Jump to content

ascended

Dormant
  • Posts

    63
  • Joined

  • Last visited

Everything posted by ascended

  1. I apologise in advance for the number of posts that follow... with the limit of 3 attachments per post this is the best I can do Each set of pictures shows the square in focus, the circle in focus, and from the side to show the layers. Attached - print at 220 Celcius, 50mm/s print, 4.5mm retraction at 50mm/s. I also changed the minimum speed to 25mm/s (up from 2mm/s). Double skin is off, 0.1mm layers, 0.1mm initial layer, 0.8mm wall, 200mm/s jumps.
  2. Thanks for the link futurejames, I can't see that in any of the categories! I wonder what other useful parts are in the store but not showing up Via the cheapest postage option for just 1 tube: Grand Total Excl. Tax €28.50 I might have to find somewhere to source these locally, as even at crazy australian prices I might be able to get it cheaper Daid, Yes i'm starting to get better results for the print which is heading back towards what I had initially. I'll hook a PC up to the ultimaker and run the Z axis calibration test again as the filament tolerances are really awesome on this stuff from BilbyCNC. I find it interesting that it appears to be over extruding yet it cant print the fine parts (we're working on a redesign of the part to try to bring it within the envelope of the printer). I'm not too fussed about retraction but i'll try the settings you mention. I might load some different coloured filament in the printer to give better contrast on the blue tape. It's hard enough taking photos of tiny parts with a macro lenses depth of field hand held without it all blending in What are your thoughts on initial layer size?
  3. There is still some slippage on the material, however with the lighter pressure on the extruder drive as Daid mentioned, it recovers. There is also a picture of a lug for a PCB mounting that is demonstrating the same failings as the thingiverse part.
  4. These new photos are all narrower for Daid All of these prints are at 215C, 0.1mm layers, 0.1mm base, double height skin, 50mm/s, travel speed 200mm/s - in Cura The first print is a 3 hole test that was posted on the forum. I also tried this: http://www.thingiverse.com/thing:22263 . These settings are with 0 retraction and the top parts of the pieces really go to crap, they dont seem to have enough density or something. Quite a few prints I'll need to do have similar features, so this is fairly critical to me if anyone has some ideas.
  5. Ok so now I have two Eclips on the tube, one at the upper wood piece, one on the lower. Hopefully this will provide some redundancy. The new plastic came in from BilbyCNC and the first lot i grabbed out varies from 2.98 to 3.01 along its length, and its very circular. It has a totally different feel to the stuff that came from Ultimaker. The PLA from ultimaker was an elliptical shape, and therefore very difficult to measure - I thought this was normal. I've got the extruder drive with just enough pressure to extrude I believe, so we'll see if it slips up. This bowden tube is starting to become a bit short i feel, so I believe if it slips again i'll have to throw it out and go back to my spare.
  6. Considering I can't even get a part number out of them, I doubt I'll get a replacement part from them. It's not listed on their store so i'm not sure where even get one. I'm hoping to get some from BilbyCNC in the next few days. Well, this is for a business so we can print prototypes and small volume parts. With our chargeout rates and the amount of hours we've spent on it so far we're into 5 figure costs already (if we consider the time spent on this and not doing consulting as cost).
  7. Daid, you are a star Thanks for the in depth response. That's where you started to go wrong. I made the same mistake (even after months, and people calling me an "expert" and everything). You had a plug starting, right at that point. Your bowden tube has moved up a bit, a plug starts forming, and then you need more force to push PLA trough it. Very interesting, in this case with constant extrusion we're seeing a plug form in the first layer or two... Machinging the brass? Where does it say that and who added it? It's a bad suggestion IMHO. Drilling the end of the tube will make it worse, as it makes it easier to form plugs. On this page: http://wiki.ultimaker.com/Jamming#Hack_ ... nt_jamming The way we did the tube makes this a very tight fit/hopefully seal and allows the tube to move up about 1.5mm before popping off completely. When the tube is too tight, it compresses at the end, making it harder to filament to get trought, and making it easier for a plug to push the tube up. I suspected we might be seeing some concertina action there and slackening off the pressure did seem to help (from using a tool to tighten the thumb screws to just getting it finger tight.) I'll keep the light pressure on from now on. The white clip? Did you replace this clip? IMHO, it's the source of most troubles, it's not gripping enough.I agree, you can see the gouges in the tube showing its just sliding along the surface removing a small amount of tube surface in the process. I dont have a replacement clip, and ultimaker support still have not told me a part number or where i can expect to source one from. I looked through all the drawers which hold similar such parts at Masters (a new hardware warehouse chain here) but didnt come up with anything - they were my best bet for something local - hence going for the eclip instead. Stick to 50mm/s until you have everything in working order. I still print at 50mm/s to get the best quality (but I have play in my linear bearings)Print at 230C until you are sure that everything works, ignore the people that say you should lower the temperature, that will only make things worse. A lower temperature will put more pressure on the tube, same for higher speeds. I must admit, our first tests were up at 220C (which the sticker on the PLA says not to exceed) however we had some pitting in the skin and overhanging/angled areas looked really terrible where the fan couldnt cool it fast enough. The threads on the bowden tube clamp on thingiverse also came out terribly at the higher temperatures as they became very "spongey" from not cooling down. I'll try going back to the higher temps and go at 50mm/s for further testing. 3.72? That's quite thick for 3mm filament that should be between 2.9 and 3.0. Variations of more then 0.1mm won't work with the current design (they are working on an upgrade to fix this)Sorry that was a typo - it was meant to be 2.72 I'm seeing variations on both spools of +/- 0.3-0.4mm within a 30cm area which makes it very difficult to know what to put into Cura. I have some more PLA and ABS hopefully arriving from an Australian supplier today or tomorrow. They were shipped almost 2 weeks ago, but Australia is a big place and we're in Perth (most isolated capital city in the world.) Hmm, i see them just fine even through a small remote desktop window using Opera, it even gives me a scroll bar Sorry, but the replicators are also not without problems, even pre-assembled. The price I cannot argue about, but it largely depends on where you live. Shipping a replicator from the US to Europe is not cheap, shipping an Ultimaker kit the other way around is also not cheap.Unfortunately I'm in Australia and have to pay large shipping fees for everything It's not an optical illusion, it's the machining mentioned earlier as the wiki article suggested. I suspect i have one of the bad white clips then. Where did you get your second one from? Considering we plan to print ABS as our main material I can only imagine our issues are going to get worse if we use a white clip rather than an e-clip. Ultimaker should hire you Daid, you are the most helpful person on these forums!
  8. Another couple of pictures of the clip slip. The filament is also shown in its heavily compressed state. This makes the problem of filament feed slightly worse.
  9. I have to admit that the Eclip took a huge amount more force than the standard clamp ever did. The part it was printing had a huge amount of fill with no gaps to allow pressure buildup to escape. This meant the varying tolerances of the material eventually overpressured it.
  10. Showing the modded brass tube in the peek and stripped teflon tube.
  11. More photos, the filament shows a compressed pattern as it tries desperately to push the filament out of the extruder.
  12. I'm feeling pretty dissatisfied with my purchase of the ultimaker. After reading these forums, and seeing peoples comments about it initially it seems like an excellent product. However having assembled it and then run into no end of problems it seems like i'm not at all alone. I've been waiting 10 days for ultimaker support to get back to me, and have tried pretty much everything I come across on the forums and google groups in order to get something to print, or at least - to finish printing and have some modicum of quality. When I got the printer, we had it up and running in about 5 hours - I realised it was a kit build so this is fine, everything went together well even though the instructions are rather terrible (eg: the part may be black, possibly white, you may not have it at all... if you do have it then do this *ambiguous instructions with poor photo*). Once we had it printing, it worked amazing for about 3 hours. Tightening the heck out of the belts got nice circles eventually and we were able to calibrate .the x/y axes for their differences (as we would with any CNC machine we have). We would expect this with any machine we build or purchase (and we do have a few CNC machines). Then, we started to get the extruder drive stripping away material and then free wheeling in the grove it made - no problem, tighten the material down until its flatting it off as it drives it. This then puts extra pressure on the bowden tube - and caused our first plug as the . From there on, we haven't had a single decent print. What we've tried: * Machining the end of the brass rod down, and drilling the end of the tube as suggested in the wiki so as to get a better seal. * Printing at higher temperatures (220c). * Printing at lower temperatures (190c). * Printing at every temperature in between (in 2c steps). * Cutting the end off the tube time and time again - tightening it hard, tightening it barely finger tight. (with the long M3 thumb screws). * Varying the extrusion rate (by changing the step count in the ulticontroller) as it prints * Cleaning the plastic clip's metal "teeth". * Replacing the whole head assembly, including tube (lasted about 1hr). * Replacing the whole bolt assembly on the extruder. (We bought a full set of spares of the extruder system to be able to go dual head later.) * Printing an alternative clamp designed by a user (8 tries later, we still dont have one) * Using different material. * Re-measuring the material and reslicing,... over and over and over. * Slotting the tube and putting a 1/4" E-Clip into the slot (we got about 5hrs of printing out of this, however the quality is terrible) - retrying all steps above with the eclip. * Printed from 10mm/s to 150mm/s. Most testing around 75mm/s which seems relatively safe when compared to other CNC equipment we have. Ultimaker support blamed it on poor quality PLA when I called them, among other things - I received the PLA from them, and I know it's not from the bad batch because they told me it had an issue and delayed my shipment 12 days as they switched the colours over. The tolerances on this plastic are pretty bad (in my opinion), it varies from 3.72 to 3.18mm - the translucent green is even worse. Support also suggested the wood may have been damaged during assembly, so there are pictures attached to show this is not the case. Other pictures attached to show the various parts/issues. Now, I'm utterly out of ideas. This product should not be sold with an issue that is plaguing so many users, looking at the forums and groups it seems as though pretty much everyone is having these issues. For the same price as the ultimaker cost delivered, we could have had a pre-build makerbot inventor delivered - had a heated bed and dual extruders (if not the speed - but what use is speed if you cant print!) I'm really hoping that someone can give me some options we haven't tried here, which don't require printing. I'm really thinking something needs to be done with the telfon tube, it's teflon - no wonder the clamps are slipping on it! Very frustrated at the moment after having spent around 120 hours trying to fix this printer.
  13. I meant that perhaps the hex file which is included in the installer is faulty, downloading and flashing a build from Marlin Builder works perfectly and does not come up with those messages - and the ulticontroller works very well. Hence, I think perhaps the hex that cura flashed onto the machine has a problem.
  14. Ulticontroller is also heating when loading a file from SD card. Might I suggest you look into the firmware you include with the latest download of Cura Daid? It appears to be very very flawed :( I've almost spent as much time trying to get this working as it did to build it!
  15. I flashed on hex files from Marlin Build and now it works (or at least, it's heating). I'm going to reassemble the electronics and try to run the full print.
  16. Print run shows this: Connecting...startPrinter is now online.echo: External ResetMarlin 1.0.0 RC2echo: Last Updated: Marlin-Mon, 11 Jun 2012 17:33:42 +0200 - b67dacdc8f1bd489e058e16d92ba29c364b2a8e5 | Author: Build-Me-Marlinecho: Free Memory: 4494 PlannerBufferBytes: 1232echo:Using Default settings:echo:Steps per unit:echo: M92 X78.74 Y78.74 Z533.33 E865.89echo:Maximum feedrates (mm/s):echo: M203 X250.00 Y250.00 Z5.00 E45.00echo:Maximum Acceleration (mm/s2):echo: M201 X9000 Y9000 Z100 E10000echo:Acceleration: S=acceleration, T=retract accelerationecho: M204 S3000.00 T3000.00echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum xY jerk (mm/s), Z=maximum Z jerk (mm/s)echo: M205 S0.00 T0.00 B20000 X20.00 Z0.40 E5.00echo:PID settings:echo: M301 P22.20 I1.08 D114.00echo:SD init failecho:Unknown command:"start"echo:Unknown command:"echo"echo:Unknown command:" External Reset" Loading a gcode file: the print goes from 81.4 mm to 118.6 mm in Xand is 37.2 mm widethe print goes from 81.4 mm to 118.6 mm in Yand is 37.2 mm widethe print goes from 0.3 mm to 15.0 mm in Zand is 14.7 mm highEstimated duration: 312 layers, 00:09:03 Clicking Print, i get: Print Started at: 18:29:26 And the printer sits idle. There is no heating at all. Printing is at 0%, printing line 0 of 21694 lines. Attached is my gcode.
  17. No, it doesn't heat up at all. If I preheat it, connecting to the printer reboots it and it starts to cool down.
  18. OK so connecting with printrun causes a reboot too. The LCD goes blank and the drives go on for a second then off again. (atmega pulling up on reboot.) So now i'm not sure what to do, i cant connect to the printer, and the ulticontroller does absolutely nothing.
  19. I'm using the SD card that came with the ulticontroller (2gb sandisk, pretty sure it's not a SDHC). Ah, could be the problem - I can't even do the calibration of the printer. I'll try printrun tonight, failing that I'll put the stock firmware back on and try it with inferior software
  20. I have my ultimaker with ulticontroller built, and am trying to get a print to work. I can autohome with the ulticontroller, move the head around manually and extrude just fine as well as heat up the extruder all with the ulticontroller. This is the limit of what i'm able to do though! I'm running the Marlin firmware that Cura flashes. When I first tried to upgrade the firmware it failed, I had to remove the arduino board from the ultimaker shield in order to get firmware back on the device. If i try to do anything over USB with Cura (ulticontroller still attached) it just resets the ultimaker and times out - it is never able to connect. If i generate a gcode file in Cura (in simple mode, or after fiddling with settings), and then from the watch screen: [press button] -> scroll to "Card Menu" -> [press button] -> scroll to file -> [press button] - it takes me back to the watch menu with the file name on the bottom row, and absolutely nothing happens. I've tried multiple files, preheating the hot end, clicking "autostart", just about every menu option - however I cannot find any documentation at all on the ulticontroller or how to use it. Is anyone able to point me in the right direction for getting this thing going? At the moment the only thing really working is the fan cooling the electronics. Everything else appears to be working fine, but it just doesn't print. All the troubleshooting information i've found assumes you're having issues with the print - not getting one going.
  21. A liquid pressure sensor or gas pressure sensor could be used if it's housed correctly. The biggest issue is going to be in housing it in such a way that it's away from the heat and raw plastic (as that would foul the sensor). The biggest gain i see in the pressure sensor is when you want to back the head off. Rather than dwindling the amount of plastic in the head and guess at volume - you can use the pressure sensor to slowly drop the pressure down so the print continues but the volume is low enough that when you back it off it doesnt drool all over your work I'm sure that with the right control software/slicer, you could control the head pressure based on the feature you are printing, and do some look ahead too. As with all things, more data/control points (and using them correctly) results in a better final product. Be it a weather forecast, engine monitoring or 3d printing
  22. R/C servos and servo motors for industrial control are not the same thing, don't get the two confused An industrial servo is usually a DC motor, or even a stepper motor, with an encoder on the back. They are extremely accurate and do not lose steps, as it has the quadrature to be able to tell it how many steps it has moved in which direction to extreme accuracy. Pretty much all industrial CNC machines will use servo motors, they are not that much more expensive (about 3x-5x) as steppers, but they do need a lot more control hardware. Even a cheap servo will have 1000+ lines in its encoder giving you 0.36degree-1.44degree per line, depending on the encoder implementation. Some encoders can even get more accuracy out of these encoders.
  23. I was thinking that flexability/sag wouldnt really matter as you would need an insulating or heat reflective surface underneath which would provide that support. If a high temperature epoxy was used to seat the PCB against the insulator, it should not move or flex. If the board was made of high temperature FR4 (which i doubt they are) it would be even more stable
  24. I hate to make yet another heated bed thread when there are quite a few up the top, however I can't find one where this question fits well. Is there a reason why a piece of glass or aluminium is required on top of the PCB? Sure, it adds thermal mass to the system but the standard heater plates have a solid copper base which would spread heat out well right? The PCB should also be almost perfectly flat if it was made properly and should theoretically (at least in our simulations) maintain a more even heat across it's surface. So, is there a reason which can be mathematically proven for needing the glass or aluminium, or is everyone doing it simply because other people have done it - and therefore thats the way it's done. Not being condescending here, I just see a lot of this happen in many things and am trying to understand if the same thing is occuring here. Has anyone tried printing directly on the PCB (well, with polyamide tape over it) and had adverse results?
×
×
  • Create New...