[UM2 Pre Purchase Advice]

I would suggest: if possible, have a small PLA test print made *while attending it*. Watch and listen carefully. Then it will be obvious if everything is okay: bed and nozzle heaters and sensors, display, fans, feeder, stepper motors, all bearings still working, belts,...

Even with the original feeder, you can get excellent results, on the condition that you manually unwind and stretch very hard filament (like PLA), so its bending radius is not so tight anymore, and it has less friction in the bowden tube and nozzle. This already allows you to start and get all the rest right (bonding to bed, slicing your designs, etc...)

[PLA curling on one side only]

Try printing as cool as possible. Also, try putting a table fan in front of the printer, after it has printed the first two layers (do these without fan for good bed adhesion). I found these things help, although they do not totally solve the issue.

But too much cooling might cause temp errors, if the bed- and nozzle heater can not keep up with the loss.

[Consumable parts?]

The belts can also start to squeak. Which is really annoying.

Yes, but you can handle that with a bit of chemically inert silicone grease. The sort of thick white or yellowish silicone grease that is also used in binoculars, camera's and microscopes. And this thick grease does not leak and does not splatter around. But do not use petrochemical oils or greases, as they might chemically attack the rubber. Gently wipe the edge of the belt with it.

[Printing a Nylon part with dissolvable PVA as support material]

Nylon absorbs a few percent of water, but then it should stay in a sort of equilibrium. And it definitely should not dissolve or decompose. A lot of clothing is nylon, and you can wash it without it falling apart. Also in industrial all-weather applications, you often find nylon blocks and rods being used for bearings. I have seen them here in automatic bridges from ships to the wall (dutch: "loopbrug").

Are you sure you have real nylon? I seem to vaguely remember that there do also exist some sorts of "biological nylon", based on plants instead of on mineral oil, which might indeed decompose? Maybe you need to google into that?

Or maybe you use incompatible lubricants in the system? In model trains I have seen nylon gears swelling (just swelling so they would not rotate freely anymore, but no decomposing), but that was due to incompatible oils.

Also, maybe you could ask your maintenance staff what they use in the industrial water? Usually this is just plain tap water and collected rain water that is stored in a big tank and recycled, with a desinfectant added (chlorine?) and or with UV-light treatment to kill bacteria? It shouldn't be corrosive, since then it would eat away all rubber seals too. Unless you would work in a recycling environment where they do clean sewer waste water by using plastic eating bacteria, and your watertank has been infected with these...

[Problems printing simple cylinder geometry using PLA]

These photos show what I mean (see my post above). This is a tiny cone: 20mm wide, 20mm high, printed in PLA. With default settings, deformation suddenly occurs when the layers don't get enough cooling anymore.

When printing at a lower temp, and with a dummy tower next to it, this deformation occurs much later. But you still can't avoid it.

[Problems printing simple cylinder geometry using PLA]

This seems to be a very tiny model? If so, excess heat might also contribute to the deformation, I think.

When I need to print very small models, and when there is a sudden change in surface area to print (from a big surface to a very small), I also see deformations. This seems to have to do with temperature: different cooling times between layers do show up as deformations. And not enough cooling time in very small details shows up as blobs and overextrusion, similar to the one you have here.

I am guessing now, but if you print the base at high speed and high temp, and then the printer has to slow down considerably to print the tiny part (due to the minimum layer time setting), you may have too much pressure in the nozzle, and too much heat accumulated, before the system has time to adjust to the new lower speed and heat requirements? Could that be the technical cause?

Anyway, I reduce this effect by printing a dummy block (e.g. a tower of 15mm x 15mm) next to tiny models, to give the printed layers enough time to cool down, before putting on the next layer. And I print as slow and as cool as possible.

If you want to get really fancy, you could give the dummy tower the inverse shape of the real model: make a cube, put the original in it, subtract the original from the cube, so the dummy cube now has a hollow in the shape of the original. Then fill the bottom of the cube with a thin layer of 0.5mm, for good bed adhesion. So, now each layer area is the sum of the cube and original model, and it always has the exact same printing surface and cooling time (except for the bottom layer, but good adhesion is more important). This method may pay off for very small models.

If there is a little bit of overextrusion, or junk accumulated on the nozzle's outside, or you are printing rather fast and hot, or if the material flow setting is higher than default, that worsens the situation in my experience.

This does not exclude other possible causes, such as Z-problems, of course.

[Cleaning x/y axis rods?]

Maybe you could try non-abrasive kitchen cleaning aids, like those used to remove burnt cooking rests in ovens? They work in a chemical way. Just make sure they do not get into the bearings, as they might be acidic or caustic, which could attack metals if staying on it too long; or they could contain solvents, which might attack the plastics around the bearings.

If you want to polish, you could consider using a soft polish for plastic lenses? These should be far less abrasive than metal polishes.

Anyway, try them on a small spot in the least used corner first (left back?) Or even better, try them on a separate piece, like a dirty coin.

Edit: I just realised: in a car shop, you may find plastic lense polish. This is used to polish plastic head light covers to fully transparant again, when they have become dull due to the sun, or when they are scratched and sanded by sand and stones thrown up from the road.

[Ultimaker Original+ Hotend issue]

First do cold pulls or atomic pulls as the others said (search for the manuals on this site). Or if you want to do a more gentle pull, try my method. This method does not require brutal pulling, but works by cooling deeper, and by gently wiggling and rotating the filament prior to pulling. See:

https://www.uantwerpen.be/nl/personeel/geert-keteleer/manuals/

Then you should be able to see through the nozzle from above, or from below with a mirror: light must be shining through.

Inspect the pulled-out pieces: are there bits of residu or foreign parts in it?

I also sanded down a needle (one with a glass ball, so I have a good grip) to a diameter of 0.39mm. Normally they are 0.50mm here, which is of course too wide for my standard 0.4mm nozzles. Then after doing an atomic pull, I gently (!!!) poke through the nozzle from below. This also helps removing accumulated residu in the nozzle tip. But do this very very gently: steel needles are much harder than the brass of the nozzle, otherwise you would damage the nozzle.

[Ultimaker Original+ Hotend issue]

Could it be that the nozzle gets blocked, so the melt has no other way to go than upwards?

[Ultimaker 3 Facebook Group]

I don't like the idea of information on one topic being fragmented over lots of sites: that makes it a pain to find anything back later on. So, info on Ultimaker stuff would best sit on the Ultimaker site, especially since they are very open to all sorts of discussions. (It would be different if discussing problems and solutions was not allowed, but this is not the case.)

Further, I don't like "Faecesbook". First of all, you can not access the info without logging-in. And if you log in, you make yourself vulnerable to all sorts of data- and identity theft (since that is Faecesbook's business model).

So, I will not join. This is my viewpoint, but of course, you are free to see things differently.

[Zero Gravity Extruder - V1.0]

I like this direct drive concept with the motor separated from the drive gear and print head. This removes a lot of weight and inertia from the system. This should be the direction to go for the factory UM2++ or UM2+++, I think.

Have you tried magnesium as the drive shaft? This would weight less, but I do not know if it is strong enough, and if it is safe enough (as magnesium can catch violent fire when hot)?

Another question: what is the purpose of the long slider with coupler, connected to the print head? Wouldn't it be possible to connect the square shaft directly to the extruder gear (with only a small conversion piece to convert it from square to the round hole of the drive gear)? But without this long block? This would save a lot of weight, and complexity, I think? I guess you will have a good reason for doing it the way it is now; it is just that I don't see it.

[general forum feedback]

...

If I search for replace PTFE the first hit is a resource page that tells you how to disassemble the hot end. If I search for replace teflon, it is the second hit.

...

 

Yes indeed, you are right; I just tried it now.

I had not tried this when writing my previous post, but since you said earlier that it only searched the forum, I had simply *assumed* it would not search the manuals. My bad. So everything is okay for me.

[What about different traveling speeds over land/air?]

... Slow travel path for combin moves will cause the drip to increase and unless someone makes a math formula to actually know how much drip/mm/heat/viscosity is lost, you will get a bit of underextrusion when the extruder resumes to print.

 

Hi neotko,

This week I first encountered this phenomenon you described, when printing in PLA at a bit higher temperature. Good prediction!

So it seems to happen occasionally when the material is quite liquid due to higher temps, and when there is a long travel through air. Never had it with my PET up till now, since that stays more rubbery like chewing gum, and doesn't leak so much.

So traveling through air definitely benefits from a fast move. While traveling over land for PET and PLA at low temp, seems to work better with slow traveling speeds, to avoid the "bits of morse code".

[First layer probs]

Have you measured if the first layer really is 0.1mm thick? In the first photo it looks much thicker, almost 0.3 to 0.5mm. Which could mean that the nozzle is too far away from the glass. (But if you had set the first layer thickness to 0.3mm it could be okay, of course.)

Also, did you clean the glass plate with soap, window cleaner, white spirit, thinners, methanol, or other cheap cleaning stuff that might leave a soap- or oil-residu, which could destroy bonding? I think it is best to clean only with isopropyl alcohol and pure warm water.

Further, I would rather start from 210 or 215°C for the first layers, and then if desired tune temp down. This too gives a better adhesion of the first layer. But if it has worked a 1000 times for you at 200°C before, this shouldn't be the problem.

[What Is The Best Low Budget Printer kit?]

I think it would be a good idea to have a test part printed on various machines, via a 3d-hub service, or via hobby clubs in your environment, and see which models deliver the quality you want.

Instead of buying a cheap assembled model, you could also consider buying a do-it-yourself kit. This will generally give you a bigger and better printer for the same amount of money. But this only works if you have enough technical insight to assemble it yourself, of course. If you are not confident in assembling such a complex thing, it could become a nightmare.

[general forum feedback]

For me the search works fine now since it is Google-based.

But I could imagine that searching through the manuals and help-section too would be useful, not only the forum. For example if someone wants to know how to replace a teflon coupler.

[Help with printing]

At first I was also confused, since some parts in the original photo seemed okay, some seemed gaps and underextrusion, and some seemed blobs and overextrusion. But when seeing the new photos, and when seeing the original in Photoshop with only the blue color channel (blue is opposite of orange, and shows more contrast), then it is clear that there are lots of gaps. So I guess it is underextrusion. If one layer is underextruded, then of course the next layer will sink into the gaps left by the previous layer, causing an irregular surface.

Could it be that printing PLA at 0.2mm layer height and 60mm/s is a bit too fast or too cold for this combination, so the filament has not enough time to melt? This maybe in combination with tightly wound filament near the end of the spool, or with a worn-out teflon coupler?

What happens if you print it at 215°C or 220°C?

[Long print filament/extruder stops but keeps moving. Next print fine?]

For nylon and ABS this is a well known problem: in a few hours they may absorb enough moisture from the air to make printing or injection moulding impossible, or with very poor results.

But I hadn't heard about this for PLA yet. Although it does degenerate over time (months, years) due to moisture absorption: this breaks down the polymer molecules.

I store all my filament in big boxes with a disseccant with color indicator (blue = dry, pink = moist). I found in a car shop: to dry car interiors and to avoid condensation on the windows. So it has a high absorption capacity.

There are also a few people who have made boxes with a spool holder, and with a small exit hole, with disseccant, so they can keep the filament dry even when printing. Maybe that would be a solution for you? Just find a suitable polyethylene food storage box, drill a few holes in it. Then design and print a spool holder and fix it in it.

[Filament waste---what to do with it?]

I once made a filament welding tool like this:

Procedure to weld: cut off filament ends at 90°. Then put both filament ends in the groves, and hold them down firmly with your fingers. Heat a metal knife in a flame, put the hot knife in-between the filament ends and push them onto it, melting both ends. Remove knife, and slide your fingers towards each other so the molten ends bond. Then wait until cooled down, and with a Dremel tool remove the excess flanges. It worked well, feeds correctly, but still, I didn't like it, so I don't use it anymore...

[Dual printing notes?]

I don't have a dual extrusion machine, so I can't comment on material (in-)compatibilities.

But you could always design mechanically interlocking features into the models, to improve strength of the bonding interface. Then even if it doesn't bond, it can't fall apart. And if the design permits it, provide a few tiny holes at the bottom where you could drip in cyano-acrylate glue along the interface line. It's not optimal of course, but it might be sufficient for showing prototypes.

[DIY Polishing Station]

If I understood things well, this works in a chemical way, very similar to acetone smoothing for ABS, but only with different materials, alcohol soluble? If so, it could be interesting to extrapolate this concept to work with acetone and other materials too. Just make sure the pumps and housings don't melt...

However, if you only need to polish objects without fine details, I think a polishing wheel on a drill works a lot faster and gives better results: this takes only a few minutes and gives high gloss. For example, see this video (or google in Youtube for "polishing dental retainer"):

 

For smaller items you could use a tiny polishing wheel, but of course it won't get into very fine details without sanding them off. So it would be perfect for door knobs, clamps, pen holders, big RC model airplanes, car parts, and similar stuff. But not for HO-scale model trains, or tiny figurines.

[2 stroke engine]

Yes, model engines are fascinating. Sometimes non-functional demo-models, but sometimes also really working models in metal, or metal with glass cylinders. Some of these Youtubers have made fantastic things.

Also, I just thought about it: there do exist companies that can 3D-print in metal. Some only for decorative items (thus low dimensional tolerances) such as Shapeways, but some also for industrial applications with higher accuracy. They often use this for injection mould making.

So what you could do is design the model, print it in plastic first to test if everything works as desired (no valves or cylinders hitting each other etc...). This will save you a lot of time and money. And then have it printed in metal. Don't expect it to be cheap.

But I still doubt if it can be done without a lot of post processing. Industrial iron or aluminum cast engine blocks also require a lot of machining to get them within tolerances.

The best known industrial 3D-printing companies in Belgium are, as far as I know:

Melotte: https://www.melotte.be (specialised in metals)

Materialise: http://www.materialise.com (specialised in really big prints, mostly plastics, for car manufacturers)

[Printing PET: less strings, hairs, blobs with lower traveling speed?]

The ICE PET is more flexible than PLA, so in applications where it has to flex a bit, such as keyrings or snap-fit clips, it works better and has a longer life.

But thin plates or structures do sometimes fracture indeed. So I don't think it is stronger than PLA, only more flexible. And it is transparant, which may be usefull in some cases, for example to make watermarks, rulers or logos totally inside the model. So I use it mainly for these purposes. In other cases the transparancy may be a disadvantage.

I also printed around the same temp and speed: 25 à 30mm/s and 220 à 230°C. Printing a bit hotter and without fan improves strength, and improves transparancy, but not very much.

[2 stroke engine]

Any combustion goes way over 1500°C. You could make an educated guess of the temperature, based on the color of the flame: compare it with color temperature charts like those used in photography or interior lighting. Plastics can only handle about 100°C maximum, and PLA even only 50°C. And plastics do burn very well. Further, they don't provide the required cooling. So there is no way you could make a functional combustion engine in plastic. You will have to go for the casting method.

You might be able to make an engine running on cold compressed air, provided that it has very good lubrication, low friction, and low RPM. But I guess it won't be easy and won't last long. Might be good for a demo-model, not for real use.

[The Clumsy Noob Goes Supersized in the Noobverse]

I like the "P" and "S" features, for easier alignment and assembly. I have to remember that concept.

Are the two middle "vertical colums" separate parts? If so, I would print them in a separate batch. Then, if something would go wrong half way a print, you don't lose everything. And you can already start post-processing the first parts when the next are still printing.