calinb
-
Posts
305 -
Joined
-
Last visited
Content Type
Forums
Events
3D Prints
Posts posted by calinb
-
-
I've mostly played with nozzle size and width settings in Kisslicer and I forgot to mention that I also usually play with the flow to get the finer infill to fill out well and overlap too. Then I use the Inset Surface feature to get the walls back where they need to be.
The nozzle size also affects the positioning of the head. For example if you are printing a 10mm cube with a .4mm nozzle, in the gcode, the head will move 9.6mm because the idea is that the plastic will stick out an additional .2mm on each end.
At least this was true with the older slicer. Haven't tested steamEngine.
- George
-
Wow--you got a bargain, Daid! I think they quoted US$7k per year to me so I didn't even bother with their free trial, but maybe the price was high because I have a small startup business. I don't know what their business was like during its first few years but startups aren't wealthy. In fact, they typically lose money far longer than the investors ever expect! There's no way I'm spending $7k every year on Spaceclaim anytime soon!
I've recently been using Spaceclaim for some modeling work:
It's not cheap (I think a single floating license is something like 2k), but holy shit. It blows everything away.
-
My UM has always stuttered in Z fairly often. I don't have to bottom the stage like you to reproduce the problem either. Moving Z more than 20 mm using my Ulticontroller is always asking for trouble when it homes. Sometimes it stutters after no such provocation too--usually when I start a 2nd print right on the heels of the first print. Like you, I've seen the problem occur very rarely in X/Y. The period of my stutter is longer than your stutter, but my Z speed might be slower than your UM too. The pause in Z-motion is quite long--maybe 1/4 to 1/2 second. It's VERY irritating and forces me to make countless power down resets (the only way to clear the problem), requiring me to reset preheat temps and / or reload the .gcode file from the SD card with the UC.
But my UM and UC electricals and electronics have been VERY irritating, in general and multiple electronics assembly defects have been my major gripe with the machine. I wish none of the electronics had been assembled by UM Ltd., because it would have taken me far less time to assemble the boards and cabling myself than to debug and correct the problems. (I'm speaking as an electrical engineer who transferred two major Intel board products to off-shore manufacturing facilities in Asia). I've spent far too many hours reworking my UC PBA and replacing solder-less crimp connectors that had been incorrectly assembled with solder than I spent assembling the machine in the first place.
I had this stuttering of the z-axis sometimes but I could never reliably reproduce it. Until now.
-
I've seen sandblasted glass plates, but have not tried it. The potential to ruin a nozzle is a noteworthy downside, Daid!
I'm using Uhu gluestick for nylon, but the best ABS adhesion I've ever achieved is with "ABS-juiced" Kapton or PET tape (and "aged" juiced tape is even better). I've broken a grand total of three of my hot-swapable plates trying to remove large parts from them. Fortunately, I pay only about $3 for a plate at the local glass shop. It's not worth my time to cut them, given my relatively meager skills and poor yield. Borosilicate glass would be an expensive proposition for me, with little or no benefit.
We have a desk full of ABS prints :wink: also to test best platform adhesion (sanded bed glass with glue stick works best. However a sanded bed can grind your nozzle ruining it. Which is why we did not go for a sanded glass plate)
We did not test nylon, PET or other materials. But I do think the UM-Original is a better printer if you want to tinker with stuff. (Buy both! one for printing one for tinkering :wink: )
-
I use a trick with Kisslicer that have not tried with Cura yet. I sometimes get much better results by setting my nozzle size and line widths much narrower than my my nozzle orifice size. I typically do it in my quest for perfect solid infill (at the cost of my patience, of course). The trick can achieve better solid infill where even slowing way down with the thicker settings results in little to no benefit.
What the nozzle size effects is the extrusion width, so the with of the lines it will try to put down. As you can easy put down 0.8mm lines with a 0.4mm nozzle entering a slightly larger nozzle is no issue.
-
Your concerns that the fan cools the heated bed are well-founded. The trick is to set the bed temp extra high for the first few layers (at least two layers). I often use 130 C. Do not run the fan for these first layers. Reduce the bed temp and run a low fan only through the problematic region. Only run the fan high enough to preclude the distortion. Again, if you err on the side of too much fan, the layer bond strength will suffer. My fan is stronger than the OEM fan (160 mA vs. 100 mA) and I never need anything near full speed to solve this problem. The correct fan setting is touchy and varies with the part. My fan speed range can usually produce a part with the distortion you have, a part without distortion, or a part with small layer delaminations--all just by changing the fan speed at various heights but not changing anything else in the process!
I occasionally use a skirt (a thin wall) around the part to protect the lowest layers from the fan's breeze. I almost always use a brim, which I typically add to the CAD model myself, because most slicers cannot create brims.
I've experienced the problems you're encountering but I have successfully solved them and printed a wide variety of large, thick, and blocky parts in ABS using these techniques. I also use the bubble wrap tent and achieve chamber temps as high as 65 C to 70 C at nozzle height within the tent. I do'h need no stinkin', expensive, and patented Stratasys oven!
For the best possible adhesion to the bed, I recommend wiping Kapton tape with a paper towel moistened with acetone. Rub it on a piece of old ABS and smear the bed. If you "age" the "juiced" Kapton with a pre-print, and can remove the part without damaging the tape job, the adhesion will be even better on the 2nd and subsequent prints. The adhesion is so good, I can actually turn off my bed heat after the first layers (if I don't need it to heat the build chamber) and the part will stick well all the way down to room temperature. Actually, I've found that 130 is a bit too warm for maximum Kapton to aluminum or glass bed adhesion. PET tape is the same, but it needs a bit of heat to stick well to my bed. The PET is thicker and tougher than Kapton, however.
I really want to know how the new UM2 automatically solves all these problems, becoming a turn-key FFF solution for newbie makers!
Okay--the heated bed will help a little, but plenty of 3D printers already have heated beds.Thank you for the tips.
Yes the photos are oriented in the build direction. I haven't used any fan at all, because it seems to cool the heated bed and affect adhesion. I may have to try with a lower fan setting.
BTW, I am using the new heated bed: http://reprap.me/Alu-Heatbed-MK3
which is working very well. But my thermistor must not be the right one because it doesn't show more than 90º C and I am getting good adhesion with large flat and fat ABS parts.
-
Any chance you could release solid model files for your design, Musti?
Thanks!
I have taken Nick's design for the E3D mount and made it more easy to print with nice large surfaces for the part to sit on.
<snip>
http://www.thingiverse.com/thing:145988
-
I think Owen nailed the cause. As he said, temps are critical with ABS--both ambient air temp and bed temp. Are the parts in the photo oriented in the build direction? I'd try to print the part in a heated build chamber (I use a bubble wrap tent.). I'd use a high bed temp (about 125) for the first two to four layers to bring the tent up to temperature and then reduce the bed temp to 110. I'd try a light fan for the first 10mm (or as high as the plastic is shrinking inward) and then turn it off until the the layer times become very short at the top protrusion.
A narrower waist from shrinkage, as illustrated in you photo, early in the print is common. Some fan usually helps. If you over-cool with a fan, darker ABS colors will often exhibit thin lighter color lines or striations, due to poor layer bonds and partial delamination.
You could also try thinner layers (longer print time). In general, thinner layers seem to reduce this effect but there are so many process combinations, I don't think you can count on any rules.
I've not found anything better for ABS adhesion than either juiced (ABS/acetone wipe) Kapton or PET tape. Hairspray on glass can work and be very convenient, but juiced tape is the best. It's so good I've broken a grand total of three glass build plates trying to remove large and blocky ABS parts (and I'm careful and know all the tricks for removal). You could also add a brim to the part, if adhesion to the bed becomes a problem while tweaking your temps.
-
I have the full set of MakerGear nozzles. They work well. The MakerGear extrusion diameter values (extrusion into free air) are quite accurate. Obviously, extrusion onto a build surface and resulting line widths are a different matter.
Mostly I use the .50 mm nozzle, because I print a lot of ABS. The 0.25 is only useful for PLA and I find the 0.35 to be extremely marginal for ABS with the stock UM filament feeder.
Using a simple test case, I swept Cura across my nozzle sizes and a little more. Cura gcode output was invariant within each of these regions:
???-.20
.21-.24
.25-.30
.31-.40
.41-.79
.79-???
So obviously Cura is just using the nozzle size setting to determine a reasonable range of extrusion rates. If your setting is in the right ball park, the results should be valid.
After a solid year of printing with the V1 hotend and nozzle, I finally destroyed the nozzle during disassembly. Because I'm in the USA and a cheapskate, I ordered nozzles from MakerGear. I purchased a .35mm and a .75mm nozzle. On their description page, it says the 0.35mm nozzle actually extrudes at 0.43mm. It says nothing about the 0.75mm nozzle. I know that this is influenced by die swell but what parameter is the "nozzle width" setting is Cura actually expecting?
-
It's good to hear that it's possible to add a knob. I'm not worried about the motor as much as the mechanism that applies the motive force directly to the filament. It's always the weak point--not the motor. I predict there will be times, even given improvements in the UM2, that it will be desirable to turn the knob by hand while simultaneously forcing the filament with one's other hand. Manually turning the mechanism will surely continue to be far easier than accessing motor drive control from the UM2 controller menu.
Also, I suspect most pre-release testing of the UM2 was performed with PLA. As with the original Ultimaker, other plastics will probably prove to be more challenging to feed reliably through the hot end.
-
Also netfabb does not do boolean in two overlapping meshes. It only joins it on the top layer. Have had plenty of failed prints to realise that happening.
If the meshes overlap, use Blender to combine them with a boolean union. Kisslicer will probably have trouble with such files too (and usually with a lot less ugliness than that :()! Kisslicer is the pickiest of all the slicers, in my experience. It doesn't usually result in a slicer failure, it just produces toolpath artifacts and ugliness.
I'm a little late to this thread but I've got netfabb figured totally dial-in and I use it frequently. I invested many hours of trial and error to get to his point. Daid is right--there's no need to buy it until you find yourself unable to control the other slicers at the level you desire or you fail to get acceptable results, due to a slicer limitation. netfabb isn't even the fastest slicer anymore. (With Daid's new "SteamEngine," I think Cura is probably the fastest!)
However, when it comes to layer by layer control of the printing process, there's nothing even close to netfabb. It can do so many things no other slicer can do, though it might not be obvious how to realize its full potentia, because, as Daid warns, it's a PITA to learn (and the documentation is very outdated and incomplete).
Finally, netfabb is relatively buggy. nf fixed its completely unusable retraction nearly a year ago but many bugs remain that are either irritating or require a work-around. I would not do without the program in my arsenal and I use all the slicers (though I use slic3r only infrequently) but nf is certainly not a program I'd recommend to every user. If nf ever fixes most of the outstanding bugs I've filed in tickets to their support system and updates their documentation, I'd recommend it whole-heartedly to anyone who has the budget for it. For now, I've completely stopped reporting nf bugs. The tickets I file to their support system just fall on deaf ears! nf really lost this community with nf's lack of responsiveness to the retraction bug one and two years ago. It would take some pretty big changes at nf to earn back interest in their products here. That's my opinion anyway.
-
I've printed many rolls of ABS parts for my company's products (both prototypes and beta customer deliveries) and I've used a torque wrench to perform empirical destructive testing on many parts. In my testing, the "grain" strength (layer bonding) always improved with higher nozzle temperatures in all the brands and colors of ABS I've tested. I'm comfortable in saying that no ABS will exhibit a reduction in strength up to the 260 C limit of a stock UM, in combination with any reasonable and practical print settings. With ABS in particular, the Achilles heal of FFF/FDM is the weaker layer bond strength.
On the other hand, the problem with 260 C is it's hot enough to increase the probability of hot end jams, due to "heat creep" to risky levels. (Perhaps not Ultimachine black so much, which requires the most heat of any ABS I've printed, but many others can be prone to jams at 260 C.) Given that most new users aren't even aware of this failure mode, I think I'll take a swag at a recommendation for 245 C--or maybe 250 C at the very most. I think 245 C is pretty well aligned with your choice of 220 C for PLA, in terms of maximizing the chances of a hassle-free and successful 3-D printing experience for a new or inexperienced user. Ultimachine black prints okay at 245 but a bit of layer strength will typically be sacrificed at 245 C vs. 260 C. I think it's a good trade-off for quickprint mode users. I often print many other brands and colors of ABS at 245--particularly the lighter colors. Of course print speed affects optimum nozzle temperature too but, like the Cura PLA settings, I'm trying to suggest one temperature for all three quality modes.
If Joergen chimes-in with an opinion, I recommend weighing it into your decision to change the default Cura quickprint value too. He probably has as much experience printing ABS as anyone!
(Back after 2 weeks vacation)
I have little experience with ABS, what would be a better ABS temperature?
-
I have always "primed" the mechanism by hand and never had a jam since I started to work with the Ultimaker in March.
I completely agree. Single-handed manual drive capability is essential. (The other hand can simultaneously force the filament too, which is often sufficient to clear jams or partial jams and preclude a hot end teardown).
-
Thanks, Daid.
I like what you are reporting here! I can mill slider blocks. Too bad about the hot end parts and (surprisingly) the Z-motor. (Why isn't an off-the-self motor used for Z, given that there's a plethora of motor designs readily available?)
I must comment again that, even before watching Erik's announcement of the UM2 yesterday, I've noticed a bit of a shift toward improvements in open-source projects support from UM Ltd.. It's proof that UM Ltd. may be "putting more of its money where its mouth is (or Erik's mouth yesterday)!" I've noted the additional parts available from the UM web store and the release of the .stl file for the injection molded parts on Youmagine. These steps are much appreciated! I guess a factor might also be that UM Ltd. is more willing to reveal its former crown jewels and increase support in the spirit of open source, now that the UM2 has been released.
Oh--I keep forgetting to ask you something. Is your Steamengine source code from Cura open? That's the only part of Cura that really interests me. You appear to be notably advancing the art and science of slicer engines and, in the spirit of the open source initiative, I think the fruit of your labor should be shared.
The UM2 parts are difficult to source. But some parts are easier then from the UM-Original. Most of the injection molded parts are printable, except for the slider-blocks (which are not strong enough when printed) The dibond panels are difficult to source in 6mm, but can be replaced with other materials. The hotend bits are custom, will be expensive in low amounts. Rods, belts, motors, doable. Z motor is a difficult one. But pretty much, you are better off self-sourcing an UM-Original, it's cheaper.
But the thermocouple has been replaced with a PT100 for example, which is easy to source.
Anyhow, I think you all know I do not like marketing too much. And yes, we are planning to OpenSource all the files of the UM2, but not from the start. But we will put up the UM-Original files (All of it, 3D files, documentation, BOM) within the next few weeks.
Source-code of firmware will be released when we start shipping. And Cura is already OpenSource (hell, those who has been paying attention will notice that I already pushed UM2 info into the github repository)
-
Thanks for the links to the great comparison photos, Nicolinux! My initial thoughts follow your bolded and italicized comments below.
1. Heated bed
This is the best improvement so far. I know that some people don't value a heated bed, but from experience with another printer and now with the Ultimaker 1 - it makes a huge difference.
The heated bed is a huge improvement--even just for PLA. It's like Erik said in the product announcement in NY. Once the bed cools, you can nearly "blow the parts off the bed!"
Even though I designed and added a heated bed to my UM for ABS use, I find it to be so much more convenient to use than blue tape for PLA.2. New hotend
I don't have enough exprience with the actual hotend to see its drawbacks (and I didn't experience clogging yet), but the new hotend looks good and if it is lighter, that'a always a plus. Also the dual fans should help with small and delicate prints.
Occasionally I wish for more even (multi-sided) cooling with my UM, but it's not a biggie for me and I've never been motivated to add another fan or ducts. The angled UM2 fan helps to balance the reduction in access possibly imposed by two fans, however. I'd like to know more about the UM2 hotend. While I'm completely happy with the few simple mods I've made to my UM hotend, a shorter hot section and higher temperature capable materials (all metal), might be nice to have for the nylon I'm now printing and other materials. I would not want to give up the versatility needed to print PLA and ABS, however.
3. New Extruder
This thing looks awesome. Switching to "direct drive" was a good idea. And overall less pieces means less opportunities for something to break.
I hope the new UM2 extruder filament drive is stronger than the original UM drive. The drive strength generated by the original UM drive limits the speed at which ABS can be printed using a 0.4 mm (or even 0.5 mm) nozzle. A 0.25 mm nozzle is impossible with ABS. Of greater concern is the lack of a manual assist mechanism on the UM2 filament feeder (as far as I can see). Even if the not end is less prone to jams than the original UM hotend, I'm certain there will be times when I will want to force feed filament into the feeder with one hand while turning the drive mechanism (large gear in the UM) with the other hand to clear jams, partial jams, or prime (preload with pressure) the hot end with filament.
4. New case
I think the new case looks slick while it retains the typical Ultimaker design. But I don't understand why it is not completely enclosed. I don't buy that Ultimaker went to these lenghts to add enclosed walls on the sides just to avoid air blowing through the printer and that could disrupt the heat flow. I guess that's important for ABS but then why not build a completely enclosed heat chamber? And by the way, I think the biggest benefit for the heated bed is layer adhesion and not fighting warping.
For large, blocky, and thick ABS parts, I often erect a bubble wrap tent around my UM and achieve build chamber temperatures as high as 65 to 70 C. I'd like to see removable side AND front panels. I'm not fond of the permanent side walls. They are not always needed and they impede access to the inside of the printer.
5. New mechanics
Thats related to the new case too and something I don't like. They could have moved the motors outside and ditch the short belts. I don't understand why that has not been considered. Maybe for aesthetics reasons. But then they could have moved the x/y and extruder motors to the back side and add a nice cover. And since the back is already "wasted" by the big and bulky filament spool and the extruder - why increase the overall depth and enclose everything? If they truly wanted to make the printer more compact they could have increased the height and moved the filament spool underneath the printer.
Moving the motors somewhere else is required if they ever decide to implement a heat chamber. So that could have been prepared aswell.
Agreed on all points! Because I sometimes run a heated build chamber, my motors are outside the frame. I have all the parts required to ditch the short belts too, but haven't found the time to make the mods yet.
6. Z-Stage
I don't know what changed exactly but the z-stage changed and that's definitely good. They seem to have switched to three-point bed leveling which is nice too. And finally some thumbscrews!!
If the z-stage is more dimensionally stable with temperature and humidity changes, I'll be pleased. However, I often run hot-swappable glass build plates and I'm not sure how well they will work with the UM2 design--especially given that there's no side access! The glass plates will need to be swapped out through the front "window." Might be okay, but I still want side access.
7. New controller
This one looks very slick. Reminds me a bit of Eva from the movie Wall-E - link
I hope they changed the finky wheel that tends to skip steps when it is turned halfways. One odd thing is the angle of the display. I guess Ultimaker expects its users to have the printer on their desk at "eye level". The old UltiPanel was tilted upwards a bit which is better readable.
An angled display (especially an LCD display) is better. Also, push buttons are more easily pushed when the push is largely downward. (ATM keyboards are not vertical!) This looks like an ergo demerit for the UM2 to me! If the new controller and firmware doesn't contain the bug that causes my Z-stage to sometimes glitch (periodically pause) after a previous print or Ulticontroller Z-stage manipulation, I'll be pleased however!
8. Printer for everyone
Sorry but I don't buy that yet. I don't see Joe average downloading model x from YouMagine or Thingiverse and hit print. You need pliers, a digital calipper a screw driver and so on. Then you need to think about speed vs. quality, deal with blobs and under/over extrusion. There is no printer in this price range that can counter all that. Besides, I deffinitely don't want to sit next to the printer (even if it is quieter now) when printing with ABS...
I think the "solution" Ultimaker tries to sell doesn't match their honest and charismatic image. Some customers will be disappointed when they find out that 3D printing isn't that easy. Better to keep it modest and true. Don't blur reality with Apple-like over polished presentation.
Yeah--a low cost FFF/FDM printer "for everyone" ain't gonna happen! The very nature of the materials processes preclude it. I don't make this prediction to suggest that there's no room for improvements in usability, however. Though I only have an Ultimaker (fortunately) and I've learned to use it very effectively, I understand all of the experiences related here and can sympathize with the author:
Finally, Erik was promising a new level of dedication to open source practices. Time will tell whether or not UM Ltd. truly embraces open source or just uses it as a marketing tool. The original UM has not been the best platform for open source activities, because several components (heater, TC, injection molded parts, etc.) have been impossible to procure or fabricate until the last month or so, They had to be "re-designed out" of a self-source build. It was the same for pulleys, in all practicality. An .stl for the injection molded UM parts recently appeared on Youmagine and it's much appreciated. A couple of months ago, I went to the Lulz site and open source archives to see what it would take to self-source their latest printer design. At the time, I thought it would be far easier to do than with the original UM. Now, because of the recent additions and support I cite above, it's probably closer to tie. In the NY UM2 announcement, Erik said an advanced user can rewrite Cura, but I don't care much about Cura's user interface. From a functional standpoint, I want the source code for "Steamengine," Daid's slicer engine. If Cura is claimed to be part of the Ultimaker project (and I think it clearly is such), ALL the source code should be released to justifiably claim the UM is "open source."
-
Hopefully, UM Ltd. will continue its recent trend to offer the few odd and single source parts with the release of the UM2. Only recently have parts like the cartridge heater and TC become available to builders and an .stl file for the injection molded parts showed up on Youmagine. It's true that one can redesign and improvise with the original UM and build a functionally identical printer but, looking at the photos of the UM2, it appears to contain many more specialized and difficult to procure or fabricate parts than the original UM. Redesign and improvisation on a self-sourced UM2 project will likely prove to be even more difficult than with the original UM.
Historically, all the parts contained in the original UM have proven to be more difficult to obtain than a typical open source 3D printer and I suspect that, in practice, the UM2 will be even worse. There's a difference between good open source design and an open source marketing tool!
-
I think I'm with George... is there really a valid use case for Raft (on an Ultimaker)? I think I've only ever used it once. Am I missing something?
Same here. I've never found a good reason to use a raft, though I've tried them many times. If a raft works at all, a brim works better!
-
When I have a difficult support challenge, or a part where surface finish is very important, I usually reach for the netfabb Ultimaker Engine and its "fluff" support. netfabb uses a different approach than all the other slicers called "fluff" support. Rather than trying to make support structures that break away in large pieces, fluff is what it sounds like. It crumbles away in many, easy to remove pieces, but it is also highly configurable so the bases can be made quite solid too. If some of the fluff breaks away (blows away
) before the support is finished, it usually doesn't matter. Very often, a few wads of fluff can fall apart but the support structure still works well. netfabb fluff is also the easiest support to remove from nearly inaccessible cavities in models, using a poker and long scraper of one's choice, like a screwdriver.But like everything netfabb, it took me countless hours of trial and error to achieve excellent results with fluff support. Though, the default fluff settings (weak fluff, strong fluff, etc.) worked quite well for PLA, it was worthless for ABS. It took much tweaking, but I like the concept and the results I get now.
I really wish netfabb would fix the remaining bugs in their UM engine and update their documentation, but I guess they are too busy developing and releasing their new software products, instead of fixing their old software. Given netfabb's apparent lack of interest in further development of the UM Engine, I really hope the other slicers catch up to netfabb's extensive layer-by-layer list of slicer controls quickly too.
For an example of nf's fine control granularity in the context of support, what other slicer gives a user the capability to turn support on and off on a layer by layer basis? Yes--I use this feature. I have models where the slicer creates support for the full range of layers, but some of the support in the mid-level layers is not necessary. Yet the models require support higher on the print. If the base of the higher support starts higher too, I can disable the support where it's not needed and turn it on back on again for the upper layers.
-
Have you ever had any situations where you would find it advantageous to print multiple prints at the same time layer by layer?
With multiple very small parts (or parts with large numbers of small layers), it's sometimes faster to print them at the same time, layer by layer, because the print head doesn't need to slow down to provide time for the layer to cool.
I often print 35 to 40 small parts on a "sheet" for our commercial products. Some simple parts print reasonably well as fast as 300 mm/sec travel and 250 mm / sec solid infill (solid only in a few places) and loops (mostly loops / hollow parts) using 0.12 mm layers / 0.5 mm nozzle, but they don't cool enough--even with my more powerful replacement fan blasting away at full speed (and this is ABS)! Then the tradeoff becomes stringing (print all at once) or slowing down to print them one at a time with enough time for the layers to cool.
-
Cura 13.06.4
Reset Profile to default (probably moot)
Switch to Quickprint
ABS
High, Normal, and Fast low quality print settings all produce 260C nozzle temp (a bit too hot for nearly all ABS filament).
;Generated with Cura_SteamEngine 13.06.4
M92 E839.100000
M109 T0 S260.000000
T0
;Sliced at: Sat 07 Sep 2013 13:01:51
;Basic settings: Layer height: 0.2 Walls: 1 Fill: 1-
1
-
-
My most cherished Kisslicer features that are not present in Cura (AFAIK) are inset surface, loops fan only, stacking, and support control. I'm unconvinced of the benefits of wipe with my typical media (ABS) and process settings.
I'd like to see more support control in all slicers (netfabb has the most control and produces the easiest to remove support, "fluff," once the settings are optimized, though the settings are pretty good right out of the box for PLA).
Obviously, all slicers need much more extensive dual extrusion control. I'd like to see the ability to assign any extrusion type to any extruder. I plan to use a second 0.70mm nozzle to print solid infil very quickly. My next interest is PVA support. I'm not terribly interested in dual-color prints, because my 3-D printing interest is the production of industrial parts and prototypes, rather than art work, toys, or trinkets.
Is it because of wipe, or a different feature? (or multiple).
It could be worth checking the latest version of Cura, i heard some good stories about it handling big (huge) files.
However, dual extrusion for support is not yet supported.
-
Well--every filament is different (even from color to color) so that's good to hear. I've been printing Ultimachine blue translucent PLA lately at low speed. I was printing my first layer so slowly that I had to reduce the nozzle temp from 220C to 210C for just the first layer.. The layer took too long long to print and I was getting a jam from heat migration up the filament.
But you are right that 230C and slow speed is not the best idea and might cause the heat to go up and mess up. Now, for this print I needed low speed, and under 230 it wasn't fluid enough I felt, especially at the beginning of the print. I will reconsider this thanks!
-
I have a 160mA fan and no problems. I think the original was 100 mA. The new fan is significantly stronger than the original.
-
I agree with Joergen. Plus 3% is the most you should need for a flow tweak. If you are running high flow tweaks to fix under-extrusion problems), the problem is probably due to filament feeder slippage (not necessarily resulting in visible grinding, as illuminarti has also observed and reported). The slippage occurs at filament feed volumes well under 10 mm ^3 /sec with the standard 0.40 mm nozzle. To some extent, you can crank the flow setting or flow tweak way up and compensate for the problem, but it's better to just slow down. I've done quite a bit of testing with 0.25, 0.35, 0.40, and 0.50 nozzles. The 0.35 is barely usable for ABS (obviously at slow extrusion rates) and I find the 0.40 mm nozzle to be marginal. The 0.25 nozzle is only useful wih PLA, if one accepts the additional challenges.
Joergen is using a 0.50 mm (or a bit larger) nozzle and my nozzle also measures 0.50 mm. Even with a 0.50 mm nozzle, sustained 10 mm ^ 3 / sec ABS extrusion can be quite iffy (10 mm ^3, /sec is often considered the upper limit for PLA with a 0.40 mm nozzle).
Mostly I recommend that you try a larger nozzle and higher nozzle temperatures. The feed pressure required goes down with increasing nozzle temperature. Also, I've done extensive destructive testing of my ABS parts using a torque wrench and my parts are always stronger with increasing nozzle temperature. The problem becomes the service limits of the hot end materials (PEEK and Teflon are both good for only 260C or so in long term service) and the increased likelihood of hot end jams at higher temperatures. Joergen and I have made simple mods to our hot ends that provide a bit more temperature headroom. (I'm very comfortable printing at 275 C, which is my standard nozzle temperature for Ultimachine black ABS). I never print any ABS below 240 C.
The next item on my wish list is a more powerful filament feeder. Either the one that mooncactus has developed or perhaps I might work on refining this dual counter-roatating drive feeder. I'm certain that a stronger feeder will enable higher print speeds or smaller nozzle orifices
http://www.tridimake.com/2013/04/rollerstruder-filament-feeder-driver.html
http://www.thingiverse.com/thing:62145
I have the Ultiamaker V3 drive "hobbed bolt" (it's really a knurled stud) so I made a conventional hobbed bolt to try. It did offer more power than the V3, but it was very intolerant of any grinding. Once it started to grind, it was game over, and it had to be removed for cleaning with a dental pick. I re-installed the more fault tolerant and easy-to-clean V3.
I print ABS on heated bare glass, Kapton, and PET. Bare glass is very convenient but far more challenging than printing PLA on heated bare glass. Kapton and PET are equivalent. I recommend them to start, because the first layer settings become far less critical when using tape. You can experiment with "ABS juice" too, if you need better adhesion between the ABS and the tape. In fact, "aged" Kapton and ABS juice can be too strong! I'm up to a total of three broken hot-swappable glass plates!
It's best to leave the fan off for at least the first couple of layers and most ABS prints require little fan (but there are times when its essential). My process parameters are quite complicated at times and highly dependent on what I'm printing, but I've written enough for now, I think. Good luck and let us know of your progress.
Okay--the heated bed will help a little, but plenty of 3D printers already have heated beds.
Spaceclaim
in Design for Additive Manufacturing
Posted
It didn't take long to discover the "catch" in the free deal. It doesn't seem to permit the importation of any standard solid model formats. (It imports only .step but read-only, which is nearly worthless.) It also does not permit the export of standard solid models required for, say, injection mold manufacture. Geesh--even the free version of AutoCAD Inventor Fusion supports import and export of standard solid model formats, though the performance of the program leaves much to be desired.
DesignSpark seems fine for designing stuff for a personal 3D printer but I guess ya' get what ya' pay-for!