jameshs

What Steve said and more! Before I bought my UM1 (I now have 2 :cool: ) I hung out on the fora of the larger printer manufacturers - if you want to see grief go to the Makerbot Gen 5 threads! I think you will see that 3d printing is a mix of art and science - kind of like getting a great hand tool, but still needing many hours of hands on experience. With this analogy you could go the route of a really low cost printer to learn all the basic skills - but you will soon hit the capability envelope of the printer. But at the same time, your first few weeks with your new UM printer may also look like pants while you learn the hand skills to use it properly. I still learn stuff every day - like the vibration of having 2 UMs next to each other can produce interesting artefacts! Few of the threads that start out angry end up angry because the problems are solved - look at the smart extruder experience of other machines! Naturally I am biassed as an owner - but then I would also be more critical if I was not getting the results out of my machines - in use everyday. There are others that come close to the UM (The Lulzbot) but have less developed communities and other drawbacks (moving bed). Let us know what you decide! James

So I did the same as you for a year and one of the most important things for me was the community that had built around the printer. MB and UM had WAY the most active and helpful forums (fora?) But for me the UM had way more demonstrated reliability - like MB did before it went closed source (I think the flashforge is now the nearest). I have now 2 UM classics and having had one for over 10 months I look at the newer printers and do not find any killer improvements - so bought another one second hand as the only negative is that I never have enough time to print everything Being able to alter the machine to do what you want is a real plus for me - I mainly play with nozzle size and build platforms etc. I will put a heated bed on the second when I get time, but have not so far needed to as PLA is still delivering what I want. (funny how no matter what experimental filament I buy I switch back to PLA) I think Nylon will be useful so I am thinking heated bed, but also have Bridge which I think may not need it. I am continually stunned at what a cranky looking lash-up can print - but then realise that the thought and engineering that went into the design is really good - and the 'look' is deceptive - these are really industrial machines. Sure they break, need tinkering, need skill, tips, technique - but so do all my high end industrial machines - which is the difference between a craftsman and a bodger So I was not biassed at the beginning but chose UM classic and AM now completely biassed James

I generally find that 0.1 in .1mm printed objects gives what I call an interference fit. 0.2 is generally a bit loose James

your youmagine page is marked as private. I think that your top/bottom layer is responsible for most of the artefacts. I would go for Xmm bottom/top thickness height where X= the height of your proposed object I would have 0 fill - this essentially produces a solid object as you have been trying to cover the top with a o.4mm thick layer of molten plastic! For something this small I print cool and slow - cooler than your 220 - something like 210 if your machine can produce a noodle at that temperature. And slightly faster than 10 as you don't want that hot nozzle on the layer too long - say 20mm/s change all the speeds to 20. then change the minimal layer time - I never remember which way, so I raise and lower it in Cura and see what the build time is - if the time goes UP then it is taking longer so this is the right way - basically on a small object you want to ensure that the nozzle gets clear of the filament to let it cool. (Though I don't fully know if that is the right direction ) Have a go with some of those settings - but the main one is to go for 100% infill (by changing the top/bottom layer thickness.) And the temp and see what happens. I tend to write the settings that I am printing - so 210/20 for 210 degrees 20mm/s and you have a permanent record of what is going on. James

Which UM do you have? On the UM1 I always wipe the hot nozzle and then spin the extruder wheel to purge the nozzle until it hits temperature when I whip away the extrudate before it homes into it Not sure you can do this on a UM2 James

With any of these machines you can get air prints for a whole load of reasons - you need to get to know your machine, research, research research and try lots of stuff, and then you will learn how to avoid them (I say avoid because printing is as much an art as a science and there are times when the machine just decides to make art rather than what you asked it Most often air prints are running too fast for the filament to catch up (i.e. out-running the possible rate of pushing the plastic through - and this is a combination of head speed, temperature, filament and layer height. running the filament too hot Feeder grinding and these happen on any printer. I am no MBI fanboy - but I think you have a user error there and can play with it to fix it - but you should have bought a UM )

Here are the images of the prints that I hang from - one set either end of the pole and I do chin-ups and pull ups with no sign that they are going to break yet so not just strong in static loads - but dynamic too. The white one is an example of a partial print with the infill showing - from memory this was 20% infill. Colorfabb PLA/PHA mix I do use quite a thick perimeter - these were early prints in my printing career - I would probably go thinner now James

Will do when I get into work tomorrow - I also have some part finished showing the internals. James

I have a UM1 and wanted printing out of the box - so bought it assembled. The current price is fantastic compared to other printers. I don't have a UM2 (I am only 8 months into printing) and currently don't yearn for one as I understand there is less tinkering opportunity than on the UM1 - but it would be interesting to hear from those who have gone from UM1 to UM2 to get their experience. I would love a heated bed - but think that will come in time with a UM1. I still marvel at what a UM1 can print - I think the UM1 and 2 compare well on print quality. On the 1 I have drilled out nozzles to different sizes and don't think this is possible on a 2? - I find it fantastic for prototyping at different speeds for different parts, and for nozzle clogs I have a few spare nozzles for fast swaps. I believe (and hope) that UM will support the 1 as long as the 2. I have a glass build plate on the 1 which works well. I do not print ABS on the one as I don't have a heated build plate - but materials like colorfab XT and their new xt range are beginning to narrow down the benefits of ABS Be aware with ANY of these printers there is a steep learning curve - models, software, tips, tricks and techniques. I regard myself as reasonably competent and learn everyday from this forum - some 'aha' moments, some investigations that are of no benefit (lexan buildplates) But I love every minute of it, and the UMs are a sound basis on which to learn, and this forum a good one to learn in. So YOU need to narrow down why you want the printer and what features the UM2 would have over the UM1 to justify going up - two that I would consider are it is MUCH better looking (living room material) and MUCH quieter than a UM1 - mine is in a workshop and I like hearing when it has finished - so I still would go for a UM1 If I was getting a second printer I would get a UM1, if a third I would get a UM2 Hope this helps! James

do you have an example of the solid version as this one would end up in a confused machine. I feel for you as I print mainly architectural models - and I either use shrinkwrap or really think hard about what my wall thicknesses are - If I am doing a large building with 100mm thick walls, and using a .4mm nozzle then I need to be aware that at 1:100 my walls are 1mm wide at 1:200 they are .5mm wide and Cura just does not see them. So my models for print are different to my models for build and I am always aware of the scale that I am printing at (like a gutter for example where the material is 10mm thick in real case is not going to actually print!) So I think cura is behaving itself and the model is not currently ideally suited to a .4mm nozzle FDM capabilities. James

find a local 3dhub www.3dhubs.com to print your model on a UM2 Normally a hub will really know how to use their machine so you will get the best if can be and make a judgement on that. Note - I am a hub but get nothing for recommending them - there are other like makexyz but I think in europe 3dhubs is better. James

except it is a valid point - you merely said cooling - and I bet the 'double fan' has both on the X axis - i.e. the x axis is double cooled and the Y axis not at all except from the x axis. I only noticed this when I was asked t print a 'marvin' as a sample for 3Dhubs - and I found that the right hand eyebrow was different to the left. As this was a print sample to prove print competence I was keen to improve it and watched the printer like a hawk - the eyebrow nearer the fan did not suffer from (a) initial cooling lag and (b) reheat from the returning nozzle (next layer) So I turned the model on the buildplate so both eyebrows received the same fan axis and voila - they ended up the same. So IMHO there is a x/y difference from cooling in the fan axis, and having two would probably maintain/magnify this -artefact - possibly in the initial deposition cooling and then in the reheating from the next layer. To experiment - drop the fan speed and see if it has a difference, or fans off - anything to make the x/y less different. so IMHO a good observation! - follow the differences. James

I print quite a few 'structural' parts in PLA - the toughest support my body weight as I have a chinup bar that is supported between two bookcases - the top bar is scaffold pole that sits in a triangular PLA foot in either end (to give it a flat bottom. These are in compression - so no big deal. From the scaffold pole I hang a timber broom handle - so 48mm diameter to 24mm diameter pole using a kind of tear drop shape. The offset from the pole diameter is about 10-15mm to make the object as fast to print as possible and they are 20mm deep - I printed with 20% infill as when I did 10% I got some tortion. I did 1.2mm top, bottom and walls. Then, from the pole I have D handles - like triangles, so I can exercise different muscles and these are again in 25mm flat profile and 20ish mm deep with rounded edges (radiussed). Have been using these continuously for almost 8 months and never had a failure or felt unstable. I did try to print the triangles with a 25mm opening in it (i.e. so I could just pop it onto the pole) and they did not survive - the tortion was just too much and they failed - the second I closed the loop their strength was phenomenal. A set I printed prior to a nozzle block has severe under-extrusion and were squidgy, but they still managed to hold me until I twisted them to failure (deliberately) So with the right 'design' playing to the strengths of the material PLA printed can be incredibly strong. This was before I drilled out a nozzle to 0.65mm which would really drop the print time, and started using Colorfabb XT which would allow me to reduce the cross sections. I also print wedges that go under my feet on a rowing machine and in a rowing boat. I use 0.8mm walls and 10% infill to keep the weight down (weight is important) - I have done 300000 metres on the rowing machine and 200000 metres in the single sculling boat - in the outside rain etc (and the boat is stored outside) with no failures. So you just design test to failure, change and print again - steel it is not, but I can't print steel:) Am about to try nylon bridge which may allow some thinner profiles, but I find every filament type takes about a month to get the best out of! Will post photos if it is useful? James

I don't have a UM2 but do have a UM1 Hypodermic syringe needles are so far the most reliable for me - they tend not to break. It will sound daft, but I would think that you should heat up high and gently push another needle in after the pin - to properly push the pin into the chamber. For the filament you can either heat a pin (fold over a paper clip), push it into the filament melting as you go down, and then wait to cool - then gently pull and it should help hoik it out. it is going to take you time and be frustrating, but try and avoid anything sharp James

Wow - clever stuff - can you tell me what the IP relay that you used is and the app? Thanks James

Likewise I don't have a UM2. But from UM1 Experience; Scrape off the remainder with a window scraper (with a razor blade) but do it gently as you can score the glass. Then wash the glass - the glue under should mean that it all washes off. If you are using PLA to print then only a slight bed temp (or none at all) and a light coating of glue will work well - trial and error to work out what works - and taller prints tend to need more as they can warp a bit - also things where you have a large infill and the nozzle is rattling around - then I go really Sloooooow and only a light amount of glue is needed. If you can get some spare glass that is good too (I run two sheets and use dilute PVA glue solution which when dry helps stick) Lots of experimenting! James

I am UK based and have to admit that I almost exclusively use colorfabb - once you find a supplier where the filament just works I guess you tend to stick to it, and the shipping cost is 'ok' if I buy a few reels at a time (and I do ) and I LOVE XT - it is my "go to" filament for most things that don't need colour. The email notification and delivery tracking is good. Personally I like spools - but may try loose at some point. James

Just ran it through netfabb and it was not a god model! The fixed version slices perfectly in Cura. So a mystery as to why inventor can output a watertight model in one thread type and not in another ..... but I have a workflow now so thank you for all your help. James

That certainly works on the solid 'bolt' but when I then introduce a hollow central section it just fills in the whole thread making it solid :( is it something leaky in the stl - strange that they are all made the same way and that they behave differently depending on pitch.

https://www.dropbox.com/s/le5aqf9dmk9slir/test%203.0.stl this is the one that works In the more reply I don't have an attach button and can't add to my gallery so hope the dropbox ink works Thanks James

https://www.dropbox.com/s/1pxnx0tslt81cfn/test%202.0.stl'>https://www.dropbox.com/s/1pxnx0tslt81cfn/test%202.0.stl https://www.dropbox.com/s/1pxnx0tslt81cfn/test%202.0.stl does this work? - a link to dropbox rather than an attachment?

Happy to but I can't work out how to attach a file - do I upload it as an image? grrr

hmm - so I did aclean version of a 30mm cylinder and then applied a thread with a 3.5mm pitch 3.0mm pitch 2.5 mm pitch 2.0 mm pitch and then popped them on the board to slice. The 2.5 and 2.0 fail, but the 3mm + slice fine. These are solid cylinders so something happens when the threads hit less than 3mm. Something in the thread is too small for cura and makes it think it is hollow. Any clever fixes I am missing or do I have to go back to trying ti figure out ISO thread tables to make a 32 3.5 thread definition ? TIA

So the odd thing is that the only difference between the two is the thread pitch. Once that slices successfully has a smaller wall thickness, but larger pitch of thread. Both seem to be large enough, being 2+mm (and the 2.1mm printed well, but the 2.5 is not. Could it be that there is too much - I think cura is being confused by the threads or that the higher frequency of thread somewhere has too small a part so cura sees it as an outside part? Has anyone printed a thread with a pitch of 2mm. I went to 2mm pitch from 3.5 as I am making a pole mount and the outer ring squeezes the inner ring to grip (like a telescopic paint handle) so the OD of the inner piece is 30. When I printed the threads as 30od and the nut at 30 id as an experiment they did not fit. The next ISo defined thread in inventor is 33 @ 3.5 which printed well, but was slightly too loose for my liking (workable and printed, but I am trying to push the tolerance) So as there is no defined thread at 32 3.5 pitch I changed the 30 od to 2mm pitch and the nut to 32 2mm pitch, which I have defined - and that is when the 'bolt' wont slice but the nut sliced and printed fine. So I think it is the pitch of the thread. I don't know enough about thread definitions to 'make' a 32 3.5 which is what i want and none of the auto table generators seem to match the definitions in the inventor thread xls tables. Anything I can d to fool cura (I don't have ANY fixes ticked and when I tried them they just sealed every hole available!) I will try pushing it through a few mesh fixers and see if they can find any holes, but inventor seems quite good at clean meshes. Thanks for the suggestions - I enjoy hacking away at these items and am continually impressed at the usable details that I can print into functioning things. Autodesk inventor has really opened up the opportunities but hence increased the challenges James

I have an odd one. I am printing some threadded items they are fairly big - 30-32mm diameter and cura won't slice the bolt. Strangely yesterday it happily sliced the bolt when it had a bigger thread pitch, but today it puts a cylinder inside and then the thread is a cobwebby mixture. the nut printed fine! The threads yesterday had a bigger pitch - 3.5, and the ones today 2.0. Have I hit a limit of slicability? - I sliced as low as 0.1 but same thing, though I am using a .65 nozzle for rough work. The wall thickness inside the thread is good and these are coming from Inventor - should I try and netfabb them? Thanks James