Forum Replies Created
-
Posts
-
I am not too keen on things that fix themselves, but at this stage I’m happy to report that PLA printing is looking good on my printer.
I theorised that my Z offset might cause the bottom few layers to be “squished”, and me calibrating extruder multiplier on these layers (incorrectly). Once the print gets to higher layers I then see the actual extrusion width.
I lowered the bed in 0.2mm steps reprinting first layers, expecting to see thinner extrusions nd having to up the multiplier to compensate. But at 0.5mm lower the printer started producing good prints.
Any ideas, anybody?
I built the latest Slic3r from source, and although it now slices the small extruder gear without the “not manifold” error message, the extrusion width looks similar to before at high layers. My test print is the Hall holder as it prints reasonably fast.
Next I will try running everything under Windows (all my ABS prints ran under Windows).
The thinning of the extrusion width with height seems to be repeatable, and not random. If I re-print parts they suffer at approximately the same level, never at the bottom.
What I have done is to stop a print when it is extruding very thinly or in one case almost none at all, and immediately press the “Extrude” button on Pronterface (set at 10mm at 100 mm/s). There does not seem to be any problem like a blockage in the hot-end, it pours filament as expected.
This was done because I thought there could be something like a heat build-up over time in the hot-end causing the extrusion characteristic to change. But as I said it extrudes normally when manually initiating extrusion.
My only other thoughts are either the heat in the workpiece, which I don’t see how it could impact the plastic delivered by the hot-end. I could understand if it did not adhere to the workpiece well or something like that. My last suspect is Slic3r.
However, in software, if your software is not doing what you expect, and you start suspecting a compiler bug, you need someone else to look at your code. Unlikely to be the compiler (but not impossible), you’re missing the bug in your s/w. Same here. Unlikely to be Slic3r but if all else fails I will try another slicer.
Oh, and I still have not been able to print a good gear (either small or large) for the extruder.
I have got a good set of calibration settings, though. Circles are round and pretty much on size.
My large extruder gear slices very differently to yours – did you use Slic3r? I can see the direction of filling being different, but the main problem is the teeth – they just do not look good.
In South Africa, the filament that is most commonly available, is either PLA or ABS.
Most of the parts of the Morgan is normally printed in PLA, as it bends less and is easier to print. The exception is the print head (ABS) as that gets hot and PLA melts at a lower temperature than ABS.
I don’t know about PVC, but I have read that HDPE contracts (shrinks) when it solidifies after printing, so the sizes of your printed pieces are not as you designed them.
So unless you scale up the 3D models, by the amount that it is going to shrink after printing, your dimensions on the printer will be off. This means your bearings, pipes, etc. will not fit.
On the J-head: you can build it yourself, or you can buy it online. I paid around US$35 for mine.
It consists of a nozzle, metal body (which has a heating element and a thermistor to sense the temperature), and a plastic housing. Google to see what it looks like.
You get J-heads in two filament sizes (1.75 or 3mm) and with different nozzle sizes. I purchased one with a 0.4mm hole.
Hein
I have quite a collection of small gears now, none quite perfect. I’m playing with settings to try and get the ultimate gear. The one I installed after loosening the motor mounts is still soldiering on, and I’ve put a few meters of filament through the extruder already.
Feel free to share your settings for a solid, small gear.
Hi Quentin
I have switched to operating under Linux, and I am using Slic3r 0.9.10b. I get a complaint that the small extruder gear has a hole that is “not manifold”. Slic3r (not sure which version) under Windows sliced the same file no problem.
I have tried to fix the stl under Meshlab, the error (position) changes, but Slic3r still complains. I have uploaded and retrieved the fixed stl under Netfabb’s cloud service, and it seems to slice properly now.
Any idea what causes this problem? Have you experienced something similar?
Hein
Hi monamour
Having just completed my build, I can comment, but only about the situation in South Africa (where Quentin also lives):
People have bought printed parts for around USD25, and that sets you well on your way to build the frame and main construction for USD100.
As Quentin said, one of the things making 3D printers more expensive is the control electronics. In my case, the (Arduino Mega) controller, and RAMPS board, was USD100 on its own, and the stepper motors another USD100. Exactly half of the stepper motors price was shipping.
Stepper motor drivers, bearings and power supply can add more than another USD100. So the final price will always depend on what you have available, or can get cheap/for free (like a PC power supply).
And if you have time, it works out cheaper. If you are in a hurry, it will cost money.
Regards
HeinHi Quentin
Got your message, did it when I got home from work. It works well now, the gears were a tight fit.
Currently I am printing calibration prints (thin wall boxes), I am down 25 percent on extruder multiplier, and still printing better and better.
Hein
After visiting House4Hack yesterday and collecting new extruder bits from Quentin, my Morgan is printing!
There are still fine calibration issues to be sorted out, but I have printed small gears, one of which actually works in the extruder.
Hein
The new brass drive gear seems to works well. The little stepper motor gear is no longer with us. You gave me two (a white one and a black one), the white one seems larger and does not fit (the larger inside-teeth gear does not fit around the small gear). The smaller one, which fit, broke two teeth.
I will need a new small gear or two, and preferably another extruder body as well (I cracked the current one although that does not stop it from being usable.
How can I get those from you most quickly?
Note to self: friends with lathes occasionally have offcuts of half inch brass rod in their junk bin…
Drive gear:
I started with a threaded M10 brass bolt. Drilled it out, tapped it cut it to length, but by this time the threads were flattened from gripping it in the bench vise, and then I cut slots across the threads with a Dremel disc. The OD is too small so the gear doesn’t grip enough (clearance to idler bearing too big).
Hein
Update: looks like it fixed itself! Makes me worry.
I reversed the wiring order of the arm stepper motors (from posts on this forum), as expected everything then worked in reverse. After changing the wiring back, it calibrates! I suspect I had a bad connection on one stepper before – that might have caused the X and Y resolutions to differ.
Anyways, movement is now sorted. The extruder drive gear is not working as it should, I will re-make that.
Hein
Hi Quentin
Thanks. I’m going to have to talk to you, I’m not having success with the calibration. Full description on my blog, but the short version is – I follow the calibration procedure, at the end the printer thinks 0,0 is VERY close to the drive shaft (actually jamming against the pipe), and I cannot convince the firmware to accept a given coordinate after manually positioning the head (using the G92 command to force coordinates).
Hein
