Thanks Thanks:  8
Likes Likes:  16
Dislikes Dislikes:  0
Page 3 of 5 FirstFirst 12345 LastLast
Results 21 to 30 of 50

Thread: Universal Hub

  1. #21
    3D Printer God(dess)
    Join Date
    Nov 2016
    Posts
    1,572
    Post Thanks / Like
    If you were stateside, I'd be happy to send you a couple of cogs.
    That is definitely a sign something is working way to hard. Rounded out of the box? Wow!
    However, if they are new cartridges, they are still product, so they are still warranted.
    Problem is that they only warrant cartridges sold by their qualified distributors.
    They are really pushing the consumer protection laws in the USA.

    I'm not into risking the heater block either.

    Matching up a pair of nozzles is a pain but now we can just leave it be.
    For the most part, this finally puts the nozzles with the machines rather than the cartridge.
    If only someone would up the count between maintenance warnings.
    I don't -want- to go through the routine again

    The only print I was having trouble with was the bulkhead tubenut in ABS.
    Haven't tried the new filament yet.

    BTW, the groove depth when I turn it on the lathe is 0.4-0.45mm.
    If you can turn a nozzle, you can sure carve a groove with a parting tool

    - - - - - - - - - -

    B2B Bulkhead parts. I've included PLA, ABS, and PETG.

    If you have a #34 drill bit, run it through the center.
    #34 is .111" or 2.8mm

    The design is a snug fit that allows us to remove the barrel nut from inside the driver mechanism.

    If you lack the right size drill bit, some sandpaper or a hobby file will do the trick.
    It doesn't need much adjustment, just smoothing out the bore basically.

    https://drive.google.com/open?id=1s5..._r9iPFwnzPX4Fi

    Remember, these drop in for a left side cartridge. Turn them 180 for right side printing.

    The tube segment which has to be salvaged from the original bowden tube is 98-99mm long.
    You can push back the nylon bulkhead and trim that section to length if nothing else.

    Interesting trick to salvage the stock bowden tube is to carefully remove the skeleton.
    If the tubing is not to severely damaged, run one of the parts with the 2.8mm bore over the tube with a bicycle spoke inside.
    This will smooth out and -restore- much of the original shape of the tube.
    The pre-threaded bits won't heal, of course. Just trim those off.
    Last edited by TommyDee; 07-15-2018 at 04:46 PM.

  2. #22
    3D Printer Noob
    Join Date
    Jun 2017
    Posts
    15
    Post Thanks / Like
    Quote Originally Posted by TommyDee View Post
    If you were stateside, I'd be happy to send you a couple of cogs.
    I’ll see how desperate I get before I potentially end up begging for you to send it to family west side Thank you!
    Quote Originally Posted by TommyDee View Post
    That is definitely a sign something is working way to hard. Rounded out of the box? Wow!
    However, if they are new cartridges, they are still product, so they are still warranted.
    Problem is that they only warrant cartridges sold by their qualified distributors.
    They are really pushing the consumer protection laws in the USA.
    Bit of an interesting one, I bought my printer from Ebay about two years ago which was previously a demo unit in a retail store, 7 cartridges and the printer 205 all in. Many of these carts are dated from 2014 with no proof of purchase ☹
    I got a bit creative with a t20 bit and a hammer and managed to recover it.
    Quote Originally Posted by TommyDee View Post
    The only print I was having trouble with was the bulkhead tubenut in ABS.
    Haven't tried the new filament yet.
    I’ve printed that and all other parts in 3ds ABS from a stock cartridge and only minor issue is the bulkhead tubenut has a teeny bit of resistance and isn’t as buttery smooth as the print head, the print head bottom is also very weak so any overtightening causes the end to pop off meaning you need to print another, I’ve accidentally done that twice now, DOH!
    On the bright side I just ended up going all texas chainsaw massacre on the tubing and it’s printing admirably, no issues from 3 hours of printing so far.
    B2B.jpg

    lack bracket.jpg
    Last edited by Ardoneye; 07-15-2018 at 05:13 PM.

  3. #23
    3D Printer God(dess)
    Join Date
    Nov 2016
    Posts
    1,572
    Post Thanks / Like
    Love to see a success!

    The bottoms of the nozzles shouldn't be weak.
    Were these printed with the stock bowden tube?
    That is where I thought my orange was junk but it turns out it was the stock bowden tube.
    Even the bad prints are stronger that I can break easily.
    My orange is about the same vintage as your green.

    I don't tighten the hot-end tubenut much at all.
    I want to be able to loosen it easily while paused without tripping the nozzle present switch.

    A rougher feel to the bulkhead tubenut is no big deal.
    This is only removed when you have the cartridge in hand anyway.
    The prints are definitely optimized for PLA.

    Does your motor shaft show a lot of wear? You can always swap motors from right to left.

    Can you post an image of where the nozzles are separating?

  4. #24
    3D Printer Noob
    Join Date
    Jun 2017
    Posts
    15
    Post Thanks / Like
    These were printed with a completely standard stock ABS cartridge without modification.
    I was fully tightening the end without paying attention and didn't stop as I didn't encounter any resistance so kept going, the trick with these is to stop tightening when movement on the nozzle has stopped. Overtightening causes the nozzle and spring to protude from the end and the plastic to snap like so:

    brokennozzle.jpg


    Measuring the thickness of the snapped circle at the bottom it's currently measuring in at 1mm.


    The motor looks reasonably fine, I have a feeling the previous owner molested a few previous carts when they were playing up

  5. #25
    3D Printer God(dess)
    Join Date
    Nov 2016
    Posts
    1,572
    Post Thanks / Like
    It definitely shouldn't do that, and yes, this is likely due to the stock cartridge.
    This is the exact reason I considered my orange cartridge bad, because the 1st couple of layers were always suspect.
    Not anymore! ...As the B2B without sidewalks is working perfectly now on these parts. PLA was never an issue.
    if you are experiencing some "elephant foot" on the bottom, just carve it smooth with a hobby knife.
    Any "edges" at the base will catch on the hot-end when removing the nozzle.

    Sidewalks will make this a little weaker, only in that the sidewalks are bonded too strongly to the part.
    The only thing I am sure to do is to "shear" the part from the plate rather then bend them over.
    There is a sharp transition after the second layer so this is a highly stressed region on removal from the plate.

    Technically, the bottom ring is only there to manage the push nut when not in the printer.
    Nothing in the hot-end makes contact with this part of the nozzle housing.
    All the work is being done between the top of the "knob" and the "keeper".
    The 3 slots make sure there is additional "bonds" where the body is only 2-traces wide.
    If you have to bend it off the plate, push it backwards while prying at the "knob" when printing from the left side.

    I had given serious thought of putting the nozzle in the printer and using the housing as a keeper only

    - - - - - - - - - -

    NOTICE: If you downloaded the bowden set or the individual tube-nut for the nozzle, you might download the set again.

    Too many variations on the hotend tubenut mixed things up on my end. I've corrected it with the latest download.
    There will be 4 files in the zip rather than the previous 3.

    Sorry if this causes confusion. The correct files have a little extra clearance around the tube and ringlet features.

    - - - - - - - - - -

    Quote Originally Posted by Ardoneye View Post
    These were printed with a completely standard stock ABS cartridge without modification.
    I was fully tightening the end without paying attention and didn't stop as I didn't encounter any resistance so kept going, the trick with these is to stop tightening when movement on the nozzle has stopped. Overtightening causes the nozzle and spring to protude from the end and the plastic to snap like so:

    brokennozzle.jpg


    Measuring the thickness of the snapped circle at the bottom it's currently measuring in at 1mm.


    The motor looks reasonably fine, I have a feeling the previous owner molested a few previous carts when they were playing up
    D'oh! I know what happened. Or at least a suspect condition.

    I've been rebuilding all my setups to B2B. While I was rebuilding nozzles, not all the 42mm long tubes would seat fully in the nozzle.
    If the 4mm tube was already in place in the tube-nut when you set the clip in place, this could force the nozzle out the end.

    What to do if a tube does not fully seat in the nozzle:
    Clean it out! Simple enough, right? I drill mine out by hand with a 3mm twist drill. It is marked for depth so I don't cut any aluminum.
    This also cleans out the muck they used to hold nozzles in place while installing the housing.
    I also do the periodic maintenance of clearing out the aperture by running a wire through backwards.

    However, this is an alternative to makng sur ethe tube isn't blocked by cold plastics:
    The idea of the snug fit of the tube in the core is to help make sure it is properly installed.
    Use -only- the tube-nut when setting the level of the core in the housing. No tube!
    This way, the 42mm long tube can move if it needs to.
    Once the keeper is installed, you can remove the tube-nut.
    Take a look at where the 42mm long tube is. Is it still seated at the bottom of the socket for the 4mm tube?
    If so, you are good to go.
    However, more often than not, it won't be if you don't clean your nozzles first.
    We can use the printer to fix this;
    Put the nozzle housing assembly in the printer. Just the housing, nozzle, spring, core, 42mm long tubing, and clip.
    Put in a cartridge.
    Run the test print... when print head moves to the plate, cancel twice...
    Then push a piece of 4mm tubing into the core's socket to "seat" the 42mm long tube.
    Once seated, you can load the filament by connecting a bowden tube as previously discussed.
    Last edited by TommyDee; 07-16-2018 at 12:20 AM.

  6. #26
    3D Printer Noob
    Join Date
    Jun 2017
    Posts
    15
    Post Thanks / Like
    Printed a good 10 hours or so with the original filament with this setup without anything keeping the tubes in on either end of the nuts and no slipping or problems at all.
    Filament eventually ran out and switched over to third party filament and the need for the circlets finally becomes clear, it prints fine but sooner rather than later:

    BowdenSlip.jpg

    Solution 1: I finally give in and faff about trying to make the circlets by getting creative with a dremel/drill and the new tubing to make a circlip
    Solution 2: I print the groove cutter set and buy an eyeglass screwdriver thingy
    Solution 3: I get creative with tinkercad and merge your m6 thread from the hot end tubnet onto the bulkhead tubenut and use a push fit connector until I fancy attempting solution 2 when I have more time available.

    Solution 3 is looking like the easiest option in the short term.

  7. #27
    3D Printer God(dess)
    Join Date
    Nov 2016
    Posts
    1,572
    Post Thanks / Like
    Wow! all I can say is "WOW"!

    I will have to try to simplify the grooving tool.

    Realizing the cutting blade can be made with a bike spoke opens up a few possibilities.

    So dumb question... if you don't have a lathe, how were you going to try to make a nozzle?

    You are aware that the previous M6 option posting includes both ends modified for M6, right?
    Bottom of the post...
    http://www.print3dforum.com/showthre...ll=1#post44567
    Last edited by TommyDee; 07-17-2018 at 06:51 PM.

  8. #28
    3D Printer Noob
    Join Date
    Jun 2017
    Posts
    15
    Post Thanks / Like
    Quote Originally Posted by TommyDee View Post
    Wow! all I can say is "WOW"!

    I will have to try to simplify the grooving tool.

    Realizing the cutting blade can be made with a bike spoke opens up a few possibilities.
    The amount of plastic and parts put me off attempting the solution. I'm in a rush to build a lack enclosure and replacement desk at the moment and needed some brackets etc like yesterday.

    Quote Originally Posted by TommyDee View Post
    So dumb question... if you don't have a lathe, how were you going to try to make a nozzle?
    Dremel off the section of the refil carts to match the length of the printing nozzle, hope it's the same width, swear that it's not, use a vice/modified dremel 3000 to turn it into a make shift drill press. Thread a nut onto the end of it with a tap and dye set to make the circle part....Didn't say it was a well thought out plan

    Quote Originally Posted by TommyDee View Post

    You are aware that the previous M6 option posting includes both ends modified for M6, right?
    Bottom of the post...
    http://www.print3dforum.com/showthre...ll=1#post44567
    Turns out I went with the hidden option 4:

    Option 4: Swear I've made a Frankenstein'd monstrosity that seems to do the job and wish I'd printed: http://www.print3dforum.com/showthre...ll=1#post44567

  9. #29
    3D Printer God(dess)
    Join Date
    Nov 2016
    Posts
    1,572
    Post Thanks / Like
    Okay, I needed a good laugh Carry on!

    - - - - - - - - - -

    Loose a nozzle housing spring?
    Find a ball point pen spring that fits...

    penspring.PNG

    - - - - - - - - - -

    For some reason I have a lot of trouble finding things on the forum after a short period of time.
    There are a lot of fragmented projects here that I am trying to consolidate in Thingiverse.

    One part of this project that should be easily found is this adapter to earlier efforts.

    If you went through the efforts to make this hub adapter, I've also made an adapter for this B2B effort.
    This requires a 57mm length of 4mm tubing in this housing to make the smooth transition.

    (crap, now Chrome won't let me go -advanced-) back to IE11...

    - - - - - - - - - -

    Ha, now I found it... the PC4-M5 thread is my placeholder...
    http://www.print3dforum.com/showthre...ll=1#post44506

    - - - - - - - - - -

    Alright; let's proceed with the B2B cartridge spool install video...



    - - - - - - - - - -

    Thought this might be interesting for merging different colors...
    A means to break the bowden tube in the middle somewhere.

    b2bsplice.PNG

    B2B Splice

    - - - - - - - - - -

    c1.PNG

    This should be fun....

    - - - - - - - - - -

    The splice already has the sidewalks. I printed a pair on opposite sides of the plate using ABS and they came out very nice.

    The plug in the bore just pushes out. This is a support for the split ring shoulder.

    I should see if I can use this is an alternative location for changing filament (hot-pull). If it gets stuck, it is a lot easier to fix on the fly.

    The concept behind the splice is to maintain alignment between the tubes. This should allow you to selectively "stock" the tube with short segments of different colors. If it doesn't catch and rip the 42mm long tube in the nozzle housing, life is good!

    - - - - - - - - - -

    Okay, the splice is a hit on one count and a miss on another.

    No matter what, removing the cartridge for filament changes is required.
    This is just a fact of life until you go through the trouble to hack a jog-driver to the motor as a service mode.

    Fortunately, that isn't a problem. I found out the Cartridge saver doubles as a push-button cartridge release.

    I was re-using a white PLA filament to push 1/2 meter segments of filament preloaded in the bowden tube.
    Remember I was running cooling towers so this is the result...

    d1.PNG

    I'm ecstatic! This has been my goal for a long time. The ability to reliably and easily change filament at the tip.
    So I tried my bowden path torture test. I ran a pigtail B2B tube with the splice just to have a little extra length on the bowden tube.
    This tube I preloaded with about 5 or 6 ~3" chunks of different colors of PLA.
    "Rewound" the white pusher filament in the cartridge and started the print. The bowden tube held about 12 minutes worth of PLA before the pusher stock would reach the nozzle. The attempt here is to make sure the pusher stock does not run into the nozzle. The nozzle is always primed this way. However, a pull of the filament while hot from the nozzle of the previous color can get you a fairly precise color change mid-print.

    d2.PNG

    And for the big question; Did the 42mm tube survive? Didn't even look phased!

    d3.PNG

    The splice isn't necessary for filament changes. The disconnect at the bulkhead is more than useful for this "merging" process.
    But I will say that having the small pigtail attached to the nozzle housing for a filament pull is working perfectly.
    Adding the pigtail just adds to the "no-touch" nozzle. It is more part of the machine rather than part of the cartridge or a dongle device.

    This system just rocks!

    Coming up... a simpler grooving tool.

    - - - - - - - - - -

    I knew I could come up with something simpler.

    This works great. Same knob as the other tool. This is where the screwdriver thingy goes, Alex.

    a1.PNG

    Slow-speed carve is all it takes. Could even go manual if you really needed too.

    - - - - - - - - - -

    Interesting color change observation.

    I swapped out natural, red, yellow, and back to natural in this print...
    sm4.PNG

    With PLA, these colors purge very quickly to the new color.
    However, ABS seems to linger a while longer.
    Notice the bleed from red to yellow... and again from yellow to natural.

    I find this interesting and thought it worthy of a quick note.

    - - - - - - - - - -

    And a new finding... Confirmed!

    We began hacking the cartridges quite some time ago.
    We are convinced that the weak tubing on the 3DS carts are garbage! And we are not wrong.
    It is in fact the secondary failure mode for the cartridges after the barrel nuts.

    But what does the stock bowden tube do, and why is it different from the CubePro?

    And why am I asking this question? Actually, it is a question I've had for a long time.
    It is you who should be asking this question. And now I have an answer for you.

    What I am getting at is that we are undoing a lot of what code developers tweaked to the nuts for the Cube3.

    What's wrong?

    Retraction parameters!
    ...and why they are all different, yet implemented the same.

    Problem statement:
    When the Cube printers have a local retraction system, meaning retraction at the hot end (no bowden tube), the system is pretty much "hardwired" or rigid.
    There is squishy anything between the nozzle and the feed mechanism. This is the CubePro and several others.
    When the Cube printers have a remote retraction system, meaning a significant length of bowden tube between the hot end and feed mechanism, the system is less rigid.
    Implementation of the bowden tube introduces "stretch" in the Cube3. And 3DS has compensated this with very simple yet specific driver defaults.
    This stretch is as much 8mm in the stock cartridge. I suspect that a 1/2 meter 4mm OD bowden tube may stretch <1mm under the same conditions.

    For those who have looked at the code and gained some understanding of G-Code and BFB, here is my finding:

    The BFB commands that manage retraction are M227 and M228.
    They have a very constant command line for this action when not associated to a general purge.

    M227 is the retraction/purge setting. It has up to 4 variables. M227 Pnnn Snnn Gnnn and Fnnn
    M228 is a retraction disable command with 2 variables M228 P0 Snnn

    These commands are sprinkled throughout the cube printing files regardless of printer model.

    What did I find?

    The default parameter that is written to the print file is all hard-coded in the drivers these days.
    No means to use a 3DS or MS provided software to set these values. This means intervention to change the information.

    I found the CubePro uses: (n denotes any number)
    M227 P250 S250 Gnnn Fnnn
    M228 P0 S250

    Ekocycle uses:
    M227 P375 S375 Gnnn Fnnn
    M228 P0 S375

    Cube3 uses:
    M227 P450 S450 Gnnn Fnnn
    M228 P0 S450


    (still working the meaning of Gnnn)

    That 450 value in the Cube3 is the amount of retraction in stepper counts.
    There is a purge element, and I am assuming this must remain the same as the retract as noted in all the reference cases.
    Furthermore, I've tested changing these specific command lines replacing 450 with 225.
    This action made for a near perfect print from a file that was horrible using the native format. This case used a non-3DS ABS filament.

    I found this issue when I looked at a single layer of a print (the tube above).
    I noticed that it had air bubbles (little holes in the print), but not consistently.
    The bubbles were only at the seam both the forward "resume" and the reverse for the inner skin.
    The rest of the sheet was near perfect! So the problem had to lie in the process.
    Sure enough, changing the retraction setting removed all the bubbles!

    That simply means that the retraction went so far as to draw air into the hot end and did not purge out fast enough.
    Somehow 3DS ABS filament is not quiet as sensitive to this.
    Yet, the print itself does suffer with "blobs" due to the trapped air expansion.
    That is the -only- reason for "drool" in a fine tuned machine.

    Bottom line:
    We improved the tool with a reliable bowden solution. We did not (and cannot*) optimize the slicer.

    I've suggested loading Axon from 3DS legacy downloads. I suspect a solution will come from that app instead.
    Anyone interested in exploring this route for a slicer?
    I suspect Simply3D can also build an appropriate profile with a generic header and footer.
    The command set is so small it should be pretty straight forward.
    Considering I am going for a true BFB capable printer so i can use Simply3D, I will see about this also at a later time.

    - - - - - - - - - -

    Now's the time to ask "what the hell is that!"...

    a.PNGb.PNG

    And the only answer I can give you right now is "prototype".

    But this is an attempt to provide a solution for the last remaining Cube3 niggle with regard to bowden tube conversion of your Cube3 printer.

    - - - - - - - - - -

    The first test of the tension compensation device (TCD) failed pretty spectacularly.

    Interesting finding however.

    So the previous images are for a device that, when spring-loaded, will act like the thinner bowden tube normally provided on the Cube3 cartridges.
    I am convinced that our more robust bowden system is actually detrimental to prints using the default slicer.

    The backstory is the MakeShaper ABS filament I was testing. Not only did I need to reduce the temperature for their ABS filament, I needed to address the excessive retraction. The retraction of the ABS filament would introduce air into the nozzle which would come out of the nozzle as "plosives"... a hole blown into the current trace being lain.

    The modification of the Cube3 file could easily "correct" the retracts but the idea of modifying every freakin' file for printing makes me gag.

    Therefore, with the knowledge gained using the MakeShaper filament and the understanding from over a year of development, I decided to give this TCD a try.

    Sure enough, with a hair tie as a spring, I got decent action from the device. However, it is obviously insufficient spring rate. it would accumulate length meaning it would stop bottoming out when the tension was released, meaning it always maintained tension after a while making some huge blobs.

    So now I understand the delicate balance that 3D Systems dealt with when it came to compensating the elasticity of their thin bowden tube.

    Overall, this setback won't deter me. I will need to characterize the original bowden tube to understand the proper settings for the device.
    Attached Images Attached Images
    Attached Files Attached Files
    Last edited by TommyDee; 08-20-2018 at 05:23 AM.

  10. Likes bolsoncerrado, bobskinner liked this post
  11. #30
    Regular 3D Printer
    Join Date
    Oct 2018
    Posts
    50
    Post Thanks / Like
    I figured this would be the spot for this. Here is what I spent most of my day on (besides one minor unplanned distraction to print a ring thanks to TommDee...)
    b2b1.jpgb2b2.jpg

    Thanks for your hard work in designing these Tommy. They printed out great on the EkoCycle in beautiful PETG. Only very minor cleanup was needed.

    The red set is currently in use but I intend to mix and match the blue and red as soon as I am able.

  12. Thanks bolsoncerrado thanked for this post
    Likes TommyDee liked this post
 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •