donmilne

Now that I use Robert's feeder I just blow away any dust with one of those aerosol duster cans. An alcohol wipe might be useful too. Of course on the original feeder you have to get it open first.

I do wonder if people in the USA understand how much the imperial measures system hurts their place in the world. When I was a little boy it was commonplace to see "Made in the USA" stamped on everyday items. Not now. The last time I bought American I learned to regret it the first time I had to replace a sheared bolt and discovered it wasn't a metric bolt I could buy in any hardware store - I had to order it from an internet supplier. Seriously, that experience was a factor in my decision to buy Ultimaker and not Makerbot. Not the dominant factor, but it contributed. The irony, I think, is that I'm pretty sure Benjamin Franklin would have been first in line to adopt the metric system if it had been proposed back then. It sort of half was - the USA had a decimal currency very early - and then for some reason they turned their backs on it.

Actually, the UK saw the light (the candela?) quite some time ago - starting with decimalisation of the currency in the early '70s the process was completed for all practical purposes I think by the mid 90's. And not only was it practically applied, it is (or was) legally enforced - shopkeepers could be fined for continuing to sell produce using imperial measures. The only major exception seems to be the roads, which are still marked in miles, and hence car speeds, still measured in mph. I think that's partly because no government is inclined to find the funding to change all those road signs! There may also have been safety concerns - people not grasping speed limits, especially if road signs were in km and older car speedos still reported in mph. A speed of 50mph is significantly faster than 50 kph!

If I might interpose in English: nobody seems to have commented on all the debris inside the feeder. That looks like a particularly large piece in the upper hole? I suspect that when filament is withdrawn for changing, a lot of junk gets dragged back through the Bowden tube and is deposited inside the feeder. I think this debris can sometimes cause problems, e.g. if a particle gets trapped between the knurled drive gear and the filament. One big advantage of Robert's feeder is that it's open and hence easy to keep clean.

Sorry, I don't speak... Dutch? I just wanted to let you know that you currently have your images marked as private, meaning we can't click to get a larger view.

I had a similar problem a while back when trying to print a metric threaded hex nut and bolt. The bolt came out fine but the central hole in the nut got filled in. The solution was to turn off the "fix horrible" features. If your 3D design tool creates proper watertight models then these fix features are not needed. Unfortunately, SketchUp is one of those that often fails to create watertight models, but you can download a plugin (Solid Inspector) that checks your shapes for correctness. My original discussion here :- http://umforum.ultimaker.com/index.php?/topic/4548-experiment-with-threads/

I see, thanks. Seems a tad over-complicated to me.

Flex3Drive mod?

Thanks.

Hang on a mo... I took "merged ... in Cura" to mean that Cura did the merging. Is this about one STL input file that contains two objects?

@Cleven, I'm not quite following this thread. Can you please expand on what important changes you have made to the spacer, what problem you think this addresses, and why?

You can drag it with the mouse to wherever you want. It will shove other objects out of the way. Bear in mind that if "one at a time" printing is selected then it imposes quite a large spacing between objects.

On a related note (similar to the wires getting trapped), that hole in the top of the hotend (for the dual extruder) is a real trap for the unwary. Working in poor light I poked my fingernail into the bowden clip and pulled it back, and lost it down that hole. How was I supposed to see a black hole in a black painted blob of metal! :-) Nor could I reach the clip with tweezers etc, I had to take the head partially apart to get it out. Took me about ten minutes to get everything back in place. Then I started again on the head cleaning procedure I originally intended... and promptly lost the clip down the hole again! :mad: Tip: rotate the fingernail edge on the clip so it doesn't invite you to drag it towards that hole!

Thanks for a replies, however as I mentioned in my first post, I cleaned the nozzle using the cold pull method just before the print, so I'm satisfied that isn't an issue. I'm also not convinced that deformation of the Teflon part is relevant either - I felt no obstructions while cleaning. I'm using Robert's feeder - which I'd rather not be distracted into discussing as I've already discussed that issue plenty in another thread: I think I've got a good handle on how to get it working well. I've pruned my area of investigation down to a fairly narrow question: from what source(s) does the back pressure on the filament originate? and why (if my observation is correct, which it may not be) does it seem to increase over the course of a print, so that skipping gets worse and worse? (or alternatively, is max motor torque reducing?). Actually I do have an additional thought: if I understand correctly, the UM1 didn't skip, it would grind the filament instead, killing the print, and possibly other problems too (head overheating?). Now that I know a little more about how stepper motors work, I'm guessing that the UM2 design used a weaker (lower torque) motor in an attempt to fix this issue. In fact it just replaced the issue with a different one: heavy underextrusion producing a useless print. I'm thinking the correct solution is a stronger motor, with multiple points of contact on the filament so that one spot won't get chewed up. I'd rather not go down the drive belt route - belt failure seems like a likely common fault. I'm thinking some kind of geared mechanism.

Surely 5 minutes is an exaggeration? What's the real value, and what print speed, what material, what temp? It just seems a big disparity to talk about 40 hours and 5 minutes in the same breath.

Both of those sound like brilliant ideas. Might drilling a wider bore in the teflon part reduce both issues?

Hi, thanks for the response. 1. I've already moved to Robert's feeder a few weeks back. It seems to be working well, but doesn't prevent the skipping problem, which seems to be purely a matter of torque on the motor. 2. The bowden tube seems fine to me. It mates well at both ends, and in any case I don't see why damage to the bowden tube would get worse during a print. 3. As I mentioned in my first post, while I was cleaning the print head using the atomic method, I didn't notice that the teflon (a.k.a. PTFE) part offered any obstruction. In any case I've only had the printer a few months and it has not been heavily used. I don't expect that it's seen more than 30 hours of use. I was thinking more along the lines of slight overextrusion causing a backlog of filament inside the head, but I couldn't explain why that wouldn't just cause the filament to exit faster.

I've noticed that, quite often, my larger prints will start off quite well, they can get 90% through without any problem, and then it will start skipping. The skipping problem will gradually get worse and worse, it's a race against time to where the print either finishes (with defects), or I'll abort it, or the extruder will stop altogether. I assume the motor is skipping because it doesn't have the torque to overcome the back pressure. Can anyone suggest what a gradual back pressure buildup might be caused by? I don't believe there is any blockage: I tried the atomic cleaning method, some gunk was indeed removed, but it didn't cure the problem. At the same timing I felt little resistance for the bit of filament I used for cleaning, so I doubt teflon deformation as well... and in any case, why would that get better on the start of the next print? I don't know why a pressure buildup wouldn't just result in a bigger volume being squeezed out?

Does anyone know how far up the teflon hole the PLA/ABS is still molten? I know from experience that if the head overheats then it melts right up to the bowden tube (and makes a lip there). But what about normally? I'm wondering if a wider bore would reduce the pressure, provide a larger reservoir, reduce underextrusion issues. OTOH, the larger the diameter of the molten end, the more likely there will be problems removing filament for material changes?

You can also reduce the torque requirement in a couple of ways: reduce the print speed and/or increase the temperature by 10C. See if that helps. A warning too: if it's been grinding filament then the bits can get caught inside the feeder casing. There's a long discussion thread about design improvements for the feeder: I've switched to IRobertl's design: it's easier to maintain, easier to keep it clean of debris, and easier to tweak the tension on.

You'll need a time component as well. I.e. the printer will execute the task in a given period of time. Hence the feeder needs to be able to deliver that volume of material in the same time or better. It's a subject that's been interesting me of late. Problems such as underextrusion are caused by a disparity in this timing (there may be underlying hardware issues that exaggerate the disparity). The normal way to deal with overly tight sync constraints is to have some kind of buffer between the supplying and consuming processes. Perhaps a reservoir of some sort which is kept topped up regardless of the current consumption rate. Of course a larger reservoir would make filament color changes a bigger deal.

This stuff looks like it might be an interesting alternative to PTFE :- http://www.precision-ceramics.co.uk/macor-machinable-glass-ceramic.htm

Microstepping? I didn't know that was possible, but I'm still reading that link from yesterday. [Edit] Oh I see. As I understand it, instead of switching each pole independantly (0->1), you can balance two poles.

Can someone elaborate on where this 282 steps/mm filament comes from? I've been trying to dig up the specs of the UM2 feeder motor, and basically have not found anything explicit. I did find specs for the z axis motor which says it's 1.8deg/step. Are all 4 motors the same? Elsewhere I find that the feeder knurled wheel is around 7.95mm diameter, so is the following calculation not correct? One step = 1.8 deg, so a full rotation is 360/1.8 = 200 steps. One full rotation pushes pi*7.95mm filament, ~=25mm. So steps/mm = 200/25 = 8. That's a bit different from 282. Where am I going wrong? Normally I would look for some kind of gear reduction, but if the motor includes a gearbox I see no mention of it.

Thanks for the replies and the links guys. It looks like I need to do some research. That "step motor basics" site looks especially good.