use 4 M8 Nuts and catch the outer clamp right, but it's not easy to use and this stops working after a while and can break if the M8 threaded rod pushes against the extrusion whilst tightening it up.

Finally, I've uploaded slicing profiles for Matter Control and Slic3r, Cura and Simplify 3D will be coming soon. This has recommended temps and settings included, which should give you a base to work from, instead of spending hours trying to find what works, believe me, it took me about 2 weeks to pin it down properly, so the profiles are there to save you the trouble.