mastory

The amount of extruder drive problems have seemed to diminish with the later UM upgrades. I would tend to take Illuminarti's advice about focusing on what the specific problem is. That said, I had some trouble with Ver 1 or maybe it was Ver 2 so I tried out the one designed by Geo Hagan http://www.thingiverse.com/thing:15897 and my troubles went away. If I add a secondary, or have troubles with the one I have, I like the designs based on the MK7 drive pulley. One possible disadvantage with the G.Hagen design is the larger dia drive results in less effective force to extrude with. Again, no problem for me. There is also a lower resolution when extruding. I am running an E value of 315 whereas the stock is something like 862. No detectable problem there either... Matt

I was trying to minimize the heated mass so I tried 3/16 (~5mm) sheared aluminum (I think 5052 alloy) for the substrate on my HBP. I didn't even try it as it wasn't as flat as the stock acrylic bed. The supplier I had chosen did not publish a flatness tolerance, and the piece was warped in the process of shearing so I went to the MIC6 which is very flat. For me, the best advantage in my opinion to upgrading to the HBP is the added flatness. The inherent flatness of the MIC6 and/or 3/32 window glass greatly reduces the need to re-level. I can now confidently start a plate of parts and don't need to be present for the start of every part. The heater I chose is an adhesive silicone type like sold by Omega in the US. Mine is 7x7" 120vac, and approx 250watts. To heat the whole thing up including an 1/4"x9x9.5" MIC6 plate, and the glass takes 4-5 minutes. MIC6 is also known as cast aluminum tool(ing) plate. Don't forget to leave enough area around the print area to fasten the glass to the aluminum with printing clearance. If I do it again, I would make my aluminum plate a bit larger. I would also strongly recommend changing to a 3 point leveling system If interested, I am controlling mine with a SSR wire to the UM HBP output and powered by a separately fused 120vac main. I used a thermistor in the usual way. There is more specifics on the google group good discussion here, more specific to my chosen solution nearer to the end https://groups.google.com/forum/#!searchin/ultimaker/hbp/ultimaker/8xjR_npZACw/QhmUO3jaWNcJ and here https://groups.google.com/forum/#!searchin/ultimaker/hbp/ultimaker/lqqVzodLcwU/Wzgop__xu44J Matt

Due to the physics of extruding plastic in a circular path and the properties of the melted plastic, the extruding material will "pull" toward the center of the holes while printing. This is unfortunately normal and expected. You will find the effect is greatest on small holes and diminishes as holes get larger. To print a 2mm hole, it might be necessary to size the modeled hole at 2.75mm or even 3mm before slicing. On 30mm hole, your model may be 30.25 or less to get the desired result. Trial and error is the key to success. When I have a hole that is critical, I make test prints that duplicate the intended print settings and even the surrounding geometry of the part. I can generally make one test print, measure the print, make an offset in the model and print precisely (repeatable and within a reasonable tolerance) the size hole I want on the finished part. Lately, the guesstimate used to make a test print tends to be right on and needs no further adjustment. If in doubt, err on the small side and open undersized holes by drilling if necessary. Matt

Why is this topic still pinned? Started over a year ago and its been 5 months since the last post. The fact that its at the top could only hurt the reputation of our favorite printer - although the hot end weakness may be largely overblown.

There should be a few links from this site to YouImagine. Maybe under "community" Here's one... https://www.youmagine.com/

One of the problems I have with Thingiverse is that I can't find designs that I have seen before and I know are there. More powerful searching would help. Maybe make the designs/models/'things' tagable by the original poster or by other registered users to aid later searching. It would seem that a moderate number and finite list of tags would be most effective. If something is obscure enough to not specifically fit in a tag category, the name/description should suffice. Matt

In spirit, thats is somewhat like what I have here: http://www.thingiverse.com/thing:20572 Single internal transition from the PTFE to the nozzle tip. The second picture shows it best. Still never had a jam in over a year of service. Everyone likes to hate on the PEEK, but no troubles there either. None of any kind actually. Honestly though, I have yet to try printing ABS. Maybe I'll do that in the next few weeks. I am looking to add a second nozzle, but waiting to see what boils out of the current evolutions. I have been watching the E3D threads, and am interested in what transpires before I build a second home brew nozzle. Matt

Yes, I did have a learning curve to clmb to get my UM to work to my expectations, and I did early on upgrade my hot end and drive to better than stock equipment. For me this was empowering work to do. In hindsight, a heated bed was less of an advantage than I thought going in. It's installed and works well when I choose to use it. When I eventually get to printing ABS, it should be a huge advantage. Without the advice and collective knowledge snd advice of other users, I could have done none of these things. What I'm trying to say is, the UM open design is a huge advantage to the forward development of the machines - both hardware and software. There are currently several choices for upgrading hot ends, more choices for filament drives, and other new mechanical upgrades almost daily on Thingiverse. The fact that the choices exist is proof that we are witnessing the evolution (And better, taking active roles when we choose.) The strong solutions will survive and the collective user group can ride on each others successes. The flexibility of the firmware faciltates this evolution when it is a factor in hardware changes. I can now capably alter my own version of the firmware. Ask 3DSystems to alter their firmware to enable a new function and see what happens Specifically, I am personally very excited about the E3D all stainless hot end that is being applied to the UM (by Lars and others) and furthermore a dual or triple hot end built on that platform. If this is found not to be the ideal solution then we will learn from it and the ideal is just around the corner. Whats more, the E3D solution is very affordable even in a multiple of 3. When it comes to software, I have never seen a process application develop as quickly and responsively as Cura has. Not much more needs to be said. A strictly proprietary machine cannot adapt at nearly the pace. There will be one source for support and you better call between 9 and 5. The open architecture and its distributive forces are fostering the improvement of our machines at a greater pace. Ultimaking probably doesn't have the resources that 3DSytems has for R&D, but the user community can make up for it at least in part. The enthusiasm of the group will in part also compensate for the desparity. I am sorry Bill, that you have had so much difficulty in getting your UM working to your satisfaction. I realize that alot of people aren't going to have the patience or accumen to suffer though it when things don't go to plan. I get the fustration with the staleness of the content in the build wiki. I think local users that could interact in person and first hand with the machines could help alot in this regard. The user base here in the states seems to be pretty sparse. What could be done to help get new users in contact with experienced ones - especially in the earliest stages? The users need a way to locate one another. At one point there was Google map that showed the UM locations. Unfortunately, any malcontent could edit it and render it useless. Could this be replicated in a more secure format? Best Regards, Matt

I am curious and I think alot of others probably are too. How many Ultimakers have been sold overall? The pace of sales and is it still accelerating? Distribution by geography? What I would like to hear is some high numbers. I want to take comfort that there will be an active and vibrant UM community for a long time to come and that company has a foundation for long term health. This may be guarded information and I would sort of understand. Matt

I think he is taking a well deserved vacation.

I've been giving some thought to some Slicer and/or firmware features that might improve printing. Mechanics are my forte, so bear with me if something I suggest is out there programming or electronics wise. I like to consider alot of ideas, good or bad. The more the better. A small percentage will develop into something real. A few will trigger other ideas. A. One thing is improving the function while pausing a print. The goal would be to make the pause function predictable, more useful and non-harming to the print. This could be the result of a premeditated stop (maybe to change filament, insert an object into a printing part, add support tooling, etc...) or an emergency on the fly thing. I started a post on the Gogle group about this. Premeditated stop points could be placed in the Gcode by the operator/slicer at slciing. A subroutine would be triggered when the stop code is encountered. Maybe for inserting something into a print it might go like: Stop, Retract, Rapid Short Z down, Jog X & Y to known postion (maybe over infill, or off the print to the side) -Pause- Resume print from break point (jump to position, Adjust E, Resume) on operator action. Unplanned stops could be intiated spontaneously by the operator for any reason. Once called, the machine would proceed to the next convenient point and then stop like above. The potential stopping points would be pre-defined by flags that were placed at strategic points in the slice (maybe one per layer, the next opportunity over infill', Z change...) If no unplannned pauses are requested, all the flags are ignored. I doubt it's feasible, but a third 'immediate' stop mode would seem intuitive. By this, I mean an imediate stop mode that could be resumed from, maybe with the sequence from above. B. Potentially borrowing from the Flat Bottom function; Lets say you have a model that you would like to split into 2 (or more?) pieces before slicing. The model could be positioned so the XY plane of the printer bed cuts the model in half. The slicer is intiiated on the model as positioned and slices from the bed up and saves the file. The slicer then inverts the model, and uses the same slicing parameters to slice the other half of the model. To kick it up a notch, maybe registration holes could be placed in both halves?? .. or snap features?? C. Its an old topic, but it would be a great benefit to make cylindrical hole printing more dimiensionally repeatable. Estimating offsets and trial and error testing becomes tedious. If this could be handled at the slicer, the benefit of shared models would be greatly improved. I don't have clue if this is possible, or how it might be done. Granted, there are alot of contributing factors. Maybe a table could be derived from a set of printed callibration models. I started logging my modeled hole sizes vs the printed result and found that the variation dimishes on fairly predictalble curve as the hole diameter gets larger. Thanks for humoring me... Your thoughts? Matt

Bill, I started a pretty lengthy disdcussion on the Google group about HBP before I built mine. It was a discovery process for me as I took the advice of others and found my own solution borrowing form the ideas of others. The thread is fairly lengthy at this point, but I think there is a lot of good info there. The thread is recent enough to still be current, I think almost everything there is still relevent. Basically for my HBP, I am using a thermistor connected to the UM board to sense the temp. This requires adding a resistor to most versions of the PCB. I have an SSR (solid state relay) connected to the heater output terminals of the UM board. The relay switches an Omega (brand) self adhesive 120VAC heater. The heater is stuck to the bottom side of a 1/4" mic6 (cast and machined alum too plate). I have the aluminum set up to accept interchangeable glass plates (std window glass 3/32") which are clamped there with black document clips. In this setup, the heater power is isolated from the UM, so there aren't any issues with the UM PSU being too short on wattage, or dealing with replacing it with a larger one. I bought a power input module and a std computer power cord to plug into the wall. The input module is fused. This set-up gets up to 70C in about 4-5 minutes and once there, holds a steady temp. There are more details on specifically which items I used in the Google discussion. I understand that alot of people will say that using AC power to control the HBP is dangerous. If I was dealing with 240?VAC like the power mains are in Europe, I would agree. SInce you're in the US, the risks are lesser. I don't see this being any riskier that using a toaster oven. I think my last post over there in the Google thread expresses my disappointment in the HBP. Although my bed functions very well, it is not the panacea for print sticking that I believed it to be before I built mine. I admittedly need to experiment some more with it. Most of my prints are smaller and I have been using blue tape on the glass but not heated recently. I have recently read that blue tape heated works very well, and that a glaze of hair spray on the glass can solve the non-stick issues. Edit: Disclaimer: I am not claiming to have the best setup for the HBP. Only one that makes sense for me and holds temp well. With some more experimenting, it will probably have the utility HBP's are made up to have. One observation about the new bed is that the printing surface is dead flat compared to the acrylic sheet that comes stock. This was a vast improvement. Coupled with a widget to fine tune the Z home switch, I can print a plate of Cura planner objects in confidence without being there at the start of each piece. Also when I changed the bed out, I went to a 3-point leveling system that also helped immensly for bed leveling. The aluminum is stiff enough it doesn't need 4 points, and 3 points is geometrically simler to level. I just noticed a couple days ago your interest in my hot end that I posted on Thingiverse. I am proud to tell you that I am still running the nozzle as documented there. It has never jammed. If not for changing between nozzle diameters, I would have never touched it since installation. With the .4mm nozzle, I can print up to about 100mm/sec. Since I usually keep it under 60mm/sec this is fine. Higher speeds beat the UM up uneccesarily and machin time is not that critical to me. If your'e still interested in trying out a nozzle, we can work something out If you want to discuss that or the HBP, I'll PM you my number. I need to add some photos of my mods, and get my Z switch adjuster up on Thingiverse. Matt

Nether, It looks to me like the extruder bolt isn't positioned with the deepest part of the teeth on the filament. I can't tell because in the picture, the filament is in the way. Sorry, it looks like you corrected this. Can you take a sharp closeup photo of the drive bolt out of the machine to see if it looks up to spec? What is the diameter of the filament you are attempting to use? It should be no larger than 3.05mm(.120"). A bit undersized is better. Measure it before any teeth marks are added, and measure it in more than one direction and in more than one place. Under normal conditions, there should be almost no restriction (or none at all) in the bowden tube when loading filament. Do you feel any binding or resistance when you insert new filament? To be sure, use a piece that has no teeth marks. When the filament is on the high side of the diameter tolerance, feed marks can make the filament diameter functionaly larger and this can cause feed resistance. This can be exacerbated by over tensioning. With your hot end at 230 or higher, can you extrude filament by hand (and with out the feeder engaged) with a little force? If everything is clean and free like it should be, you should be able to push filament thru the hot end with a couple pounds of force. Depending on this simple test, you can tell better where to look for the source of the problem. I don't see any pics of the item you printed, and I haven't seen where anyone has considered the idea that maybe your extruder is over feeding and causing the filament chewing. Did you calipbrate the feed? and/or what is you M92 E__ value? I am not running the stock extruder but I think its normally something like M92 E862 if memory serves me for the stock set-up. Someone else should verify that number. Are you sure that the filament you have is PLA, and not ABS? Does it smell sweet or acrid when it melts? Where are you located? Maybe there is someone nearby that might be willing to take a look in person. I hope not to send you on a wild goose chase. When troubleshooting anything, you have to start with the basics, and don't assume anything without direct observation. Be patient and we can help you get rolling. Matt

J4ysen, I looked at the BOM that you linked to. It is pretty sparse on information. It lists qty's and part names and nothing else. No manufacturers or part numbers as far as I can tell. If you have done some foot work on manufacturers and part numbers, please share them. I and others are particularly interested in the heater element and the thermistor. Thanks, Matt

Bertho, Your last points were among my design objectives. I wanted to reduce the number of connections, and improve the connection integrity. Inspiration was drawn from industrial hydraulic flared tube connections. I felt the melted zone should be smaller. It can't be too small, the melted volume needs be balanced against the therm trans rate and the flow rate so there is always a buffer of melted plastic based on the highest desired flow rate. I don't know how to add a clearance between the insulator and the heater block. This would reduce the melted volume, and I'm not sure reducing it more would be helpful. Maybe the sides of the cuff on the heater block could be relieved? Matt

Sorry for the delay getting back on... 1.I haven't measured the temp of the peek near the top, but I it has not shown deterioration or weakness. The PEEK in the stock set-up is really just as close to the heater. With stock, the PEEK surrounds the brass piece, and is flush to the heater block. I think the longest single print so far has been just over 3 hours. 2.printing at 220C, the divide between the solid and melted PLA is somewhere between the top end of the heater block (cuff) and the bottom of the PEEK. This is hard to measure. I estimated by pulling the filament back immediately after a print and measuring. I am pretty sure the melted zone does not go above the height of the cuff on the heater block. I don't find oozing to be a problem. It may be improved because the melted region is smaller than stock. One of the goals I was aiming for was a smaller melted zone. I don't think it has hurt speed as I have printed up to 90mm/sec with a .4mm nozzle. 3.The picture shows the orifice much longer than what I actually running. Currently, I have them drilled out to what I believe to be the maximum depth. The small diameter is apprximately 1/32" or less. 4.I am changing nozzles between prints by exchanging the tips. Your suggestion could work. As shown, it is designed to "drop in" to an otherwise stock carriage. If the PEEK were threaded full length on the OD, a double jam nut set-up could offer an easy method for leveling dual nozzles. Matt

By the way, I am also running Geo Hagans Extruder drive.

I think my nozzle design addresses most of the OP's points - specifically 1,2, 4 & 5. Regarding #3, it does not leak. The bowden tube to nozzle junction is based on a high pressure industrial flare fitting. I can tell you that it has been trouble free since first usage. Surely the original equipment would have failed several times since I made the revision. The heater block (Alum) is threaded through with an M12 or 1/2" fine thread. The insulator (PEEK) has an M7 thread on the ID to thread the bowden tube into. The bowden tube is first threaded with an M7 die, then screwed through the PEEK, then flared with an automotive tube flaring tool. The PEEK is then threaded into the heater block where the nozzle (ALUM) is then tightened against the flared end of the tube causing a compression fit. Zero leaks. Zero plugs. I can change from .4mm to .8mm orifice in about a minute. The nozzles have a tapered lead from the filament diameter to the orifice diameter to aid flow. I have printed at 90mm/sec with a .4mm hole and am using the standard Marlin PID settings for the thermocouple feedback loop with temperatures just as stable as the OEM parts. I am not waiting for Ultimaking to fix my plugged brass tube. http://www.thingiverse.com/thing:20572 Matt

Daid, I haven't printed for about a week or so. Don't have the current SPP running, so this comment may not be relavent. I am going to wait until your next udate that might be coming later this week. I haven'tt done much printing with support turned on. Just tried to use it thru your beta2 version. I don't see any way to control the support fill density. The default looks like it might be 50% fill. Anyway, I think it probably could be alot less most of the time and it would be nice to be able to define it. It would save some print time and material. Don't know if you've already considered this, or maybe implemented it in a more recent version. Another very minor comment; I am running dual screens on the machine I run SPP on. As a default, the program starts on my primary screen. If I drag the program to the other screen, the pop-up help tips appear on the first screen where the window was. This is in Win7. Like I said very minor Thanks for everything! Matt

Thanks

Does anyone have any experience or feedback pro or con with aluminum extrusion nozzles?

You mean NF has more then 260 options found in Skeinforge? Well, I must humbly defer. I will try to clarify with some opinion. Skeinpypy is more straightforward and easier to understand compared to Netfabb. Skeinforge in its native format has more options than Netfabb. The user controls are of Netfabb are comprehensive but more organized than the maze within in Skeinforge. The controls in NF are not as tedious as those in SF, especially because some kind people have gone to lengths to make some standard profiles for us. My preference is with Skeinpypy for now. I hope to have done justice to the facts. I invite corrections. Now before I put another foot in my mouth... Matt

I've had my machine for just under a month. No one with more experience has answered you, so I will. I paid the extra for Netfabb, and have used it. It does slice very fast. AFIK, it does not have 'comb' feature (see below). Currently it does not allow connecting directly to my machine to print - supposed to, but does not work. This is a well known problem. I haven't seen any discussion about printing from NF except that it doesn't work. You have to use another program to upload the gcode to your machine. The NF user interface is more complicated compared to the Skeinpypy & Printrun routines. The high complexity of set-up options (in NF) may offer more function - if you really understand what they all do. The NF learning curve is a bit higher (still a nube on everything here really). The few prints I have made with NF & printrun had really good quality, but have alot of stringing. There is a feature called 'comb' in Skeinforge (part of Skeinpypy) that forces the gcode to make non-printing moves over printed areas - not over voids. This makes the prints come out alot cleaner with much less clean-up time. There is supposed to be a new version of Netfabb released soon. I would think these issues will be addressed. Knowing what I do now, I would hold off on NF and see what they come up with for their next version. I would learn how to use my machine with open source tools (arguable as good) while shaking it out mechanically. Things are evolving at a quick rate. I could not have known the current state of affairs when I bought my machine & NF at the end of Dec. The price of NF is the same if your buy it with your machine or separately so what is there to lose? Matt

The readout panels seem to have a cool factor, but does it add any function. I already have a separate pc that returns the diagnostics I need thru ppronterface, rep, or netfabb. What could I gain from having one? Matt

Here's an Idea. I realize the extra motor gives twice the torque, but why not just use one motor, add a couple of pinions to the drive train so the 'idler' hobbed wheel can become drive, slaved off the main drive wheel. This could be done just by adding a pinion on the same axle as both hobbed parts that mesh when the clamp is closed.