It has been a long journey failing and printing over and over to be able to print at faster speeds. All I did mostly was trial and error. I am currently trying to wrap my head around a problem that has been asked by user:1998 (mhelvens): Can we manage to make a uniform formula for printing very fast with high temperatures? (Increasing hotend temperature to compensate for increased filament throughput)

Using Ender 3, Hero Me Gen5 with E3D V6 volcano, mdd kit 1.2, Klipper firmware

When I start printing at 215 °C (recommended temperature for my filament) everything is alright, because of initial layer speed...

When I come to 100 mm/s the print clearly fails, the plastic doesn't stick, it even swirls up and is just too solid and then it damages the nozzle when it bumps in to it...

If I start at high temperature, the print is failing at the start and then doing ok, which is equally as bad

When I have high temperature and high speed the print wont stick at the beginning...

The solution in my opinion is to gradually increase speed with temperature, adding that as a function of the firmware. Also probably involving filament flow % and pressure_advance...

Is there a formula for what I'm asking?

Can we implement it in the software or does this have to be done through trial and error all the goddamn time?

  • 1
    $\begingroup$ At that speed, you may be barely melting the material enough to flow, but not enough to stick. Try upping temps to compensate and let us know. $\endgroup$
    – Davo
    Commented Aug 19, 2020 at 15:42
  • $\begingroup$ @Davo Are you referring to the second paragraph, because I tried upping the temp and it resulted in oozing, cracking, bubbling at start when the speed is slow... $\endgroup$ Commented Aug 19, 2020 at 16:02
  • $\begingroup$ @MatthieuMorin My comment was pre-edit, but I believe an approach as suggested in FarO's answer should work. You can set a temp for layer 1, and another temp for layer 2, etc. $\endgroup$
    – Davo
    Commented Aug 19, 2020 at 17:15
  • 1
    $\begingroup$ @Davo, yes indeed, I tried it, however I've done a mistake in choosing a thermistor. I will fix that up and come back to this issue $\endgroup$ Commented Aug 20, 2020 at 9:56

1 Answer 1


There are no ready solutions. Check on Klipper Github to see if someone requested that, but I doubt it has been implemented.

100 mm/s with 0.8 line width and something like 0.3 mm layer height results in 24 mm^3/s. It's quite high already, since the Volcano is rated at about 20 mm^3/s.

What you could try to do is to take the GCODE you have, calculate the layer time and the total path length for each layer to get the average extrusion rate, and introduce at the beginning of each layer a M104 command with a temperature dependent on said rate.

For example, you could add 1 °C for each mm^3/s of extrusion rate starting at 15 mm^3/s.

If you want, you could also calculate the average extrusion rate not every layer but every 50 mm of extrusion. In this case, better anticipate the temperature command by one stretch, since it takes time to adapt.

Otherwise you need a Supervolcano, or a slight overvoltage of the hotend (remember that power goes with the square of voltage, so do not go above 10% overvoltage).

  • $\begingroup$ I have asked on Github already (github.com/KevinOConnor/klipper/issues/3211). Thank you for the info! I didn't know that it was only 20 cubic mm, this might very well be the root of my problem... I will try to manually change the gcode of my test print how you described and then change if necessary. Is there any way to make a command so I save time on changing it all? Already ordered the SuperVolcano, because I'm planning to use a 1.0 mm nozzle in the future... $\endgroup$ Commented Aug 19, 2020 at 16:30
  • $\begingroup$ I found out Volcano actually has about 35 to 40 mm3 max flow. I am going to use the changeatZ 5.1.1 postprocessing cura plugin to grudually increase my print speed from 60mm/s to 166%, my temperature from 230 to 250 (I think this will need more oomph... I also think I have a broken thermistor..), and fan power from 127 PWM to about 180. $\endgroup$ Commented Aug 20, 2020 at 8:48
  • $\begingroup$ E3D themselves tested the Volcano only up to 1176 mm^3/min=19.6 mm^3/s 3dprintingindustry.com/wp-content/uploads/2019/03/Capture.jpg I would say that above that reliability of the extrusion is not guaranteed. It seems unlikely to me that you can reach such high value on a hotend which is just a little longer than the standard V6, which is rated at 10 mm^3/s. $\endgroup$
    – FarO
    Commented Aug 20, 2020 at 9:53
  • $\begingroup$ I just got to 19.32 mm^3/s, more than that and I experience slight layer separation, which increases with more flow rate, however I think it is tuneable, but to a minor degree. The 20 mm^3/s is kind of a sweet spot. @FarO I saw the same picture and based on it my starting profile configurations. I think I could get to about 25mm^3/s while sacrificing strength and quality. $\endgroup$ Commented Aug 20, 2020 at 15:36

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .