Wednesday, June 18, 2014

Solving my 3D printing issues




I've built my 3D printer for about 2 months now and one of the most difficult thing to get right was the extrusion.   At this point in my build, I can say that my prints are now consistent and at par (and sometimes even better) to the quality of prints I see posted online.  In this post, I will share what were the factors I changed that helped me get better prints. 


For those who were lucky enough to afford a commercial 3D printer, this might be less of a problem but I've seen people complain about the same issues on printers they've bought so I hope this will help someone. 

Temperature


There's a reason why I listed this first.  Printing in the wrong temperature can cause different kinds of printing issues from clogged nozzle to ruined prints.  Since it can be a source of a lot of problems, it will be difficult to troubleshoot the other issues so it's best if you can get this out of the way first.

Here are some signs that indicate you need to tune your temperature settings:
  • Temperature climbs quickly but slows down as the temperature gets higher and gets near the target temperature - At one point, it took my nozzle more than 10 minutes just to get to 210 degrees Celsius.  Then it takes another 3 minutes or so to get to 220 degrees. 
  • Temperature is not being maintained within the target temperature - When I start the print with the fan off, the nozzle can reach the target temperature relatively fast but when the fan turns on, it cools the nozzle down by about 10 degrees Celsius and doesn't get back near the target range.
This problem bugged me for a long time. Initially, I thought the resistor or heating element on my hot end nozzle was just not powerful enough to reach or maintain 230 degrees.  When the resistor on my hot end failed, I experimented with different kinds of resistors.  I even used two 10-ohm resistors in parallel to heat my nozzle but it was still not able to maintain the target temperature. Then I used 5.6 ohm vitreous enamel resistor which heats faster but the temperature still drops by more than 10 degrees when the fan turns on and does not recover.

As a last resort, I ordered a 30W heating cartridge thinking that it will be the ultimate solution to my temperature problem and I got really disappointed and frustrated when it didn't perform as well as I thought.  Even turning on autotemp in the smart controller didn't help.  It was only then that I started to suspect there could be a setting somewhere that needs to be changed.

I knew about PID but there was no documentation on how to set them properly.  There were some guides but nobody confirmed a specific method or setting worked for them.  Finally, I discovered PID auto tuning!  Doing this step will heat the nozzle and cool it down a few times and give you an optimum PID setting you need to use for your specific hot end.  It will not change the PID settings for you though so you will need to note down the last PID values it gives out and change the PID settings yourself either via firmware, pronterface, repetier host, or whatever you use to communicate with your printer.  Don't forget to save it in EEPROM or you will lose the settings when you power down.

Right after going through the PID autotune procedure, I immediately noticed that my nozzle can maintain the temperature within 2 degrees when the fan is off.  This allowed me to print faster, too, because I do not need to compromise speed just for the temperature to catch up.

This, folks, is how important PID autotune is.  You may even benefit from doing it once if you've never done it before.  I had to redo PID tuning with the fan turned on so that the effects of the fan will be considered in tuning the PID settings and gave better results.

If your temperature does not swing wildly anymore, you can skip the next step but for me, I had to do an extra step which was to change this setting in the configuration.h of the firmware:

#define PID_FUNCTIONAL_RANGE 20
Setting the range from 10 to 20 keeps the PID active even when the temperature drops by more than 10°C caused by the fan.  A setting of 10 was too low for my setup and this explains why the temperature could not recover after falling off the range.


Extruder motor skipping (A.K.A. kickback or missing steps)


Here's a video of this happening. You may need to wait for the full animation to load (6MB) for a smooth playback.


My smartrap 3D printer originally had a direct drive configuration but I switched to a geared extruder thinking that it will solve the kickback problem.  It did help a lot in other things but there were other causes that needed to be addressed to completely get rid of the kickback.
  • Blockage in the hot end - One thing you need to make sure is that you can easily push the filament down the nozzle while it's at the right temperature for the filament you're using. Remove the bowden tube and try to push the filament. If you find it hard to push down with your hand, you may have to have it replaced or if you made it yourself, clean your nozzle. Depending on the type of hot end, you may need to take it apart and get rid of the whatever is preventing the filament from moving inside. 
  • Nozzle too cold - If your PID is tuned well and you still get kickbacks, it could be that your filament just works best at a higher temperature.  Most PLA filament works best within 180 to 200 deg. C but the filament I bought from robotdigg just works at temperatures between 230 to 240. Any lower and the filament doesn't flow well out of the nozzle.  At first, I thought there might be something wrong with my temperature sensor until I found out that there are people also printing at more than 230°C with PLA depending on where they got it from. So you need to experiment with this.  Note down what temperature every roll of filament works best at and put a label on each roll. You may get the kickback because it's harder to push out the filament when it's not completely molten.
  • Printing too fast - Commercial printers can normally print at speeds well above 100mm/s. But for my smartrap build, I couldn't print at more than 20mm/s if I wanted the quality to be even just acceptable.  20mm/s is sloooooow. But this was when I was still tweaking the settings. In my experience print speed is proportional to the temperature. To give an example, I can print with 30mm/s at 230°C but when I print at 50mm/s, I had to bump it up to 240 to get the same quality.  I now use Cura's "Tweak At Z" plugin all the time to adjust the temperature according to print speed.  Having said that, there will be a speed at which you can't go past without compromising print quality or having the kickback so you should know the limits.
  • Wrong setting on stepper driver potentiometer - As I explored ways to get rid of skipping extruder motor, I increased the setting of the potentiometer in the stepper driver to give it more power. I could hear it push the filament harder.  I had less kickbacks but still happens and just caused my extruder motor to heat up too much. It gets hot that I wasn't able to touch it for more than 5 seconds. In some instances, adjusting this just a touch can help but there's really no need to set it differently than the other motors. I wouldn't touch this until all other factors have been checked.
  • PTFE bowden tube too long - One contributing factor on a bowden extruder setup is the friction of the filament with the tube as the motor pushes it down the nozzle. The longer the tube, the harder it becomes to push especially if there are a lot of bends and loops along the line. I cut down my tube to just enough length for the hotend's full range of motion but it's hard to tell how much of it helped in making it easier for the filament to pass through.  One way to test is to try to push the filament with your hand while the nozzle is hot.  If you can pushing the filament through without much effort, it should be good enough.
  • Flow rate too high/over extrusion - Among all of the causes of extruder kickback, this was the setting that made the biggest difference.  I use Slic3r and Cura and I need to turn down the flow rate to about 70-80% to get the optimum quality. Despite setting the diameter of the filament accurately at 1.7mm, and calibrating the extruder steps per millimeter, I still need to set it within that range or get kickbacks often. Maybe it's in how the slicer program calculates how much filament it feeds but you need to experiment with this. I am now using a flow rate of 70% and what a difference it makes! One sure sign that you need to dial this down is when the filament keeps flowing out of the nozzle after a print. This means that too much pressure is building up in the nozzle.  Before changing this setting, it takes more than 10 seconds for the filament to stop flowing after a print.  At 70% flow rate, almost no filament comes out of the nozzle as soon as the print stops. This is where the flow rate can be adjusted in Cura and Slic3r.  Flow can also be adjusted using the LCD controller.  During a print, press the dial and choose Tune > Flow.  You have to commit this value to take effect. 

  • First layer height -  Whenever I get the kickback when printing the first layer, it's usually because the nozzle is too close to the glass so the filament couldn't come out. Setting the nozzle a little higher fixes it. 


Slanting print along one axis


I've experienced prints that have one side slanting like this:
.
It took me a few prints to figure out what was causing it.  Some people thought it was the stepper driver potentiometer setting that needs to be adjusted but what fixed it for me was a lose fishing line on the X axis.  I saw people having this problem as well with belts so it is not isolated to just the printers that use fishing line. Either way, this problem has to do with the drive system on one axis where the slanting happens.







11 comments:

  1. Are you having printing issues with your Dell printer? Specifically, is the print quality not as good as it used to be? Is the printout blurry or are there smudges or horizontal lines where there should not be on the printout?

    ReplyDelete
  2. Printers can be really annoying sometimes. Not only do they refuse to print, but also stop scanning and create terrible noise if you apply force, just like an arrogant child. In such cases ensuring proper connections and checking power supply is mandatory but what if everything's fine and yet the printer refuses to work? My Blog

    ReplyDelete
  3. More and more of our customers are using their printers to print their own photos. This is an amazing time that we are living in. Not many years ago, I remember taking my film rolls into the grocery store, filling out an order form, placing both in an envelope and dropping them in a collection bin to be processed.

    ReplyDelete
  4. You are lucky to create the prints of this quality. This depended on the adherence to the certain parameters, including the temperature regime.

    ReplyDelete
  5. That's curious. I think that it would be really interesting if you'll make more notes of this type.

    ReplyDelete
  6. I like this post,And I figure that they having a great time to peruse this post,they might take a decent site to make an information,thanks for sharing it to me. fifty five print

    ReplyDelete
  7. I have another opinion, but this one is also very bright and interesting. Of course, there are a lot of points of view concerning this issue.

    ReplyDelete
  8. Great job for publishing such a beneficial web site. Your web log isn’t only useful but it is additionally really creative too. There tend to be not many people who can certainly write not so simple posts that artistically. Continue the nice writing cheap every door direct mail® postcards eddm® printing

    ReplyDelete
  9. The typical inkjet printer usually includes inkjet printhead assembly, paper feed assembly, power supply, control circuitry and interface ports. The inkjet printhead assembly contains several components. cheap 3d printer

    ReplyDelete
  10. This comment has been removed by the author.

    ReplyDelete
  11. After perusing this blog, I can state that another incredible author really taking shape as of now.
    table top 3d printer

    ReplyDelete