I know based on my current spool of filament but as I said that changes to a degree for spool to spool and vendor. Even rolls from the same batch can have different shrinkage factors. But just make your self a cube of known dimensions like mine which is 30 x 30 x 30 and then put a vertical hole of known dimension through it, (I use 15 just to keep things square) then measure the print and see how much it deviates. Then you know what to adjust by.

If you use the same brand, but maybe different colors you should even make a spread sheet for each color and it's shrinkage factor or compensation amount. There can be a variance in batch and the same vendor per batch but it is usually consistent enough not to be an issue. The variation comes when you try other vendors. So if you find a brand you like, stay with it to  assure better consistency in both print and shrinkage.

Great idea, thanks for the help and info!

Oh of course there is shrinkage and that is a downside to all thermal process manufacturing. It must be accounted for and the design should reflect changes in geometry due to thermal shrinkage/expansion. I know full well that that is clear and present in FDM. I don't mean to negate that aspect by "super calibrating" the printer (if it were even possible). I simply want to get the extruder where it needs to be in order to start determining rules of thumb for shrinkage (per part).

Long story short: I want to get the extruder settings at a good benchmark on the hardware/software available in order to start adjusting designs for ABS shrinkage.

It seems like you have an understanding of the DaVinci that I want to be at- where you know what dimensions you need the output part to be and adjust the input to compensate.

I know based on my current spool of filament but as I said that changes to a degree for spool to spool and vendor. Even rolls from the same batch can have different shrinkage factors. But just make your self a cube of known dimensions like mine which is 30 x 30 x 30 and then put a vertical hole of known dimension through it, (I use 15 just to keep things square) then measure the print and see how much it deviates. Then you know what to adjust by.

If you use the same brand, but maybe different colors you should even make a spread sheet for each color and it's shrinkage factor or compensation amount. There can be a variance in batch and the same vendor per batch but it is usually consistent enough not to be an issue. The variation comes when you try other vendors. So if you find a brand you like, stay with it to  assure better consistency in both print and shrinkage.

Long story short: I want to get the extruder settings at a good benchmark on the hardware/software available in order to start adjusting designs for ABS shrinkage.

It seems like you have an understanding of the DaVinci that I want to be at- where you know what dimensions you need the output part to be and adjust the input to compensate.

Ok I see what you mean. If I understand correctly, you are calibrating your single wall thickness against 2x whatever the layer height you set for the print was. So a correctly calibrated davinci extruder should output something in the ballpark of;

.1mm layer height = ~.2mm wall thickness
.2mm layer height = ~.4mm wall thickness
.3mm layer height = ~.6mm wall thickness
.4mm layer height = ~.8mm wall thickness

(on single wall extrusion)

I appreciate the information. I only am asking about esteps and minute calibration because I am working on printing parts that interface with machined bearings, nuts, and the like. I am really trying to determine the amount of Davinci clearance needed to produce parts as designed in Solidworks/Inventor.

I have been using FFF based printers for several years now and the best accuracy I have been able to achieve is down to .01. Yep just two decimals. Also you will never get the dimensions on paper to be printed unless you account for shrinkage which varies from batch to batch of filament. For example I am making a carriage for my Da Vinci so I can use the E3D hot end. I am going to use 8mm linear bearings which have an OD of 15mm. In order to get the bores for the bearing to fit snug but still allow the bearing to be put in I had to make the bores 15.25mm on paper to get 15.1mm on the printer. This is not an issue with the printers calibration, but an issue that once ABS cools it shrinks slightly. PLA does it as well  but not as bad. All fused filament materials will have some degree of shrinkage. So if you are making precise parts you will have to figure out what that shrinkage is for your current filament and then compensate for it in your design on paper or CAD.

What I do is print a 30mm cube with a 15mm vertical bore through it then measure the finished print to see what my shrinkage is like. I then adjust my design measurements to compensate.

So that being said, depending on your tolerance you may not be able to achieve it on an FFF based printer such as the Da Vinci. Even a high end printer like a Makerbot will be hard pressed to do much better simply because the limitation is not the machine but the material.

.1mm layer height = ~.2mm wall thickness
.2mm layer height = ~.4mm wall thickness
.3mm layer height = ~.6mm wall thickness
.4mm layer height = ~.8mm wall thickness

(on single wall extrusion)

I appreciate the information. I only am asking about esteps and minute calibration because I am working on printing parts that interface with machined bearings, nuts, and the like. I am really trying to determine the amount of Davinci clearance needed to produce parts as designed in Solidworks/Inventor.

carl_m1968 wrote:

rockosmodlife wrote:

I was running some single wall calibration tests to calibrate extrusion rate when I found that I had to set my extrusion multiplier to .62 in order to achieve the design wall thickness of .4mm. At an extrusion multiplier of ~1, my single wall thickness is somewhere in the range of .63mm. The problem is when I then try to print something with infill and "substance" - not just a wall, I get lots of gaps in the fill and an overall under extruded print.

Have you run any single wall calibration tests? If so, how close where the printed results to the design?

For single wall cal you dont print a wall of a known thickness. You just print a single wall with a known layer height. I use a .2 layer height. If my extruder is properly calibrated I should then get a single wall with a thickness of .4 to .45.

So the model you use to print the single wall has a wall thickness that forces the printer to do a single pass? I am a little confused as to what you are calibrating against? How do you know the printer has reached a "goal" thickness (or extrusion rate)? I am very interested in your process!

I am just quoting what I was told once. But the best and easiest way to calibrate esteps is very simple. Mark the filament a known distance like 20mm or so from the top of the extruder assembly where the  filament enters at. The heat the machine and tell it to feed a set distance like 10 or 20 mm and then remeasure how far the line moved. If you feed 20mm and originally measured 20mm then your line would be right at the entrance of the extruder. If it is further in or not actually there then your esteps are off.

As for the previous post I was just told once that of you print a single wall with a known layer height of .2 then the wall thickness should .4 or twice the thickness of your layer height. That being said if you want a single wall of a specific thickness then you need to set your layer height to accomplish that. Of course the limitation being you can or should never print a layer height that is greater than 40% less of your nozzle diameter. Basically you should never go higher than .26 on a .30 nozzle and so on. Otherwise you will not get the expected squish you are supposed to get.

rockosmodlife wrote:

I was running some single wall calibration tests to calibrate extrusion rate when I found that I had to set my extrusion multiplier to .62 in order to achieve the design wall thickness of .4mm. At an extrusion multiplier of ~1, my single wall thickness is somewhere in the range of .63mm. The problem is when I then try to print something with infill and "substance" - not just a wall, I get lots of gaps in the fill and an overall under extruded print.

Have you run any single wall calibration tests? If so, how close where the printed results to the design?

For single wall cal you dont print a wall of a known thickness. You just print a single wall with a known layer height. I use a .2 layer height. If my extruder is properly calibrated I should then get a single wall with a thickness of .4 to .45.

So the model you use to print the single wall has a wall thickness that forces the printer to do a single pass? I am a little confused as to what you are calibrating against? How do you know the printer has reached a "goal" thickness (or extrusion rate)? I am very interested in your process!

I was running some single wall calibration tests to calibrate extrusion rate when I found that I had to set my extrusion multiplier to .62 in order to achieve the design wall thickness of .4mm. At an extrusion multiplier of ~1, my single wall thickness is somewhere in the range of .63mm. The problem is when I then try to print something with infill and "substance" - not just a wall, I get lots of gaps in the fill and an overall under extruded print.

Have you run any single wall calibration tests? If so, how close where the printed results to the design?

For single wall cal you dont print a wall of a known thickness. You just print a single wall with a known layer height. I use a .2 layer height. If my extruder is properly calibrated I should then get a single wall with a thickness of .4 to .45.

Have you run any single wall calibration tests? If so, how close where the printed results to the design?

As for results I have not noticed a need to change them. Just speed and temp is more than enough to achieve this printers maximum ability.

It is possible you might be expecting too much from it.

Or those running XYZ open or Repetier- do you get good quality prints using the respective "stock" esteps?