Friday, January 25, 2013

Warning to Curious Cube #1 Users

While the new Cube Client can be used with both First-Gen Cubes and Next-Gen Cubes, and some Next-Gen features might safely be experimented with on First-Gen Cubes, using PLA is NOT one of them.

I am NOT planning, under any circumstances, to test printing PLA in my First-Gen Cube until someone at 3D Systems confirms that it is OK.  And, I do not think that is likely.

Frankly,  I can't think of any good reason for doing so.  ABS is stronger than PLA and we have a heated bed.

No one at 3D Systems has confirmed my theory; but, my experience with PLA in a RepRap machine tells me that not only will PLA not work well at the First-Cube's temperatures; but, may damage the feed mechanism.  That potential just does not seem worth the risk to me, for so little gain.

Printing temperature for ABS is around 240C to  248C.  Printing temperature for PLA is begins at 190C and typically averages around 210C.  That's quite a big differential.

Here's my theory of the probable outcome...

First, remember that the filament is being pushed down a shaft before being melted at the tip.  In order to work properly the filament must be solid at the entry of the shaft to be heated and only reach the right temperature at or near the tip.  With ABS, the shaft at the entry point is just slightly lower than the melting point of the ABS itself.  So, the feed gear pushing on the filament is able to push the filament some distance into the shaft before it reaches the melting temperature at the tip.

Since the optimal melting temperature for printing ABS is quite a bit higher than the optimal temperature for PLA, running PLA through the head of the First-Gen Cube is likely to heat the PLA to the melting point higher up in the shaft where the melted plastic could back up into the threading mechanism. 

So while the entry to the shaft might be below the melting point of ABS it is likely ABOVE the melting point of PLA.  The PLA material will be too molten to be able to be pushed deeply into the shaft and I expect the molten PLA will flush back into the housing and then cool, causing more than a simple clog.  It could cause actual damage to the head feeding mechanism.

I'm aware that I may be completely off base.  But, I'm not willing to risk it.  I'll stick with the ABS for which the First-Gen Cube was designed.


  1. Tom: all things being equal, a problem between switching from ABS to PLA is that any residue of ABS that remains in the barrel will not melt at PLA temperatures. This leads to blockage of the tip. Note that the Cube 2 comes with some sort of tool to clean the extruder from jams. So while it is interesting to see that they are including new materials (and PLA is less likely to warp) I don't need to use PLA on my Cube.

    I would also think that the extruder temperature is a matter of software so the extruder in the Cube 1 should be capable of being controlled by the job file and not just an internal preset. But that is guesswork too on my part! - Eric

  2. I hadn't thought about the fact that the reason for the strong warning might be residual ABS that would not melt. Nice observation! :)

    This kind of contribution is makes makes this blog beneficial to all of us!

    1. Here's one technique I've found that helps with clearing (or diagnosing) possible nozzle clogs, on the original cube at least:

      After using the cleaning tool shown in the manual (which is larger than the actual nozzle opening), you can run a .015" music wire (first suggested by Tom) down through the print head from the top filament opening, and with a little fiddling it finds the nozzle exit hole and will come out of the nozzle. When you see the music wire sticking out of the nozzle, you know for sure there's no clog.

      I was told by the Cube support folks that in a pinch you can often use a large paper clip (straightened) in place of their declogging tool. I did that and it worked, but since a paper clip is plated steel, I worry about a little bit of the plating coming off and making a clog worse.

      Their own declogging tool isn't plated, so it's better to use that, I think.

  3. FYI, today I asked the Cube support folks if I could, say, create a file for the 2nd Generation Cube, using ABS and setting it to the "Solid" mode, then send that file to my original-generation Cube and have it print.

    They replied that that won't work. At least one reason is that the 2nd Gen file won't contain build plate heating instructions, and without the build plate being heated the original Cube won't print anything. I expect there are other issues also. Sigh.

    Oh, well, if you don't ask you don't know.

    I did ask them to consider adding the "Solid" mode to the original Cube, and they said they'll forward the request to their software folks.

    But I expect the original Cube isn't where they're going to be devoting much resources to, now that the 2nd Generation unit is out. That's completely sensible and reasonable. But, one can still hope!

  4. Actually. No that is not sensible or reasonable at all. Why can the 2nd gen cube print solid objects, hallow objects, and hallow objects with internal supports... yet the 1st gen cube cannot? I personally believe this is just a ploy to get people to upgrade their cubes. There is absolutely no reason why the 2nd gen cube can do these things while the 1st gen cannot.

    I emailed Cubify, and this was their response:

    "The Cube Software will automatically hollow out objects internally. The intent of this feature is to preserve filament and reduce build times."

    "The Cube software is designed to minimize the amount of material used in order help our customers save material."

    "The new build features for the second generation Cube have not been tested on the first generation Cube because it was not designed to be able to print in that fashion.

    We are not trying to neglect our first generation Cube owners. We are simply trying to improve on our product in any way we can. There are new developments happening daily, and we are excited to introduce these new features to our customers to give them new options."

    Why are we forced into using this proprietary software that forces 1st gen users to accept the hollow print or purchase their newer model? Why is the 1st gen cube designed to not print solid ojects, yet the 2nd gen is? Doesn't anyone with any kind of 3d printer knownledge know that, that is BS?

    I am willing to bet that more or less everything between the 1st gen and 2nd gen cubes are indentical between both models. The print platform and those mounting points are different obviously. The only thing I can think of on the print head being different is possibly the diameter of the nozzle (since it can print finer now), and possibly a change on the temp probe. I'd wager that beyond the print head diameter, print platform... the firmware is the only thing that is different.

    I completely regret buying my Cube in early December. I've had 3 print heads replaced, the whole cube replaced... it has easily spent more time being broken than being used. I've now had many parts fail, plastic cartridges fail, and really poor quality prints. Many of those problems would be solved by the ability to print solid objects and hallow objects as the 2nd Gen cube can. There is absolutely no reason why they couldn't add that into the software, as I'm sure all the same processes could be used... as Mike said.. they just need to add the 'heat up' code at the beginning of the file.

    1. AJ, my brand new 2.5 generation Cube is dead after 3 days. Day 1 printed fine. Day 2 started with "Filament Feed Errors". Day 3 the head wouldn't heat up. 3D systems (Andrew) wants me to record a 10 minute video showing the head not heating up..really? (10 minutes of "please wait" on the LCD? Coming right up). Anyway...I had purchased this for my company as a way of getting our feet wet for a CubeX system. I did NOT want to go with MakerBot because I feel like their pricing is reflecting popularity not capability. After this silliness (a 10 minute video of a "please wait" screen???) AND realizing their customer service is the same for the CubeX...I might have to think about this.

    2. Oh..and about my "Generation 2.5" comment. My print plate (glass) is a new 'magnet mounted' design with no external clips to hold it tight. This design has a lot of 'jiggle' to it but, when it worked, it seemed to print just fine.


      My brand new Cube had a glass plate that someone had glued the magnet on at about a 15* angle instead of straight. (see my blog for pics). Candace is shipping me a new one. (except my cube is now dead so it doesn't do me a dang bit of good).

      All in all...this is not developing well...and I've only had it 4 days now...(SIGH)...

    3. Wow! I'm sorry to hear that. But, I can assure you that most of the feedback I get is in praise of the Cubify support. I should be getting a new Cube in the next few days so I hope I can be of help to you.

      I'm wondering if a piece of filament is lodged in the head. Have you removed the filament and tried reloading it? And, do you have a way to monitor the head temperature? For purposes of testing for the blod, I use the Centech Infrared thermometer and they are available at hardware stores. That would allow you to document what you are seeing. Feel free to email me at the "request" address at the right side of this blog.

    4. Regarding printing solid objects in the 1st gen cube, I can verify that the file as meant for the 2nd gen cube will simply not print if solid. You can try, but it'll either error out shortly into the print, or you'll get a ball of filament.

      Now, this doesn't mean that the 1st gen can't do solid can. It's just fiddly. Take an stl of a cube. In your favorite 3d editing program, copy it, shrink it just a bit, reverse all the normals. Put it inside the first cube. Rinse and repeat until you're at small enough cubes that the whole lot is full. Print, and bam, solid cube. It may take a bit of practice to get the hang of how far apart they need to be, but printing solid objects is essential for a lot of what I do, so it was worth the trouble.

  5. It's not my job to defend 3D System's decisions. But, I will say that I do my best NOT to design solid objects or completely hollow objects. There are blog entries yet unwritten that explain and demonstrate some of the things I have been exploring that minimize the need for both. The first reason is that ABS DOES warp with huge solid areas. That is not a function of the printer. It is a function of the material.

    Now, as my primary design tool, MOI allows me to design from nature and nature tells us that a honeycomb is the ideal structure for lightness and strength. I'm not sure that Gen-Two Cube owners are going to be using the features that are missing at the moment for Gen-One users. It would be interesting to hear from them.

    1. To be fair. That is true. However, the Cube software decides when and how to hallow out an object for you. I have drawn thin walled items, and seriously the cube decides to hallow out a wall 0.75mm thick. It builds it as two completely separate walls, 0.25mm thick (1 single layer). Some objects I make are so small, only a 6-7 layers, that it hollows out a tiny bit in the middle and then doesn't fill out the whole shape... effectively ruining the part. I understand you don't want to go out and print a huge object solid... but we should still be given the chance to decide on things like wall thickness, and whether or not it is solid. As it is, we're stuck with buggy proprietary software that has the options built in... but won't give them to Cube 1 owners. Cubify wants us to buy the new cube, which is nearly identical to the Cube 1.

  6. I take your concerns very seriously. But, the solution (solid) that you think will solve the issue you are describing will not, in fact, solve the issue. It certainly is logical to assume that. But, that is not what "Solid Fill" means.

    What is actually happening has to do with nozzle size and the material itself. I created an array of shapes having walls from .1mm to .85mm. Using a microscope, I examined each one to see the condition of the wall. It is remarkable that it even tried to print the .1 wall. But, it printed every wall. Every wall, except .85mm, was created in two passes. The two passes started to separate, creating a gap at .70mm, .75mm and .80mm. At .85mm a third pass closed the gap.

    I plan to post the results as soon as possible.

    1. This is interesting to know, and I look forward to seeing the post. I will set the wall thickness at 0.85mm and give it a try. I think I also tried even thicker, but that also created two walls (2-3 passes thick)... with no support material between the two.

      Also when doing wider, it didn't bother drawing the 'floor' on the part either.

    2. You were right about 0.85mm. Sort of.

      At 0.85mm thick, it closes the gap with two passes.. not just 1. So basically it jammed 4 passes in the width of 0.85mm, where only 3 should go. It caused a bad surface finish with a very bloated looking part in places. Perhaps 0.95mm would be better? If it fits 4 passes in that width, it should be fine.

      I think at 1mm it starts to split them up into two again.

      I was in the middle of a 6 hour print on that part and had the cartridge jam and the print fail. Of course now there isn't enough plastic in that cartridge left to start over.

    3. Any time a cartridge has jammed it has been due to one of two reasons. Some early delivery cartridges were spooled wrong by their supplier and would tighten up and fail. Haven't seen that since my first cartridges.

      The other cause is letting the room temperature rise too high. This happened in the summer in the outside studio.

      By the way, I just took delivery on some new cartridges and they have gone to a new form factor that minimizes gluing the two halves together.

  7. I cannot emphasize enough how use a lot cost stereo microscope is in helping us learn to design to the best affect. A perfect tool for this is the C&A Scientific SMD-04. lists it for $62.10. A similar scope, at a similar price is the SE104-P. I don't know who makes it. But, it's easy to find on the web.

    It is well worth the investment.

  8. I purchased the Cube 3D 2nd Generation for 10 months. The print tip doesn't heat up it just turned me down during the printing process. I cannot contact them since it's a Sunday night. I sent an email but no reply. I live 80 miles from the Headquarter in SC so I drove all the way there early morning. The receptionist saying that "they don't do anything at the headquarter". She gave me a phone to call so I called and spoke to Drew and he saying that "my printer is out of 90 days warranty" so I have to purchase the complete PrintJet for $249 plus tax and shipping. I did alot of research before I purchased the Cubify 3D. On it website saying that "the PrintJet is $249 but rarely has to be replaced". But it gave up on me after 10 months and they won't replace it. It's a $1299 printer and break-down that easily which is crap.

  9. Write to me with the address in the Personalized Help section at the right of this blog.

  10. After I installed the 2.08 firmware the PrintJet stop heating. I talk to Gary about it and he said that I must of dropped the machine. Was he serious ? they want me to purchase a new Printjet for $249. Why waste my money. The machine is JUNK

  11. Sorry to hear that. But, I would look elsewhere, besides the firmware upgrade as the source of the problem. We have upgraded at least 9 machines without a problem.

    Check to see if filament has built up inside the cover of the print head. That would cause the head to overheat and shut off the head's heating cycle.

    I'm certainly willing to try to work through the problem with you. Simply email me at the address to the address found in the sidebar.

  12. This comment has been removed by the author.

    1. This comment has been removed by the author.