This topic contains 63 replies, has 12 voices, and was last updated by 
Omega64 2 years, 10 months ago.
-
Posts
-
Ok so really getting into the Robox now and although I still have a few minor issues I wanted to try to print in the magic 20 micron layer height.
So I thought the best way to do this was to find out how Ultimaker do it. However after a few hours of searching Ultimaker forums I have came to the conclusion that the general opinion among Ultimaker owners is that 20 microns is mostly a waste of time. Although it is possible the print time is increased so much in comparison to their “Ultra” prints of 40-60 microns it isn’t worth it, and also some features are actually printed much better at 40-60 microns.
So 40 microns is my target. I downloaded Cura and selected Ultimaker as my printer so I could “steal” their 40 micron print profile (this is a standard feature btw) but hit my first problem. I read somewhere on this forum that halving the speeds of the Robox Fine profile is a good place to start working toward lower layer heights. However the first thing I noticed about the Ultimaker 0.04mm profile is that the speeds are faster than even the Robox 0.3mm profile!!! I have attached a few speeds for comparison.
I am now very confused.
STEELMANS 3D PRINT.
3D Hub: www.3dhubs.com/service/steelmans
Website: www.steelmans3d.comI did some 20micron experiments a while back, never messed with the speed tho, just extrusion rates, I did notice that I had to lower it a lot to not get it to smear the edges. I think you would be better off to just go faster, 20microns is a very thin layer to put down and if it goes too slow I think it will just mess it up and heat up the layers beneath and make more of a mess, that is what happened to my experiments.
We have been dropping the print speed of Robox in our default profiles to concentrate on quality. You can generally double the speeds with ABS without any issues. Experiment with an increase to”Feed rate multiplier” in the filament profile, this will increase everything including the movement speeds.
I expect we will increase the defaults soon, we have experimented up to 4x speed with ABS with minor flaws appearing at these speeds.
Lower layer heights struggle with bridging, even tiny gaps are very difficult because of the small amount of material being extruded. Once you have a flaw in the print the small flow of material will struggle to repair it.
I would advise printing something tiny in your experiments, watching the plinth of a statue print for a whole day without appearing to change size is painful.
For official support please visit www.cel-robox.com/support/ and create a ticket@pete Not sure what direction to go in with this. I just printed an Owl model that took 3 hours 7 mins in “normal” mode. So I created a copy of the “normal” mode and used all the speeds from Ultimaker. The result was ok, not quite as good but did print in 1 hour 53 mins.
I understand the Robox is a very different machine but I would be looking to target the speeds Ultimaker use. The technology and innovations of the Robox combined with the speed and print quality of the Ultimaker would be a world beater.
The main thing I noticed about the Ultimaker profiles is that their speeds change very little (if at all) between “Ulti” (0.04mm) and “Fast” (0.15mm) modes. Also of interest to me was that 0.1mm layer height is the “Normal” mode on Ultimaker as opposed to being “Fine” on Robox. In fact their “Fast” mode is a lower layer height than the Robox “Normal” mode… and it is very very very fast!
STEELMANS 3D PRINT.
3D Hub: www.3dhubs.com/service/steelmans
Website: www.steelmans3d.comUltimaker’s head is lighter than Robox so in theory a higher speed movement speed will be possible.
I haven’t had any experience with Ultimaker profiles. Regular is small at mcdonalds, grande is large at starbucks.
Rather than editing speeds try editing the Filament Rate Multiplier. Leave everything else at defaults for the test.
-
This reply was modified 2 years, 11 months ago by
Pete.
For official support please visit www.cel-robox.com/support/ and create a ticket@pete LOL very funny Pete.
I wrongly assumed that everyone at CEL working on Robox would be very familiar with Ultimaker profiles.
The point I was making is that the Ultimaker has 4 standard profiles between “Ulti” (0.04mm) and “Fast” (0.15mm) modes. Their fast mode is a lower layer height than “normal” on Robox but if you don’t take anything else from my post the most important point is that print speeds on their 0.04mm is way faster than 0.3mm on current Robox profiles.
I have spent a lot of time troubleshooting two Robox machines, replacing parts and helping identify an underlying issue with the slicing on OS X. So I thought it would be good to try to help with print profiles but I get the feeling that I am wasting my time and CEL won’t want any help.
Could you let me know what the priority is at CEL? It is confusing that you are about to release the dual material head whilst the basic printing doesn’t seem to be as good as it can be yet? The lowering speed to help improve quality and removing the 0.8 nozzle infill from the profiles seem to be backward steps to me.
I suppose the ultimate question is that, whilst all the Robox features are nice, do you think the Robox will ever compare with the Ultimaker speed and quality?
STEELMANS 3D PRINT.
3D Hub: www.3dhubs.com/service/steelmans
Website: www.steelmans3d.comAs I said we have reduced our default speeds to concentrate on quality. All the settings are available to edit so feel free to increase them to find the best balance of speed and quality. Ultimaker users have been messing with profiles since 2012, I expect most of the data came from users rather than official releases.
We revise our default profiles with each software release. If you have a better profile we would love to see it. If it is compatible with a decent range of models we will even promote it for you.
Robox https://www.3dhubs.com/3d-printers/cel-robox is already as good as Ultimaker2 https://www.3dhubs.com/3d-printers/ultimaker-2 but is still being improved and has many more potential upgrades before its end of life.
The scores on 3D hubs for UM2 are from over 2 years, Robox was released 10 months ago so if we project another 12 months we should be on top.
Oh and how about Dual material? UM2 has cancelled their dual material add on head.
Our priority is reliable quality. Even a fast print is slow. If we want a fast print we go to the 400 micron/0.8mm nozzle which even while moving slowly can build a strong model in a short time (biggest nozzle+highest layer height). Want it faster? change the setting 🙂
For official support please visit www.cel-robox.com/support/ and create a ticket@pete Please don’t feel you need to defend the Robox. I own one and actually love all the features. I have had problems with it but in fact I like it that much I still have it!
Let be real though, the rating on 3DHubs is hardly an exact science whilst both machines score highly in different aspects. I think the test I did with my friends Ultimaker was more telling. Two printers in the same room, printing the same thing with 0.1mm layer height. When we had finished waiting for the Robox to finish the resulting prints were very different. I simply cannot be blinkered and say the Robox can compare.
Yes I heard that Ultimaker cancelled the dual material head. I suppose they weren’t happy with it and I imagine they didn’t want to release something that wasn’t finished or up to standard. Maybe this will return in UM3? (Just to clarify I don’t want an UM2, I want the Robox to be better).
Whilst it is interesting playing with the settings, at some point I would just like to select “ultra-fine” and print a really good print in a really fast time.
STEELMANS 3D PRINT.
3D Hub: www.3dhubs.com/service/steelmans
Website: www.steelmans3d.comAnyway attached are the 2 prints, both at 0.2mm layer height.
Ignoring the line around one of the heads (this is an intermittent fault) can you tell which one was the standard Robox profile and which was running much higher speeds?
Not too much difference considering one took 3 hours 7 mins and the other took 1 hour 53 mins.
STEELMANS 3D PRINT.
3D Hub: www.3dhubs.com/service/steelmans
Website: www.steelmans3d.comThe right hand owl looks much better to me. I would guess that was the one that was printed more slowly.
@simonhoult I didn’t want to jump the gun and immediately agree with your previous post (#21630) and now a few days later I have far cleared view on it. It think your (and as was mine) view what Robox is supposed to do does not align with CEL’s view. Mine was along the line of dynamic, young, modern startup that values community and gets big following base based on to listen and interact with us (the community). All of it enabling relatively fast initial development and with bit of skill and luck sustaining such pace. That would lead in people having idea of CEL’s vision, insight in internals of printers they bought and be able to contribute in many ways in improving the product.
What I found so far is that CEL’s opted for ‘traditional’ approach with closed development aimed to the easiest and least ‘interested’ part of the market (education and occasional novice home users rather than enthusiasts) with safety features, not top-of-the-range quality but steady and repeated results. Also, they have old mentality of ‘admit nothing deny everything’ making them completely disassociated from the ‘followers’.
My vision (hopes!) for the Robox (which I still think is beautifully crafted piece of consumer product) were that by now (a year after release) we would have at least a dozen home made projects to replace/improve head plus another dozen of other improvements done by community and similar think with wide range of improvements in software and tools for Robox. Not all ‘improvements’ really something everyone would like to do nor all with the same quality, but at least some of them released by CEL as paid for options.
In reality we have one or two simpler suggestions done by the ‘community’ and one from CEL in hardware improvements and Cura added to the AM. That’s pretty much all we’ve got within a year after printer was release. In the mean time I’ve noted some people here who came with ideas they can implement and some even started working on them just to slowly disappear (disappointed and disillusioned about Robox) and find other printers to put their efforts into.
Even the simplest thing like dimensions for GeckoTec surface which I thought would CEL really want to do won’t appear there unless someone of us make it happen. Simplify 3D is distant dream which most likely won’t happen or will when Robox finally becomes obsolete. Also all the enthusiastically put possible improvements on original Kickstarter page are very low priority at the moment and probably won’t happen in the lifetime of my ownership of Robox.
But all of it are not really failings of CEL nor are they ‘bad guys’. It is just, as I said mismatch between I hoped/wanted it to be and what in reality it is.
Oh, sorry for my long post… Even worse is that I failed to get to the point about quality: I am afraid that needle valve Robox has in head above nozzles are those that cause all the issues. Or, to put it differently - they appear to be proven much harder to control precisely as CEL wished and it affected quality. Imagine you make a replacement head for Robox with standard heater and through path and use Cura directly on it. I bet we would be able to see results so close to Ultimaker’s that one wouldn’t be able to tell difference. Maybe they would be even better for Robox. Having all this said - I am not advocating it - just observing what *might* be the cause of all the issues we have (blown head seals, blobs, missing lines, etc). There is only a handful of people who are working on rectifying those issues and to me it seems that they are struggling to pull what theoretically would be expected from current head Robox has. Or even worse we’re seeing what in developers’ world would be equivalent to ‘It works on MY machine’ scenario. But if I am right about what I posted above, then we can hope for some breakthrough done in CEL or just learn to live with what we have and not hope for much more. I am working on the second one 🙁
-
This reply was modified 2 years, 11 months ago by
You must be logged in to reply to this topic.
