Optimizing Filament Extrusion Multiplier, Print Speed, and Extrusion Temperature

All,

I recently got into 3d printing and purchased a TAZ 6. I have been doing a lot of research on how to determine the optimal Filament Extrusion Multiplier, Print Head Speed, and Extrusion Temperature for printing a particular filament. I am trying to find an efficient way to do this (so that I can spend the least amount of time and money generating this information for every new filament I use) and was wondering if I could get some input from you all.

I would like to base my discussion today on the following test results compiled by 3D MATTER, a startup located at New Lab in Brooklyn, NY. Please keep in mind that for their “Influence of Extrusion Temperature on Mechanical Performance” test, they did not test the layer bond strength which would be affected by Extrusion Temperature.

It seems to me that the first step is to dial in the Extrusion Multiplier so that there is no under extrusion and no over extrusion. The way to do this would be to set correctly (in hardware) the idler tension and (in software) the filament diameter and nozzle width then print (“slowly” so that the extruder gear will not skip and the filament will not slip and at an “acceptable” temperature so the filament can be properly extruded) stairs while varying Extrusion Multiplier for each stair level. The Extrusion Multiplier that provides the best stair face (that does not exhibit under extrusion or over extrusion) would be selected. This would be all that is needed. Am I correct ?

Next, I believe that Print Speed optimization should be performed. This would be done by printing (at an “acceptable” temperature so the filament can be properly extruded) “suitable” complex test objects at different Print Speeds and selecting the Print Speed that provides the most visually pleasing test object. This Print Speed would be the optimal Outer and Top Shell Print Speed. The optimal Bottom Shell Print Speed would be one as close as possible to the optimal Outer and Top Shell Print Speed but slow enough to not allow the print to detach from the print bed. Now how about optimal Inner Shell Print Speed and Infill Print Speed ? It would seem that these should be printed at the maximum speed that the print head can move UNLESS line bond strength lowers as speed increases and in that case, they should be printed at the maximum speed that provides acceptable line bond strength. Does print speed affect line bond strength ?

Lastly, Extrusion Temperature optimization should be performed. It would seem that the obvious thing to do to optimize Extrusion Temperature would be to print something (perhaps a single wall hollow cuboid with no top/bottom while varying temperature through the layers) and select an Extrusion Temperature based on the its print quality and layer bond strength. However this would only give you the optimal Extrusion Temperature at a given Extrusion Rate. For example, if through this method, I determined that the optimal Extrusion Temperature is 230C at w=0.5mm, h=0.2mm, optimal outer shell speed=20mm/s, then I only have the optimal Extrusion Temperature for an Extrusion Rate of 0.5x0.2x20=2mm^3/s. This means that I would need to do this optimization for every nozzle width and every layer height that I would like to print at creating a 2d map of optimized Extrusion Temperatures. Am I correct ?

Does this process seem like an efficient way to determine the optimal Filament Extrusion Multiplier, Print Speed, and Extrusion Temperature or have you all a even more efficient method ?

I really appreciate your time.
David

Thanks for the link, a very good article and it fits to my own findings :slight_smile: But as you point out, it looks as they had tested the strength in line direction which is very sad. It would be very intresting to see measured data from inter layer bonding due to temperature…

Back to your questions, yes you are right in 99%. Here is the 1%:

Now how about optimal Inner Shell Print Speed and Infill Print Speed ? It would seem that these should be printed at the maximum speed that the print head can move UNLESS line bond strength lowers as speed increases and in that case, they should be printed at the maximum speed that provides acceptable line bond strength.

The maximum speed of the head isn’t important, it will be nearly always much higher than other limiting factors. The most important here is the ability of your hot end to melt the filament fast enough for your given flow rate (line width * line height* print speed. 0.5mm0.2mm50mm/s = 5mm³/s).
I did the test for PLA at 205°C and found a save limit at 9mm³/s, maybe 10 if you feel “angry”. If you go want to go higher, you would have to increase print temperature.
Second thing is the layer bonding, which is affected by print speed (I guess, I have no device to measure that in figures :frowning: ) and in a high magnitude by print temperature. The “bad” thing about the second one is, that there is no limit. If you print too cold, your print will fall into single layers as soon as you tough it. If you print so hot that you print not even holds it’s shape, you have the best “layer” (ok, you have no layers than any more :laughing: ) bonding. You have to find something between best looking and best bonding…

Finding all that numbers - especialy numbers that are true for all possible print object shapes - is a huge time consuming work. That’s why first step of all should be finding one filament supplier and stick with that. Or you will waste a lot of filament and time in calibration work and not printing useful things :wink:

Sebastian,

I really appreciate your reply. Things are starting to come together. I guess the main problem is that there are so many degrees of freedom and you gave one example about the temperature vs layer bond strength vs look. I will do some more optimization but like you said, it will take a lot of time and filament because there are so many degrees of freedom.

I will see if I am able to get access to a Universal Testing Machine and do some layer bonding ultimate tensile strength tests versus extruder temperature for some different materials. That way we will be able to see just how much temperature affects layer bonding strength and also compare them throughout the different materials.

I plan on hosting a website and writing out my findings. My goal is to compile as much data as I can so that 3d printing will be less confusing for beginners because I had a hard time understanding everything when I first started. I am glad though that you confirmed that my understanding isn’t too far off.

-David

Would be awesome if you could do the layer bonding test sometime :sunglasses:
Its true, there are some degrees of freedom.another one would be printing of overhangs. The maximum printable angle depends on the layer height, you can go higher when layer height is lower. But if you go to low, a new problem named curling edges will appear. This happens on sharp corners due to internal stresses of the filament line in the corner and the only way to fight them is to print incredible slow. It’s like removing internal stresses in steel,where you can also heat it up and cool it down very slowly. I found the sweet spot for pla at 0.15mm,where curling starts at the same angle where the overhang will start to fail.

But to be honest, for normal average parts its not absolutely necessary to have everything tuned to perfection, it will be ok if you are in the right range :wink:
But it helps a lot to understand all the limiting factors to solve a problem if you see one in a specific print.

Typos may be in this text because this is written on my mobile phone.