Poor Man's Tool Changer
OK... after being inspired by @jurgen-7's 'Poor man's automated chamber top vent' and the video that Prusa teased about different purpose toolheads for the XL, it got me thinking about a simple solution to apply permanent marker to the interface layer of supports.
I wouldn't even call it a prototype yet...more a proof of concept. I know much more complex solutions have been put together in the past, but I wanted something quick, easy, cheap and with minimal hardware.
I have to say I'm quite pleased with the results I'm getting for a very early 'prototype'.The hardware and construction really is the simple part, 2 x M2.5x16 hex screws, and 4 x 8mmx3mm neodymium magnets. The magnets, in pairs oppose each other creating a spring effect on the sharpie...which allows a bit of legroom for setting the z offset...So far its survived about 20-30 successful prints, the only wear is slightly on the sharpie tip which is expected, at £1 each I can handle 1 or 2 of those per week as a consumable.
The most challenging part I've found is trying to work out a quick easy workflow to create the gcode to do what is required.
Basically, a pause print (M601) is executed after the interface layer of the support is printed... the sharpie thingy is attached, then on resume print the gcode from the interface layer (of the support only) is repeated...then another M601 to remove the sharpie thingy, then the print resumes as normal.Its really nothing complex.
As someone who makes and sells a part that requires me to manually add permanent marker, this alone will save me at least 30 minutes per day....so i feel the time spent on this will be paid off relatively quickly.
Theres a few things I can (and need to) improve on:
1: Accurately determinig the offsets of the sharpie, X/Y/Z...at the moment I've just eyeballed it 😂
2: Rotating the sharpie clamp 90 degrees cw...it will make the footprint smaller and mean less potential printbed space is sacrificed.
3: A more suitable mounting solution...right now it just grips the 2 screws on the hotend fan. Not ideal, but good enough for a POC as it is only on there for as minute or so per print.
Apologies for the terrible quality videos:
RE: Poor Man's Tool Changer
Better quality video here:
https://gbdesigns.ddns.net/static/IMG_1099.MOV
Honestly the pictures (or more precisely my shoddy photographic skills) don't do the surface finish from the supported surface justice...they're comparable to the surface of the unsupported side of the bracket.
RE: Poor Man's Tool Changer
That looks awesome!
Would you share a "sample" 3mf + final G-Code so I can engineer something on my own from there?
RE: Poor Man's Tool Changer
It’s up to you mate. I’m happy to share my design but there’s a few snags I want to address first.
The gcode editing really is the awkward and tricky part.
I think this is much more useful for parts that you need a constant supply of, as the legwork at the moment to manually arrange all the code just isn’t worth it for a one off print…I need 20-40 of these parts per week, so once the gcode is final I can just reuse it over and over again.
I’ll see if there’s any way the gcode editing can be completely scripted…that’s the ideal scenario.
Currently I’ve scripted applying the required offsets to the interface layer, but that gcode then has to be included in the main gcode for the part and support.
RE: Poor Man's Tool Changer
I'd be very interested in that! I totally expected, what you say in your last post. I can code some C/C++ for automation/statemachines/buttonboxes and a bit of Python and Javascript.
I'm nowhere near the skillset to complete your script, but I know enough to be very curious and look through your project/script/final gcode.
I agree, that for single prints, it would be easier to just pause and draw manually, but I printed some projects in the past with a lot of little supports and would've liked to just put a pen on the print head.
I bought the GPIO Hackerboard with my Core One kit, so maybe there's a way to route another cable along the swing arm and build something to automate the pen positioning.
Which permanent marker are you using btw?
RE: Poor Man's Tool Changer
The script I’m using is python, I can use other languages but I’m most comfortable with python.
I too have the gpio board but haven’t had the time to experiment with it, to truly automate it so no intervention is required at all, it’s just a question of dropping the sharpie below the nozzle a few mm, and raising it above the nozzle a few mm.
Im using Sharpie fine points, primarily because they have a fine point, but also because they’re really cheap. Amazon sell 2 for £2 so at that price you can’t really go wrong.
When I get the chance I’ll put the fusion file and the script up on GitHub.
RE: Poor Man's Tool Changer
Oh wow, that's awesome! I can't wait to see your model with the GCODE implementation on Printables.com haha!
Seriously tho, that's the kind of thing I would completely share with Prusa team, I think it should be "quite easy" for them to had something like that in the Slicer.
Amazing work!
RE:
I know that the hotend/nextruder was never open sourced, but is anyone aware of any accurate cad drawings for this? Accurately finding the offsets would be a complete breeze if they were available.
RE: Poor Man's Tool Changer
I know that the hotend/nextruder was never open sourced, but is anyone aware of any accurate cad drawings for this? Accurately finding the offsets would be a complete breeze if they were available.
Someone posted it on Printables I think, based on the model you can see when slicing by objects instead of layers.
HERE
RE: Poor Man's Tool Changer
I know that the hotend/nextruder was never open sourced, but is anyone aware of any accurate cad drawings for this? Accurately finding the offsets would be a complete breeze if they were available.
Someone posted it on Printables I think, based on the model you can see when slicing by objects instead of layers.
HERE
It's an STL, which isn't ideal, but I should be able to get what I need, thanks.
Now I'll find found out that my eyeballing method was way out 🤣
RE: Poor Man's Tool Changer
Hahahah, that means that the STL is better than nothing. 😆
And indeed, there are no "official 3D model" splitted in parts unfortunately...
RE:
Blimey I couldn't work out why my z offset was so far out...around 3mm which didn't make sense.
Took me about 20 minutes before I realised I've put an ObXidian HF nozzle on my Core One which is around 3mm taller than stock 😂
My eyeballing guesstimates weren't too far off overall...a few 10ths of a mm here and there, but it's good to have something more concrete than my sketchy eyesight, so cheers for that.
RE: Poor Man's Tool Changer
So I finally got around to spending a bit of time on improving this, and I'm quite pleased with the improvements so far.
It's now completely automated by post processing so no user interaction is required which is great, although there is one caveat that I'm hoping to address next time my stupid life lets me have some more free time.
Apologies for the poor video, you can't really see any ink going down as it's black ink on black filament , and also im very bad at cameras. My previous videos show better angles at that.
RE: Poor Man's Tool Changer
Sharpies dry out quick when they're left uncapped, in hot windy environments....who knew ? 😂
RE: Poor Man's Tool Changer
I really like your build plate level prime line after the pause, even though the print is well underway. It would be cool if you could do progressively move the that line to 4-5 locations based on how many print pauses you've done on the job (so as not to crash into previous lines). I rarely need to pause prints, but when I do it's a pain to try and minimize the ooze before resuming, and even then there is often a void left in the print because of lack of prime.
If a progressive method was possible, you could get *5 pauses per print job without having to deal with as much ooze or voids.
-J
RE:
I really like your build plate level prime line after the pause, even though the print is well underway. It would be cool if you could do progressively move the that line to 4-5 locations based on how many print pauses you've done on the job (so as not to crash into previous lines). I rarely need to pause prints, but when I do it's a pain to try and minimize the ooze before resuming, and even then there is often a void left in the print because of lack of prime.
If a progressive method was possible, you could get *5 pauses per print job without having to deal with as much ooze or voids.
-J
It's achievable in post-processing mate....You could keep the same purge Y position, I think I use 219, and then just subtract 30-50mm from the purge X position for each pause....you should be able to get at least 4 maybe 5 across the back of the bed....I'd have to do some maths.
I've only got one pause during most parts I make so I've never really considered having to add more.
PETG is such a great filament in so many aspects, by my god the oozing is bad....for PLA prints could probably get away with no purge at all.
RE: Poor Man's Tool Changer
I just realized the humor in my appreciation for the prime line, which is a minor piece of the much more impressive work that you've accomplished here. 😆 This is genuinely cool. Thank you for sharing this progress.
-J
RE: Poor Man's Tool Changer
That's cool buddy, I was quite proud of the purge in its own right anyway, took me a while to get it right 😂
Im amazed it's not a more requested feature to Prusa, any sort of pause print with PETG is hopeless without purging imo. You can kind of get around it with a sacrificial part but that just seems wasteful to me.
RE:
I'm just touching on custom gcode, and haven't done any of my own writing of it yet, so I'd need to see a framework for where to start on this, and work out the script for logging pause count etc..
I don't have any scripting/coding training, but I really need to learn because of stuff like this. Almost all of my prints are functional, so up until the core one, I was limited to PETG, so the oozing was always front of mind. I've only recently designed an exhaust adapter and some PC case fan adaptors so that I can vent my exhaust out of the garage. I'm just getting into ASA and other filament testing, and it's wonderful to not have the oozing problems. On that front, I've found that (at least in the hot summer months) I can run a low speed case fan on the far end of an exhaust hose pulling slight negative pressure at the exhaust vent, and still maintain 55c + while printing PC or ASA. I turn this remote case fan up to high for PLA/PETG to help the chamber fans do get rid of those fumes as well.
-J
RE: Poor Man's Tool Changer
There’s not really that much to it. What caught me out was that sometimes gcode commands aren’t always executed in the exact order they appear. Obviously all the extrusion moves are, but sometimes other commands are buffered and executed at seemingly random times.