Topic: Three weeks with my Robox

Home Forums Chatter Three weeks with my Robox

This topic contains 39 replies, has 12 voices, and was last updated by Profile photo of Rob Rob 1 month, 2 weeks ago.

Viewing 12 posts - 1 through 12 (of 40 total)
  • Author
    Posts
  • #8580

    MiffMaster @miffmaster

    This is what I have found out so far.

    TLDR; I’ll discuss first prints, the print bed, leveling, nozzle offset, precision, bridges and filament. If any of this sound interesting, locate the topic bellow.

    - Short background so you know where I am coming from.

    I got in to 3D-printing about 2-3 years ago. Got my self a kit for a Rostock. All parts included, assembly required! This was before automatic bed leveling/calibration for the delta bots. All that had to be done by hand (tweaking values in the firmware). I am sure many of you are familiar with this process. Patience being the name of the game ;) And once the bot was calibrated. It was time to dive in to the millions of options in Slic3r. The learning curve sure was steep.

    Anyhow, this bot is printing very reliably now. I always print in ABS, as the PLA always get stuck.

    After all this, I consider myself reasonably experienced.

    - The Robox finally arrived. :D

    After the initial setup and getting to know AutoMaker and the Robox. It was time to test what it could do. Loaded my favorite calibration piece and hit print. The result was terrible!

    Spent some time messing around with the few options available in AutoMaker. Switch material to ABS. Prints got a little better. Eventually switched back to PLA. Was rewarded with my first clogged nozzle. Fixed it and decide to only use PLA from then on (this has been a good decision).

    Beginning to get comfortable with the Robox. The prints (now always in PLA) are barely acceptable, but slowly improving. It was time to separate the issues and tackle them one at a time.

    - Prints coming loose.

    The main issue so far has been with the prints not sticking to the bed. And the further from the center, the worse the problem got. This translates to the bed not being flat. Which could be seen with the naked eye. Flipped the bed over and problem solved.
    Well, not really. The bed is flat now alright. But now it is only possible to print in the inner left corner. Anywhere else and it prints in the air.

    - Leveling the bed, and calibrating nozzle height.

    Performed these procedures and things got better. Prints now stick to all of the bed. But for some reason the process is not perfect. The first few layers are to close to the bed. It was easy to tell bu printing a cube. The base would be wider (smooshed out?) than the walls/body of the cube. So some manual adjustment of the nozzle height was required to get it just right.

    - Nozzle offset

    This issue presents itself when printing in “normal”, where the exterior is done with the fine nozzle, and the infill with the large nozzle. There would be a good connection between the exterior and the infill on say the left side, but on the right side there would be a small gap of 0,3-0,6 mm. Found a setting for this and adjusted it until there was a good connection in all directions.

    - Precision/accuracy

    Eventually I made some larger test pieces to get a feel for the precision. My prints come out being too small.

    X = 99%, Y = 99.5%, Z = 100%

    As you see, all axes are different. And AutoMaker only offers a combined scaling option. So far I have not found a practical/convenient fix for this. External software to re-scale the *.stl is not convenient.

    I believe that the correct way to go about this, would be to have a “pre-scalar” for each axis in automaker. Then the reel should hold a material specific “shrinkage” scalar that only affect X & Y. Because X & Y prints quickly and does not have time to cool. While Z is super slow, and has lots of time to cool during the print.

    - Bridges

    This one is annoying. When printing in “fine”, during the first (and most important) layer of a bridge, I estimate that it squirts out about 300%-500% the necessary material to span the gap. Needless to say, the bridging result is awful. However, in “draft”, the bridges are close to spot on.

    Is there a way to adjust the Slic3r settings for the different profiles?

    - Filament

    Switching filament. Now, I have spent some time thinking about this. And I only see trouble, no matter how you do it, and despite purging. There always seem to be some little residue of the previous material left in the head.

    So let us say you go from PLA to ABS. The PLA residue will eventually harden, come lose and clog up a nozzle.

    Then you go the other way, ABS to PLA. Some ABS residue will eventually come lose. Because of the lower temperature, it will not be soft enough and clog up a nozzle. So either way it will likely cause clogs.

    An observation. I see several posts here about how to store the filament in plastic bags, with silica and out of sunlight. PLA in particular, as it is supposed to be the most sensitive.
    Now, I am not saying that these instructions are wrong. However, I do feel that they are wildly exaggerated. I have a roll of PLA that I bought at the same time as my first printer. Almost 3 years ago… It did not like this material. The roll has been sitting on a shelf, out in the open this whole time. I live just a stone throw from the sea, so humidity is always high (though not like tropical, that is moist, not humid). Today I decided to test this material. Used the same profile as the Polar White PLA. No problem what so ever. And by that I mean it prints the same as the new Polar White PLA that came with the printer.

    One more thing.

    At one point I did some manual purgeing. While doing this I raised the print head to the top position. When satisfied, I started a print. The procedure when starting a print is to home all the axes. The Robox begin by homing the x-axis. This resulted in a very ugly noise as the print head collided with the screw holding the lid, unable to reach the end stop switch. Had to abort, manually lower the head and restart the print.

    At this point I am fairly happy with the Robox. There is still work to be done. But I see no unsolvable problems.

    That is all for now. Though I look forward to hearing your thoughts and suggestions.

    • This topic was modified 2 months ago by  MiffMaster. Reason: Formatting/readabillity
    #8591
    Profile photo of
    Anonymous

    @miffmaster - very interesting and informative post.

    You cover a fair bit that has already been discussed on here, so I won’t go into that.

    Interesting stats on the shrinkage - bit surprised to see the horizontal rates differ, can’t think why that would be the case, unless it’s something to do with the direction of the fan.

    Although I got the printer to print in ABS, I must admit to getting much better results and reliability with PLA. The surfaces are cleaner and the parts are more accurate. I do have a problem with big prints lifting off the bed towards the end of the print. Not solved this yet (tempted to try pritt stick), but for small prints I’ve found just raising the bed temperature, particularly for the initial layer works well.

    When switching filament, PLA -> ABS often seems to result in a piece of PLA getting stuck in there for ages, sometimes appearing after several prints. I now always purge before every print if I’ve switched recently. I don’t know how long you have to do this to feel confident, but I’ve had PLA sputter out of the fine nozzle after half a dozen (mostly bad / aborted) prints using ABS.

    Nozzle offsets - this is interesting, I’ve not explored the link between “gaps” and the nozzle offsets. I assumed most of this was down to incorrect fill levels for prints with certain wall thickness. When I get my printer back I’ll investigate this I think.

    Bridges - yes, my attempt at printing le FabShop’s mini (at 50%) resulted in rough bridges using the fine profile in PLA, but I think this is just a software issue that can be sorted:

    #8594
    Profile photo of Don Colley
    Don Colley @doncolley

    I have been using my Robox for a few days now and thought that I should show a few of my prints. They are all done in normal mode and with ABS I am very pleased with the results with no adjustments, I find that if I need to be accurate with size I print at 102% the Dremel attachment screwed strait on and is a good tight fit.The knife block was designed in Tinker cad the rest from Thingiverse. Very happy chappy up to now.

    Attachments:
    You must be logged in to view attached files.
    #8597
    Profile photo of
    Anonymous

    Nice prints, great surface finish @doncolley

    #8598

    @miffmaster Enjoyed your post. Strongly second your comment about PLA storage. I’ve had a heavily used Ulitmaker 1 for 18 months or so and never noticed anything I could attribute to moisture absorption (I live in Sydney, well known for high humidity in summer time).

    My Robox is wending its expensive way through Australian customs….

    I hope that the Robox head is successful but I have my doubts. It is very complicated and the single input splitting into two is a big worry. I’ve only used PLA in my Ultimaker and had enough problems with the original head (with no moving parts) to change it for a E3D head (also with no moving parts). For 4 or so months I had to periodically drill through the head to clear trapped material. I think caused by transitions that pointed upstream rather than downstream. Now it has settled down and I haven’t had to do that for maybe 6 months. So the Robox head without, as far as I can see, any way of readily mechanically cleaning it is a worry. CEL may have to invent a purging material, capable of scouring out residual ABS or PLA.

    I was disappointed to read of the flexibility of the PEI sheet. The Ultimaker I has an Acrylic bed that one covers with Scotch Blue Printers tape cleaned with Isoproply Alcohol. Over time this warped and I’ve mounted glass on top of the Acrylic. Much more satisfactory. It is important to have a flat surface.

    I admire the ambition of CEL and that is mostly why I ordered a Robox. But ambition can sometimes over extend…

    Mike.

    #8599
    Profile photo of gid
    gid @gid

    As far as the bed’s concerned, the PEI sheet is held against a solid metal plate. The bend might actually help to some degree as the clips hold down the edges while the downward bowing presses flat against the metal. If CEL can implement a more comprehensive height mapping of the bed than the current four-point test, it should be irrelevant anyway as the firmware should be able to compensate for macroscopic height changes anyway.

    The sheet’s flexibility is also beneficial when a printed part is stuck to the bed: while cooling usually releases the part, it’s handy to be able to take out the bed and flex it if all else fails… it’s better than scraping the bed.

    Tom Gidden -- Bristol, UK -- New Roboxer? Check out the wiki, and add yourself to the map! http://roboxing.com/user_locations
    #8605
    Profile photo of Jack
    Jack @jack-strong
    My Robox is a Green Kickstarter Beta

    @doncolley - Got links to the models made ?

    What were the parameters, infil, brim etc ?

    They look stunning!

    #8610
    Profile photo of Don Colley
    Don Colley @doncolley

    @jack-strong

    I just leave well alone, normal quality, 3 brims and 50% infill and push the button. I bought it to plug and play and that’s what I do I dont know enough to fiddle with the other settings and I have a stack of ABS from a friends defunced printer so that’s what I use, I print at 102% to allow for shrinkage and it seems to work for me.

    The knife holder was designed in tinker cad and the other things came from thingiverse checked with meshmixer I will try to attach the stl for the knife holder but I would modify the size of the base if you can to make it a little more stable

    #8614

    Clicky @clicky

    Thanks for the cutter’s stand. Is it updated or you’re going to update it and post it again?

    Could you share links to thingiverse objects on the picture as well?

    #8620

    Tanel Vilol @tanelv

    Good stuff! You mentioned in the Nozzle Offset part you changed settings to avoid that gap between layers. What were these settings?

    Thanks, Tanel

    #8626
    Profile photo of Don Colley
    Don Colley @doncolley

    I was not going to update it but you can load the stl file into tinkercad quite easily and work on it with no trouble

    #8627

    MiffMaster @miffmaster

    OK, so I do not have the printer at hand. It is at my job (got one of those nice jobs which overlap with ones hobbies/interests, without taking the fun out of them).

    @biscuitlad, @tanelv

    About the nozzle offset. Tomorrow I will provide settings and arrange some pictures, as pictures are always nice (felt a bit ashamed writing such a long post without any).

    As for the difference in shrinkage. My suspicion is that the X-belt is not as tight as the Y-belt. The variation is only 0.5% so not a dissaster. Though it might be noticeable in pieces that should fit into each other.

    I will also atempt to get some video demonstrating the bridging problem, and the difference in “draft”, “normal” and “fine” bridges. I am sure this is only a Slic3r issue (bridge feed rate). However I have not seen it in the AutoMaker interface. And I have sort of made it a point not no mess around “under the hood”, but only use the official settings/tools provided by CEL, simply to see how this would work as a consumer printer. So far, the Robox is obviously not the PnP printer adverticed. But honestly, I would have been surpriced if it was. Considering my previous experience with countless hours of tweaking firmware/Slic3r settings for my other printer. It is not too far away.

     

    @mikeg, @gid, I must admit that when I first noticed the curve in the PEI-bed, disappointment was my first feeling. Though I am not the one who gives up easily. After turning it around and instead having it press “into” the the metal support underneath (as Gid suggests), I feel the heat transfer is more efficient. And after re-calibrating the nozzle height (with manual fine tuning) it is now able to lay down a text book perfect first layer over the entire print bed. No matter which nozzle is used.

    Oh, and as Gid say about the bendy PEI-bed being useful for removing large prints. This is something that never struck my mind, but I have to agree. I have had some large ABS-prints on Kpaton/glass bed on my other printer, which were just about impossible to remove.

Viewing 12 posts - 1 through 12 (of 40 total)

You must be logged in to reply to this topic.

Subscribe now to be kept up-to date with the latest Robox® news. Subscribe Now