Update on Draken improvement based on feedback from ultimate backers

Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
edited August 2015 in 3DFacture Announcements

We appreciate all the feedback from the ultimate backers and are working hard to assure best quality of Draken for the next batch.   Here is a summary of recent progress and improvements:


1.) Z Axis Wobble

 We have come up a solution for Z
axis wobble issue.  A new two rail linear stage system
has been designed and is being manufactured. 
See attached photo of the new system.  
The stable two rail system is very stable without any wobble in our
test. 

https://drive.google.com/file/d/0Bx-KevVanX-hZzNXS2Q5R1pMUXM/view?usp=sharing

https://drive.google.com/file/d/0Bx-KevVanX-hZkdqVVZLQU5sdWM/view?usp=sharing

https://drive.google.com/file/d/0Bx-KevVanX-hUWduQUZKUWpXRlE/view?usp=sharing

https://drive.google.com/file/d/0Bx-KevVanX-hcm5VZFljc1JNWFk/view?usp=sharing

https://drive.google.com/file/d/0Bx-KevVanX-hZkNOTEtmZFFidzg/view?usp=sharing

All the second batch will be
shipped with the newly design linear stage. 
Any ultimate backers who experience
wobble issue can request to get a new linear stage (we will provide
instructions on self-installation).

2.) Build Plate screw tightness

We have designed a disk-shaped
stainless steel insert to be placed between the build plate and ball joint so
that the screw could be fully tightened (photo to be provided later).


3.) Resin tank spalling
 

The small cracks on resin tank
acrylic is found to be related to solvent in Makerjuice Resin.  These cracks are cosmetic and don’t influence
the function of the tank.   If you are
concerned about the cosmetic condition, please use other resins that don’t have
solvent such as B9 resin. We will test more resins and let everyone know which
resins don’t generate cracks on acrylic. 
Meanwhile, we will find out more acrylic to see which one is resistant
to solvent.

Other improvement.

We will use a new wiring system for
power inlets and board wiring.  All wire ends will be covered by
plastic socket to avoid any exposure/short-circuit.  Current users are suggested to use insulation tape to cover all power inlet sockets.


4.) Shipping schedule

There is at least 2-3 weeks delay in
shipping due to this change in linear stage. 
This means no shipping at all in this 3 week time-frame, so please don’t
send us emails on checking shipping status during this period.  Our next round of shipping is expected to
happen around end of Aug or early Sep.



5.) 3D Facture Resin

We have made a stable version of our
own resin and get it tested.  Will be
able to provide samples soon and move to mass production.


6.) 3D Facture Software Release

We are developing a Java based
software that can cross different platforms.  The progress is delayed due to the busy
hardware production and shipping recently. The overall framework is almost done
but need find tuning and testing.  It
will be released for all backer once finalized.

«13

Comments

  • ThijsThijs Member, Backers
    Thanks a ton for this update and the good work you put in
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    That is one beefy linear stage! I am looking forward to receiving one of them to install in my Draken.

    Ron
  • David_BarryDavid_Barry Member, Moderator, Backers, Ultimate Backer
    Thank you for update
    Hudson Valley, NY Draken Ultimate Package Arrived 7-6-2015
  • satoersatoer Member, Backers
    What kind of bearings are on that rail? Are these nylon sliders or real ball bearings?
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    Are you asking about the guide rods? My first guess based on the photo is some kind of teflon/nylon/brass? slider.

    Not sure how using ball bearings on that part would work due to the sliding action and the smooth rod. I guess the balls would be captive in place and rolling/sliding along the rod? You wouldn't have nearly as much metal contact to provide the stability we need to keep the Z wobble in check. :-?
    Ron
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    Good links. Thanks for posting them!
    Ron
  • chizchiz Member, Backers
    Thanks Andrew for these updates and for the great work enhancing and resolving the issues!

    Thanks as well go to the Ultimate Backers for taking the time in reporting back all these issues.
  • David_BarryDavid_Barry Member, Moderator, Backers, Ultimate Backer
    Where is the top end stop switch? Have you considered adding a bottom end stop switch?
    Hudson Valley, NY Draken Ultimate Package Arrived 7-6-2015
  • EdwardEdward Member, Backers, Ultimate Backer
    Great news! Thanks for the effort to make things right and thanks the ultimate backers for their feedback.

  • satoersatoer Member, Backers
    edited August 2015
    rkundla said:

    Are you asking about the guide rods? My first guess based on the photo is some kind of teflon/nylon/brass? slider.


    Not sure how using ball bearings on that part would work due to the sliding action and the smooth rod. I guess the balls would be captive in place and rolling/sliding along the rod? You wouldn't have nearly as much metal contact to provide the stability we need to keep the Z wobble in check. :-?
    image
    Left: old Right: new

    The previous version has real Hiwin-like (or are these real Hiwins?) linear ball bearing sliders, but I’m not sure the new one has a real ball bearing slider or some kind of Teflon / nylon low friction slider.

    Look at this video:


    The cylindrical ball bearing sliders are not as good as these rails (although fake-hiwins do have less quality). But since the 3D facture uses 2 it might be more rigid. But… if these new bearing aren’t the ball-bearing type it will eventually introduce backlash resulting in Z wobble all over again.

    Still, aren’t they solving the wrong problem by fixing it with a more rigid slider? Isn’t the core problem a bad quality ballscrew or spindle. These things shouldn’t wobble.


  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    Nice video! Does alot to explain how these items work.

    When I took my linear stage apart, I only saw a single loop of ball bearings on each side of the rail. Even then, one of the loops was 'stiff' and didn't rotate properly. The other loop was very smooth and I could take my finger and spin the bearings around quite easily.

    The tolerance between the platform and the rail in the vertical axis seems like a not ideal use of the product. I think it is better suited for horizontal applications, especially after watching that video. ;-)

    Based on Andrew's comments, I think the ball screw was deflecting under the load since it was doing double duty. It was the guiding element as well as the driving element. The rail wasn't sturdy enough to contain the motion in X-Y relative to Z (backlash in the video above). Using two guide rods with hopefully multi ball bearing loops (on four sides), they will keep X & Y in check while the ball screw only has to worry about driving the platform up and down.

    I didn't dissect the linear stage to the point of checking the ball screw for flatness over the length. I also didn't get a chance to see if the stepper motor to ball screw interface was cockeyed. Those would have been the big tests to see if the screw was another contributor to the problem or not.

    Ron
  • kniborknibor Member, Backers
    @rkundla the spindle is squized into a bearing of the steppermotor, so the spindle motor connection should be fine. As far as the new guiding rails, even if it isn't a ball bearing by default than you can easily change this to ball bearing ;)! Just losen up the 4 screws and replace it :)! But you have huge difference in quality and price in these lineair ball bearings. You can buy chinese for 0,50€ or ina fag bearings starting at 20€ a piece or so (depending on the size but I guess the guide rails is 10mm in diameter)
  • Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
    The problem lies on the design of the original linear stage, where both the ball screw and the rail are rigid and lead to wobble under both constraints. The new design uses the two cylinders for vertical guidance while th ball screw provide pushing force. Except for the ballscrew and nut, all other parts are made by high precision CNC machines.
  • Joseph_OsbornJoseph_Osborn Member, Backers
    ... Which leads to the question of why the linear stages assembly that was proven to work in the prototype machine was changed to a new design and not rigorously tested to ensure quality.

    But that's water under the bridge now and as long as the new design tests well, the new batch of machines will be wobble-free.
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    @Andrew_3Dfacture, any update on the timeline of when we can see those replacement stages?

    I've pretty much stopped work on the Draken until this is resolved.
    Ron
  • JHSJHS Member, Backers
    Thanks for the updates Andrew and team. It's great to see you guys putting in the work to ensure the quality of your machines. The development images are certainly making me eager to get my unit and begin testing and rapid prototyping parts for my university automotive design graduation project!

    4.) Shipping schedule

    There is at least 2-3 weeks delay in
    shipping due to this change in linear stage. 
    This means no shipping at all in this 3 week time-frame, so please don’t
    send us emails on checking shipping status during this period.  Our next round of shipping is expected to
    happen around end of Aug or early Sep.

    If this is true, I hope the shipping company won't take too long + has no trouble getting past customs as it will be getting dangerously close to the deadlines I have to meet in order to deliver a model for the graduation show!  :-SS

    Thanks again and keep us posted!
  • Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
    The new linear stage will be ready in the week of Aug. 24th.  We will resume shipping once the new linear stage is received.
  • Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
    We received some of the linear stages today, but sadly the bearing is in wrong size, kind of a little big to fit on the cylinder linear guide.
    https://drive.google.com/file/d/0Bx-KevVanX-hbV9qMzNhR2NHLVk/view?usp=sharing
    https://drive.google.com/file/d/0Bx-KevVanX-hc2locV9FRTQ1MmM/view?usp=sharing

    We will have to order some new bearing to put on them, likely to get them by end of this week.
    Thanks for the patience.
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    edited August 2015
    @Andrew_3Dfacture, will you also be able to ship out some of those shims for the build plates to help reduce the chance of movement between the stem and the plate?

    Also, are you shipping from China or the US? How long can we expect shipping to take if from China?
    Ron
  • Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
    Sure, we produced lots of shims for the build plates for the second batch. Will include a shim together with the linear stage. 

    It will be shipped from China directly and will take the same time as last round (Fedex is a week or so?).   We might consider ship to US site and re-distribute too...might not be that effective and take a little longer though since we only have a few US units from last round.
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    Sounds good - looking forward to the updated parts!
    Ron
  • CHJEWELCHJEWEL Member, Backers
    ??? When ship out printers for April
  • Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
    The overall printer assembly has been done, we are just waiting for the new linear stage to arrive sometime next week.
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    @Andrew_3Dfacture, any updates on the linear stage arrival?
    Ron
  • CHJEWELCHJEWEL Member, Backers
    Where Where the mews?
  • JHSJHS Member, Backers
    edited September 2015
    Weren't we promised shipment "sometime this week"? 


    8-|
  • Andrew_3DfactureAndrew_3Dfacture Member, Moderator, Backers
    Sorry about the delay.  The bearing will arrive next Monday and we will
    finish the linear stage assembly in 1-2 weeks, thus expect to start
    shipping in 2 weeks. For the ultimate backers, we should be able to start shipping out the new linear stage as soon as we get the linear stage assembled.
    For
    the ultimate backers, we should be able to start shipping out the new
    linear stage as soon as we get the linear stage assembled. - See more
    at:
    http://3dfacture.com/help/index.php?p=/discussion/212/james-3dfacture-user-account-has-been-deleted-wtf#latest
    For
    the ultimate backers, we should be able to start shipping out the new
    linear stage as soon as we get the linear stage assembled. - See more
    at:
    http://3dfacture.com/help/index.php?p=/discussion/212/james-3dfacture-user-account-has-been-deleted-wtf#latest
  • satoersatoer Member, Backers
    edited September 2015

    Quote from the Kickstarter page

    The bearing arrives at the linear stage manufacturer today...but it is slightly smaller than expected. Will need to some post-processing to get it work. The manufacture expect to get them fixed by Friday, thus we decide to delay the update post to Friday when shipping schedule will be clear.

    How can you rework a bearing? Is the bearing too small for the shaft, or is the bearing carrier to wide and need the bearing to be shimmed up?
  • rkundlarkundla Member, Moderator, Backers, Ultimate Backer
    satoer said:


    Quote from the Kickstarter page

    The bearing arrives at the linear stage manufacturer today...but it is slightly smaller than expected. Will need to some post-processing to get it work. The manufacture expect to get them fixed by Friday, thus we decide to delay the update post to Friday when shipping schedule will be clear.

    How can you rework a bearing? Is the bearing too small for the shaft, or is the bearing carrier to wide and need the bearing to be shimmed up?

    The second is probably the issue. The holes in the carrier were probably cut too large or the bearing with tolerences figured into it were too small diameter-wise. :-P
    Ron
Sign In or Register to comment.