How often do you need to replace the PTFE tube?
Are there any telltale signs that you need to replace the PTFE tube?
Recently I started to have many failed prints, due to clogging/jamming of the filament inside the hotend. I unclogged the nozzle (cold pulls and needle), but it helps for just an hour of printing and then the clogs come back again.
Any piece of advice?
For 4 years I had to change PTFE tube just twice. Once when upgrading to MK2.5S (new design of extruder) and second time after the big filament clog caused by cooling fan failure (broken wire). In this case filament was melted high in the hotend. So the ptfe tube was significantly damaged by long time pushing melted filament inside.
On the opposite side, you should change the nozzle quite regularly, so maybe the issue is not PTFE tube but the nozzle.
Thanks for your answer. I will follow your tip and replace the nozzle (it doesn't cost anything to try). At the same time I am doing this I will also open the extruder head and visually inspect the condition of the PTFE tube
Sure, this can be caused by many things, but more probable is nozzle than ptfe tube. I believe the recommended time for nozzle change is about an half a year. But depends on what filament and how often you print.
I have never replaced the PTFE because of failure of the PTFE, only because I've had the extruder apart and did it as a preventative measure.
I opened the extruder this afternoon and extracted the PTFE tube. It was immaculate. I changed it nevertheless.
I also changed the nozzle. I believe it was grand time to change it (see picture).
I had at hand only a hardened metal nozzle, that I believe requires 5C increase in all temperatures.
Calibrated the first layer correctly and resliced the test model (the Tree Frog) with the increased temperatures.
The result is far from optimal. The under extrusion is still there, at lesser extent, but now I have a massive stringing problem, which is unusual with PLA.
I am running out of ideas.
Sorry, just forgot to mention a detail. I hear a repeated "clack" noise from the extruder (every few seconds).
Could it be the stepper motor missing steps? Or the bondtech gears slipping over the filament?
Take a look at the gear on the idler door. The metal axis holding the gear is just fixed to the plastic part. It is quite short and can be loose on one side. In that case the extraction is wrong.
Clicking is from the extruder motor trying to drive filament into a clogged or under-temp nozzle. Or, just trying to push more plastic than the hot end was designed for. And then some plastics are difficult to push - too viscous - even when properly melted.
Generally, many underextrusion issues are self inflicted - in nearly every case the user never reads the E3D or Prusa hotend assembly manual nor the nozzle replacement guides. These guides are absolutely required, each step precise and necessary to prevent printing issues. Miss one step, or even the nuance intended by a step, and, well, you get what you get.
Does your assembled hot end nozzle look like this?
If yes, then you assembled it wrong.
Thanks for all the suggestions and the help.
I will revise completely the extruder assembly to check for mistakes. I will also remove the hardened metal nozzle and use the stock brass nozzle from Prusa, this in order to remove the additional variable.
I followed closely the nozzle replacement guide from Prusa, respecting also the 0.5mm gap between the nozzle and the heater block.
When I removed the idler door (I have the MMU2S extension), I found quite a lot of filament dust inside, but the Bondtech gears were aligned on the filament, the axle for the Bondtech idler was correctly centred and the gear that is coupled directly on the stepper motor was well tightened to the flat part of the motor axle.
One thing that I notice though is that the screw that is controlling the tension between the two gears was a bit loose. I will try to see if changing the tension helps.
It sounds like you've done a fair job checking everything. A loose idler tension can cause under extrusion, and is usually the first thing to check. The second, when the nozzle has been changed, is hot end assembled right? Is the nozzle too far in, does it meet the heat break and is it tightened properly.
Rare, but does happen: accumulation of dust and debris inside the heat break. This can be cleared by removing the nozzle and using canned air to blow air up through the hot end. I do this every six months and sometimes get a good cloud of dust.
And - trying a new brass nozzle to verify the steel isn't causing issues.
Since I am still waiting for new brass nozzles to arrive, I cleaned the interior part of the hot end and placed the old brass nozzle. This time I placed the nozzle regardless of the 0.5mm gap between nozzle and heater block, but I exclusively used a torque wrench set for 2nm. The resulting gap is around 0.2mm
I also increased the tension of the bondtech gears.
There is definitely an improvement for everything that is printed with 0.15mm layers. For everything that is 0.1mm or smaller there are signs of underextrusion. I will wait for the new nozzle to arrive to give a final verdict.
In the picture below the tree frog on the left is printed with 0.1mm and the frog on the right is printed with 0.15mm layers.
Did you ever disassemble the hot end, remove the heat break from the heater block or heat sink? If you did remove the heat sink, did you use heat sink compound on the threads when you reassembled?
And the question that most hate to answer: did you follow the E3D-V6 assembly instructions?
Your final torqued assembly should look like this:
Thanks for your message.
I assembled the MK3S with the "MK3S ready" hotend in the Prusa kit, meaning that I never opened it (it came preassembled).
But now that you are mentioning I will disassemble it and reassemble it following the E3D-V6 assembly guide. I guess you are thinking about this guide: https://e3d-online.dozuki.com/Guide/V6+Assembly/6?lang=en
Just a question. Can I use any thermal paste? The only thermal paste that I have at hand now is the paste that I use on the interface between microprocessors and heat-sinks.
I used normal old white alumina thermal paste; though the silver stuff for CPU's is probably a bit better. Any paste is better than no paste.