JohnFox
-
Posts
198 -
Joined
-
Last visited
Content Type
Forums
Events
3D Prints
Everything posted by JohnFox
-
It's all down to the quality of the bearings, the pillars and their tolerances. Someone in the forum gave me some info on high quality Japanese parts and I started to realise the difference between 'ordinary' bits and the ultimate. We get what we pay for.
-
That looks like the dreaded Z sticking problem. I had this and eventually cleared it by cleaning the two Z axis pillars completely free of oil. My theory on this is: The two Z roller bearings are supposed to roll up and down the pillars. If there is oil on them, they try and slide, causing a minor jam as the balls lock up. Since I cleaned the pillars I have had zero faults like this, ( although this is tempting the fates .
-
I use wet and dry emery paper when I want really smooth top surfaces, Start with 180grit and move up grade by grade to 400 or 600 grit. Make sure you do it on a flat surface such as a work top. If you want to avoid those straight edges on the sides of your stamp, specify a large number of segments, ( I use 99 ) when drawing the circle.
-
This was wet polishing on a machine similar to the one below:- http://www.sharmic.co.uk/index.php?option=com_content&view=category&layout=blog&id=25&Itemid=45 I did try and build one from a polythene bucket and a plaster paddle, running off an electric drill. http://www.powertoolsdirect.com/vitrex-mixing-paddle-65-x-410mm?utm_source=google&kwd=&utm_medium=merchant_pla&gclid=CjwKEAjwuPi3BRClk8TyyMLloxgSJAAC0XsjCQJ5KMY2kjIE-GRQC2vVg_KTET3JXJX6LCJ60Z1jORoC89nw_wcB The drill was too fast, I needed something like a 300rpm motor. -
I tried vibratory tumbling - too slow. However I had great success with a similar method - centrifugal disk finishing. Similar to vibro finishing but a lot more energy goes into it and it is a lot faster, typically 45minutes for a smooth surface. The machines are quite expensive, >£2000 for a starter machine.
-
ABS Print Splitting at same place every print
JohnFox replied to Waternut13134's topic in Improve your 3D prints
Grease on the lead screw is good, it needs to be there to lubricate the thread. It's grease on the two pillars that can cause sticking problems. Can't see that bed height of the first layer would affect the print many layers in. That usually only affects the quality of the initial layer and how it sticks to the bed. ABS is very prone to warping, any weakness in a layer will cause a split. I eventually gave up on ABS because of this problem. Thanks for the insight, I moved the bed a tad bit tighter to the nozzle when leveling and it looks like that solved the problem as I was able to successfully finish the print. I am however going to keep an eye out and keep this in mind if it happens again. Thank you so much for your help. PS: there was a ton of the Greese on the lead screw,so I wiped down all the excess maybe that could have helped? -
You need a Cura expert.
-
Suspect it is where the outer layer stops and starts. The rest of the print looks pretty good.
-
ABS Print Splitting at same place every print
JohnFox replied to Waternut13134's topic in Improve your 3D prints
I suspect it is a Z axis bearing problem. At that exact height, the bed is dropping down a bit more than it should, giving a weaker layer. In ABS, that layer will fail due to warping. I had similar problems some time ago, that prompted me to switch to PLA. Cured it by removing all lubrication from the Z axis pillars and getting new roller bearings under warranty. -
Not sure but it looks like the feeder slowed down or you had a blockage - which eventually cleared.
-
Could you either put in an insulating section into the rotary encoder shaft or fit an earthing 'wiper' onto the knob? ESDs aren't as common as they once were, components tend to be more robust now.
-
Although the UM2 could just about make these holes, the big problem would be the surfaces. A FDM printer, ( like UM ) has Z axis layers, ( ridges ) and these would affect the flow, creating very severe turbulence, especially at this scale. As |Robert| points out, an SLA printer should give much smoother surfaces and hence reduce any turbulence to more realistic levels. Even so, it's not going to be an easy project, you may have to consider upscaling the model.
-
Version 1.0
2,175 downloads
This was the first clock I printed. The designer is Christoph Laimer, aka 'The Goofy' who published it on Thingiverse a while ago. http://www.thingiverse.com/thing:328569/#files A much more forgiving design than his pocket watch. I used a 1.3Kg block of wood as the weight and it runs for several hours. On this build, I eventually re-designed some of the gears in Sketchup in order to get tighter tolerances, mainly on the co-axial hours-minutes-seconds gears. The frame is held by M3 bolts and the pins were stainless steel. However, brass would have been a much better material to work with. The key to get it working is to make sure the anchor and escape wheel have very low friction. All the hands were supposed to be yellow but I had to re-print them and had run out. -
Just wondering if the PTFE coupler and nozzle are exactly in line. Get a length of filament and try and slide it into to coupler/nozzle (cold) , see if it goes in OK. However I tend to get something like the blob on the end but it feeds OK.
-
I suspect you are seeing the sides of the polygon which approximates to the circle. I use Sketchup for design which assumes a 24 sided circle, which you can see the sides. I now set the number of sides to 99 and these are now invisible.
-
I've just printed the same part using Cure 15.04. Turn combing OFF - quite important. Then it prints a treat. I used translucent PLA, reckoning that it is stronger than the coloured stuff. Got the clock working for about 5 minutes at a time. It's a challenge, the holes must be drilled very carefully otherwise the drill bits melts the plastic.
-
Really difficult to explain without hands on the machine Try doing a 'Change material' from the menu. That should remove the filament, then you can cut the end off and try inserting again. It may be that the tip of the filament is too large to go into the coupler.
-
Most likely a feed problem. Sometimes the filament gets stuck in the PTFE coupler and won't push down into the heated nozzle. As your machine is new it isn't likely to be the nozzle blocked.
-
Version 1.0
1,747 downloads
Christoph Laimer posted his clock on Thingiverse a few weeks ago and I eventually decided to have a go at it. A superb design and it actually works, ( although the Swiss watch making industry isn't going to be worried yet . It's really a sophisticated design that does need some care in preparing the prints, especially when drilling out the holes. Printed on a UM2 running on default settings. Had to buy a set of micro drills specially for the job, ( about £12 on Ebay for a set of 150 drills ). Typically, for the 2mm pins I drilled the holes out to 2.1mm and got a nice sliding fit. The pins are brass rods from a model shop on Bay. Used Faberdashery Arctic White for the body, blue for the gears and translucent for the main spring. I used the translucent because I thought the plain PLA would be stronger than any of the colours. Don't forget that Christoph Laimer, ( who designed the clock ) is Swiss. I suspect they start clock making in Switzerland around the age of 5. :-) https://www.thingiverse.com/thing:1249221 -
SMALL amounts of light oil to the X&Y axes, I do this every 3 weeks or so. Remove any excess oil from the bottom of the X&Y bearings. If the first layer is going down OK, then don't alter the bed height. Remove any bits of filament from the feeder. Wipe the Z pillars to remove oil. Do an atomic clean whenever you change filament. Build a little 'wiper' to clean the filament before it goes into the feeder, it's surprising how much muck it removes.
-
Never used it but found some info on the web. Use with caution, it may destroy your print. http://www.fargo3dprinting.com/using-d-limonene-dissolve-3d-printing-support-structures/
-
I often have to print 3mm holes to take a M3 bolt. I always specify a 1.6mm radius and the bolt is a perfect fit. It isn't shrinkage, just the PLA squeezing into the space. The elephants foot on the bottom layer adds a bit more but that is only 1 layer thick. The overall size the objects I print are certainly within +/- 0.2mm of the specification.
-
Cura is the second stage in the process. You design the 3D object in some sort of modelling software ( I use Google Sketchup ), then export the object as an .STL file so that Cura can slice it. After slicing, save it as a .gcode file for the printer to work from. As well as slicing, Cura offers a very good error check with its Xray and Layer view. I check every object this way before printing.
-
That's not bad at all. A good indicator is the left ear of the robot - which is pretty good.
-
You really must use PVA glue for PLA and the heated bed. Once the object starts to move around that's the end - you get the tangle. Don't think it was the room temperature, I often run at 15C. Simply the first layer not sticking to the bed. PVA glue solves this 100%. At the default 60C bed temperature and glue, you simply cannot pull the object off the bed.
