Jump to content

So it starts - Voron 0.1 build


smirk
 Share

Recommended Posts

Well after confessing, in my intro, that amongst other sins I am lazy I have decided to pull my thumb out and get on and build my second Voron (0.1). Clearly the term "get on" is a relative one and should be viewed in context. Somewhere between the thermal heat-death of the universe and the half-life of Copernicium-235, I'll try and aim for better than Continental Drift as a good measure of progress.

I spent last night inserting the heat-inserts into the necessary parts and managing to melt a couple and misalign a few others courtesy of my sausage fingers but fortunately Voron #1 came to the rescue and churned out the necessary replacement parts.

For the purposes of this build, I'll be fitting the kirigami bed from the outset (much easier than a retrofit) and fitting an ADXL345 accelerometer. I have to confess that I utterly love the Resonance Compensation features of Klipper (saves printing a zillion test prints and never quite getting decently ghost free results).

Clearly because I am still lazy I did not lay out all the parts for the printer 😉

 

the-beginning.jpeg

  • Like 1
  • Haha 1
Link to comment
Share on other sites

  • 2 weeks later...

Gosh! Some progress. I've got the Y and Z axes done! I would love to say I've learned from my first Voron build and that this went swimmingly without a single hiccup. Clearly no one likes to be struck by lightening, so I will confess to several hiccups, perhaps more like a couple of bouts of nasty acid reflux 🤢

I did remember to pre-load the nuts for the back-plate into the opposite sides of the extrusions with the z-rails so I guess that was a plus. The main issues related to fitting the Kirigami bed and tramming the z-axis. Despite using various jigs there was a good amount of fettling required.

Issues/woes that I had:

  • The 2mm nuts/bolts that come with the Voron kit are too long for holding on the Kirigami bed. It's been so long since I bought it I cannot remember if it came with mounting nuts or if I'd just lost them. Anyway I'd ordered some M2x4 BHCS to fit the bed to the MGN9 carriages. Ok, the bolts are small but I'm sure they could've made the sockets a little larger. The bolts used a 1.3mm hex key. I don't have a 1.3mm ball-ended hex key, just a straight one. So a combination of fat-fingers and awkward angles caused me to destroy a few bolts trying to screw them in. I also think the holes on my kirigami bed are very tight with not much clearance. I decided not to drill them out on the basis that really would be asking for trouble as my drill press doesn't like fiddly little drill bits.. So it took a few attempts ("few" hah!) to get the bed mated cleanly with the carriages. Of course the poor mounting caused additional difficulties aligning the z-axes....

    My first Kirigami bed that was retro-fit to my other Voron-0 was not half as much hassle. It came with bolts which had a larger socket (certainly larger than 1.3mm); the mounting holes for the MGN9 carriages had better clearances,as well, so the bolts just went in. If I can remember who made this bed, then I would explicitly not recommend it.
     
  • Stress fractures/melting ABS. It never struck me before  that the 2mm nuts used to mount the rails are smaller than the internal width of the 15mm rails. Yeah, I guess it's bloody obvious but I'd never thought about it as I always used the printed nut-carriers. When I assembled the rails I just used regular Locktite (the anaerobic stuff), I know it attacks ABS but I just took the attitude that the nut-carriers weren't structural components and were only convenience features to space out the nuts to match the rails. Back to my first point, the nut carriers also hold the nuts and stop them rotating when tightening things up. The 2mm nuts are too small to wedge themselves against the internal sides of the extrusions (that's why we can fit 3mm nuts which do wedge themselves against the internal sides of the extrusions). Like I say, bloody obvious....but only when you see it!

    With all the fettling I was doing (courtesy of my horrible attempts to mount the bed to the carriages) I was loosening and re-tightening the nuts on the rails quite a few times. This then caused the ABS to fail, in no small part because I hadn't used my ThreeBond (plastic safe threadlocker). Please don't point out that if I'd used jigs the most I should have need to do was loosen the nuts hold the z-axis extrusions. I know,but I'd  completely borked the mounting of the bed. Anyway this all leads to the final act of stupidity.
     
  • When I realised the nut carriers had failed I had to remove the entire rail and extrusion. In the process of stripping that down, I committed one of the cardinal sins. I allowed the carriage to slip off the rail and loose a number of it's bearings 🤬 Fortunately I had spare rail and printed a new nut-carrier and all way saved.

 

I did get this stage of the build completed after way more time than it should have taken. Despite all this I'd would still recommend a kirigami bed - just not the one I'm working with 😉

 

 

Z-and-Y-axis.jpeg

  • Like 2
Link to comment
Share on other sites

Oh, I've just decided to fit a Klicky probe to this machine......naturally that means reprinting the hot-end cowl as I want the nice integrated mount rather than the one the bolts over the cowl.

  • Like 3
Link to comment
Share on other sites

4 hours ago, smirk said:

Oh, I've just decided to fit a Klicky probe to this machine......naturally that means reprinting the hot-end cowl as I want the nice integrated mount rather than the one the bolts over the cowl.

Just another day in building a Voron 🙂

 

  • Like 1
  • Haha 1
Link to comment
Share on other sites

Posted (edited)

I found an order for my  Kirigami bed - looks like it came from PrintyPlease (I also note they don't seem to be selling them anymore). I really would not recommend a Kirigami bed from PrintyPlease. I've used them for other items and generally the stuff seems to be OK. I also did not mention that the bed is made of "stainless" steel so is arguably heavier than the stock bed. Although I did not weigh things to start with. It now comes badk to me why I bought a second (aluminium) bed to retrofit to my first Voron.

Oh well, if I can find better M2x4 then I think a retrofit will be on the cards with a better version.

Edited by smirk
Link to comment
Share on other sites

<Sigh> After completing the Z & Y axes I moved onto assembling the rest of the frame....and it worked.....however, reading a couple of the [serendipitous] posts elsewhere about Kirigami beds it got me thinking. Well possibly riled up in a grumpy old man sort of way 🥵 so I decided I'd had enough with my horrible bed and it's awful mounting bolts. It didn't help, that I'd found a box of M2 SHSC with the "proper" 1.5mm sockets (for which I do have a ball-ended hex driver). So I removed the Kirigami bed, with much cursing, and managed to strip  the sockets of a couple of the bolts so I had to be a little creative. Perhaps not a complete restart but definitely a step back. Ho hum such is life 🤔

I don't seem to be able to find a supplier of Kirigami beds in the UK, and I don't want to wait a month of more for Ali, so I guess I'll keep using my super-heavy PrintyPlease bed....

  • Like 1
Link to comment
Share on other sites

On 6/27/2022 at 1:43 PM, smirk said:

....and it worked.....

No it didn't! While I contemplated trying to reassemble the bed between meetings I noticed I'd missed the two nuts (between the Z axes) for the Z motor mount. To paraphrase Ford Prefect " I counted them twice when checking my work." and clearly they weren't there both times.....

I think I'll strip the entire frame and start again 🤪

Link to comment
Share on other sites

Disassembled the entire frame this morning. That was quick (once I'd put all my labels for which extrusion is which back on). Then spent the rest of the day rebuilding the z-axes. It gave me a chance to try out my new Mitutoyo calipers. I am happy with the basic frame, nice and square and everything parallel. Then I tried the Kirigami bed with my (proper) M2x4 1.5mm SHSC bolts, and at that point it all went to 🤬

Before I ever used the bed, I made sure the mounts were "flat"/aligned in the "y-direction" (that seemed to be the only misalignment). The previous time I had fitted the bed I noticed the mounting "flaps"were slightly out of alignment in the "z - axis" direction (up and down from the printers perspective). Not by much but enough to twist things, so I'd straightened that out. I did recheck the alignment of the mounts again before fitting the first time but we saw how that ended. I'm not sure if I did not recheck properly or if I've  mucked things up with my "inventive" means of removing the old stripped rubbish bolts but I've just re-checked (again again) and hey presto,as if by magic one of mounts is slightly bent (in the y-direction). Only one this time. Thinking about it I probably did twist it when removing the old bolts.

Now what amazes me is,if I try to deliberately bend things it won't budge (cause I'm an elderly weakling) but using the power of my mind and simply looking at it and it goes all kinds of misaligned 🤔

Anyway, I got it all strapped down, as it were, so hopefully by the morning it will be back to aligned and keep it's shape. Don't hold your breath though, I'd hate to be the cause of your expiration ☠️.....

  • Like 1
Link to comment
Share on other sites

Partial results, managed to realign the mounts....then it struck me. I'd never checked the "deck" of the kirigami bed. I assumed that was flat and it was only the mounting flaps that were out of whack.

It's always a humbling experience when the universe reminds you of what assumptions do.....nope, the deck of the bed is also out of skew. Only bent by slightly less than a mm but bent all the same.  Clearly I'm not surprised. Fortunately, being stubbornly cantankerous is a super-power so I'm not giving up........(yet)......

So the clamping continues, and no doubt once I've got the deck straightened I'll need to realign the mounting tabs (better word than "flaps") 😵

  • Like 2
Link to comment
Share on other sites

5 minutes ago, claudermilk said:

So what issue is the Kirigami supposed to resolve? It's sounding like a lot of trouble.

Apparently it's meant to resolve the issue of the stock bed sagging over time, and certainly on  my first V0 I did find myself having to adjust the bed more frequently until I replaced that with a Kirigami bed. Guess I could've stripped/rebuilt the stock bed, however, I invoke the universal law of "just cause I  can" 😋

To be fair to the notion of the Kirigami bed (or any rigid bed) I think the problem here is a combination of the generally increased rigidity/coupling of the solid all-metal bed-frame (vs a certain flexibility/forgiveness of the "composite"/multi-component stock bed-frame; [in this case]  poor implementation (questionable dubious quality of manufacture); general fecklessness and overly optimistic assumption on the part of the owner (that the bed had sufficient basic quality) and lastly fat-finger-syndrome coupled with piss-poor piddly-little mounting bolts (see dubious manufacturing) that conspired for the first round of frustration.

My first Kirigami bed which was a retro-fit was nowhere near so problematic. That was aluminium rather than stainless steel; that come with decent M2x4 1.5mm SHSC rather than those odd 1.3mm M2x4 and it only required very  minimal fettling to get the mounting tabs parallel with the MGN7 blocks.

Beyond the immediate frustrations (and no doubt at the time  cursing) I think I'm enjoying the challenge. Either way I don't seem to be looking for an easy life (more of that in a while).

  • Like 2
Link to comment
Share on other sites

Well continuing the subject of not making things easier. Perhaps it's a deep-rooted subconscious desire to sabotage my build  so it never completes.....although I really think it is an altruistic humanitarian concern for Jeff Bezos. He's approaching his twilight years, I'm concerned he's down to his last hundred billion or so....how will he cope?

Anyway, a combination of discount deals and vouchers (and a humanitarian concern) convinced me it would be a good idea to purchase a BTT Pico from Amazon. I have a couple of PI Zero 2 looking for a use so why not?

I suppose I was also not looking forward to trying to get the SKR E3 working in the new build. I don't like the boards, I've run into a few of them and they've all been a disaster (I know VoronDesign use them but I've never had luck with them). I had to resort to using an SKR 1.4 in my current V0.1 which  is a smidge large and I have two cooling fans on the back of the unit (one for the PI and one for the SKR). The only other boards I have on hand are Spiders and S6s and beyond physical size they're a little OTT for the V0.

So yeah, when the opporchancity came up to grab a PICO I thought why not. I'm hoping I'll be able to make a neat electronics package with the PICO+Zero and get away with a singular decent fan flowing across both.

My only concern (beyond niggling quality concerns over BTT/Biqu stuff) is the PICO appears to have the same design "flaw" as the E3. (Going to use clumsy words here) The TMC drivers and their passive cooling (heat sinks) are on backwards. The heat sink(s) are mounted to the (thermally insulating) plastic of chips rather than mounted on the copper flooded area on the opposite side of the of where the drivers have been mounted/soldered to the board (and connected using vias). There is a very small copper area on the back of the PICO board (which seems way smaller than the foot print of 4 2209 drivers) but that's not adequate (especially as the heat-sink is on the wrong side). I'm pretty sure Trinamic recommend 4oz copper for the purposes of heat dissipation but no doubt this board will be using the more common/cheaper 2oz copper.

Anyway, I'm not about to prise the heat-sink off to check, so hopefully I'm just being mean-spirited and overly pessimistic. It'll be fine once I get a decent active cooling set up, and I've still got that warm glow from knowing that I've helped keep poor old Jeff from destitution.

pico-unboxing.jpeg

  • Like 2
  • Voron FTW! 1
Link to comment
Share on other sites

Thanks for the heads up. Was going to look at the pico as an option for another build. Maybe I'll stay clear of it. Have the SKRe3 Mini v3 currently in the Voron with the Ice Cooling fan on the pi and printing ABS with chamber reaching 60 degrees. have not had a problem ....so Far.....

 

  • Thanks 1
Link to comment
Share on other sites

Perhaps it's just the balmy sub-tropical climate we have here in Scotland, after all we have the gulf-streaming bathing our shores 🤣 , and no aircon?!

I was thinking that perhaps a largish heat sink on the bottom of the E3 board (across the copper heat dissipation area) might work but my experiments with the small heat sinks I had on hand made no real difference but I think I was struggling to get decent airflow across them.

I'll see how I get on with the PICO and update here.....guess I've always got a spare Spider board.

Link to comment
Share on other sites

Bodgetastic!

Some success. After severely  underestimating how out-of-whack that Kirigami bed was I've come to a compromise, a bodge for lack of a better word. Not great, but I didn't want to burn more time on it (diminishing returns and all that) and I didn't want to wait a month for a new bed and potentially a whole set of new woes.

To kind of sum up:

  • The tabs (as expected) were out of alignment in the y-axis (so they weren't parallel with the faces of the MGN7 carriages), this is just a matter of bending them into alignment. That should have been the limit of the required fettling to get the bed fixed (it's common with the kirigami beds).
     
  • Sadly the whole deck had a slight twist in it.  That's not easy to bend, certainly not by hand so I put that down to a fault in the manufacture. However the twist caused the mounting tabs to go out of alignment in the z-axis (so they weren't parallel with each other).

Both issues were "easy" enough to fix with a lot of bracing and clamping. Which then brings me to the core-issue, the real source of al the woes:

One of the mounting tabs itself is twisted so it will never be parallel with the face of the MGN7 carriage. This meant when I was mounting the bed that twist was being transferred (as it were) to the carriages and causing them not to run true.  The twist is yet another defect in the manufacture - I had thought I'd bent the bed when I'd removed it originally but without a hydraulic press you couldn't bent it this way. Clearly,  I have been unable to straighten that twist out which brings me to the solution (well bodge). SHIMS!

Placing a couple of shims at the "high" side stops the mounting tab (and whole bed) from rocking/twisting when it is attached to the MGN7 carriages.Which in turns prevents the carriage being twisted.

Everything now runs smoothly and the bed even falls under it's own weight.

It's not a great photo but you can just make out the shims

v0-shim-bodge.jpeg

  • Like 2
Link to comment
Share on other sites

Déjà vu time, finally got back to where I was days ago and have the frame built (again) and all the nuts installed. The frame is square and the bed moves smoothly. Hoorah.

v0-deja-vu-frame.jpeg

  • Like 3
Link to comment
Share on other sites

Well got all the bits for my klicky probe printed out. I printed a couple of the dock mounts (top and side screws, since I  added the necessary screws at  the start since it's a new build) and a couple of the klickyprobe "arms", a long and short version. Also  glad of @Demosth build tip for dealing with ABS stress marks, saved me having to print another hotend cowl

Not sure what the advantages of one  klickprobe arm is over the other? I guess the long arms position the switch nearer the nozzle, which I presume means you're more easily able to probe the whole bed (i.e. everywhere the nozzle can reach)? I'll see how it looks when i assemble it.

I also got some samarium cobalt magnets as they're good to 300C (apparently). There again not having ever heard of a rare-earth called samarium they could well have been made from vibranium or illudium fosdex.....

 

v0-klicky-parts.jpeg

  • Like 2
Link to comment
Share on other sites

  • 2 weeks later...

Well, after a weeks hiatus down at the seaside (kicking over sandcastles and complaining about the heat; the sea being cold; sand getting every where and the kind of music people listen to these days which doesn't have a proper rhythm you can tap your toe to and you can't make out the words (unless it's profanity)) I'm back and refreshed.

I got the A/B motors built and attached to the frame today...[alarmingly] without any drama

v0-assembled-motors.jpeg

v0-fitted-motors.jpeg

  • Like 3
Link to comment
Share on other sites

Got the rest of the motion system installed (idlers, x-gantry, belts). Quite pleased with that. Managed to get the belts run and tightened reasonably easily, there always some corner where it proves a little fiddly feeding the belt.

I getting ahead of myself but I am thinking about the BTT PICO mount that I will need. I've not seem one that I like (and I dread the thought of trying to make one). I want a compact set up that manages the cabling properly. Most of the PICO based setups just seem to wrap the cabling around the back of the board. That might give a clean appearance on the front/top side but there's a rats nest behind the board. Definitely need to look into that some more.

 

v0-assembled-1.jpeg

  • Like 3
Link to comment
Share on other sites

Garn! I'm even grumpier today. Didn't sleep well, it was too hot (in sub-tropical Scotland) so kept waking up and when I did I'd think about mounts for the SKR PICO. Sadly I didn't come up with a solution (probably the real reason for my grumpiness).

When I was seduced by the rinky-dinky size,my hope was to come up with a nice little package not much bigger than the Raspberry PI foot print, that housed all the electronics for the V0. My current V0 (as I've mentioned) uses a full size PI, a SKR 1.4 and two 80mm fans on the back to keep each device semi cool. I don't think I'll ever get to the stage of being able to put the back-panel onto the electronics bay (to my mind an inexplicable choice by VoronDesign) but I was hoping for a neater set up.
 

The PICOs diminutive size is it's draw but that is also it's biggest weakness to my mind. Possibly by trying to match the PI footprint they've made it too small. To save board space (and presumably cost) they've stuck the majority of the connectors out the side, parallel to the board, rather than the more usual perpendicular to the board setup. That arrangement might have been fine if the board was being mounted flat  to something in a space with only 20mm of head-room but with all the other issues you would never be able to do that. Another trouble is the connectors jut out from the edge of the board (see picture) which means the setup isn't really Raspberry PI sized (the board might be but the package isn't). This side orientation also constrains your cable management choices/routing. With top traditional mounting you could decide which side of the board to run the cables off of, with the side mounted option without [serious] looping it around you've got to route the cable on the side the connector pokes out. The board has cables on every side so there's not a nice sleek setup - it's a hairy pancake.

The notion of mounting the PI and the PICO together is really only a nice idea, given all the flaws. If you mount the PI "underneath" the PICO, then you have full access to the cables on (the top of) the PICO but you'll be buggered trying to access anything on the PI (Camera connector, GPIO pins). If you mount the PICO underneath the PI then the issue is reversed (you can't access the top-mounted connectors on the PICO).

Even using the PI Zero format and a top mount you'll still have issues. The PI Zero is still large enough to obscure about 50% of the top-side connectors on the PICO (see picture)....unless.....you mount the PI Zero over the heat-sink on the PICO. Somehow I think it is asking for [cooling] trouble mounting a hot-piece of electronics over the heat-sink of another hot-piece of electronics.

Despite having one looking for a project, the PI Zero (v2) in this context is concerning me. I'm sure it only has 512Mb of memory and I'm sure I read somewhere that trying to run klipper (even with mainsale/fluidd) and a camera would cause it to struggle. Might just have to experiment.

Cooling, as I've already grumped, is going to be a major problem on this beast and the more I think of it, sandwiching two "hot" pieces of electronics together, without decent air flow is troublesome. My gut feeling is the sandwich orientation would require airflow across the boards (i.e fan at side rather than top) but that runs into problems if you've got all those PICO cables wrapped around under the board (thanks to the side mounted connectors). Given the limited space in the V0 electronics bay you're looking at a fairly gutless 40mm Fan (which probably is only capable of 5-6CFM). Especially as  I have my doubts/concerns over the effectiveness of the PICO heatsink.

Then there's height. The electronics bay on the V0 is about 50mm deep (I think). The SKR PICO  from (roughly) highest to lowest is about 14mm (and by lowest I'm taking into account the circuit leads/pins poking out of the bottom). A raspberry PI is roughly 19nn tall (again roughly from highest to lowest point). Naturally given the variability of how far the lead/pin from any electronics sticks out from the bottom of the board you will want some clearance. Unless your base mount and the gap between the boards is only going to be 5mm (which is ludicrously small) you stand a good chance of having an electronics stack (without cooling fans) that is going to be the height of the V0 electronics bay. Having said that you will need more than a 5mm gap to provide enough room for cable management, connectors on top of boards, etc.

Well, I feel better for getting that off my chest. If you did read my ramble, then thank you 😃

 

skr-pico-topside.jpeg

skr-pico-underside.jpeg

skr-pi-sandwich.jpeg

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

4 minutes ago, smirk said:

Youtube (as well as google) can be your friend. Nero3D has allayed my concerns over the PI 0 v2. It does seem to work.

Find Nero3D channel very helpful and informative.

13 minutes ago, smirk said:

When I was seduced by the rinky-dinky size,my hope was to come up with a nice little package not much bigger than the Raspberry PI foot print, that housed all the electronics for the V0. My current V0 (as I've mentioned) uses a full size PI, a SKR 1.4 and two 80mm fans on the back to keep each device semi cool. I don't think I'll ever get to the stage of being able to put the back-panel onto the electronics bay (to my mind an inexplicable choice by VoronDesign) but I was hoping for a neater set up.

Have the back panel on and using an Ice tower cooling fan on the pi. Seems to be working well. But I feel you pain.

IMG_2488.thumb.jpeg.f9d9c94490b8f66401f1080994e72a4d.jpeg

  • Thanks 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...