I came upon this blog page that gives some very interesting and useful information about how prime/retract values affect the output from slicer programs. Some of it corresponds to information I’ve previously posted here, but some is different. So I decided to do a test following Michael Hackney’s recommendations.
What I did was change my slicer settings as follows:
Prime and Retract changed from 7.00 mm to 2.00 mm
Prime/Retract speed changed from 120 mm/sec to 35 mm/sec
Using these settings I printed the same test part as before; here are the results:
You can compare this with the results I posted before that used my original settings:
Why is it that the blue part has much worse stringing that then yellow one? I think the reason is something Michael did not mention, and that is pressure – specifically the pressure on the melted plastic inside the hot end’s nozzle.
The nozzle remains at a constant temperature while printing. Melted filament is forced out of the nozzle anytime the pressure on the melted plastic inside the nozzle exceeds the 1 atmosphere pressure on the outside of the nozzle. Stringing happens because the pressure on the inside cannot go immediately from what it was during printing to zero (or more precisely, anything less than 1 atmosphere.)
To understand why this is, consider what creates the pressure in the nozzle. What does it is unmelted filament being forced into the melt zone by the extruder gear. Ideally it would be nice if the pressure in the nozzle dropped to zero (or actually anything less than 1 atmosphere) as soon as a retract operation happened. But this is not the case, primarily due to the viscosity of the melted filament. This viscosity results in a much more gradual drop in pressure inside the nozzle when the filament is retracted, and this gradual decrease results in some melted filament being pushed out of the nozzle during the retract function.
The net result of all this is that it takes some trial and error to determine what parameter settings work best for your particular printer and filament. Michael’s page gives some very helpful insight on how to understand what these settings are and what results you can expect. But, as is often said, there is no test like a real test.
Last Update: 31 May 2017