Why 600 steps on E axis in Firmware?

K8200 3D Printer K8200 - Printer software (repetier, slicer, ...) a
1

Hi guys,

After I while I started to get Z banding on my prints and having upgraded the Z system to eliminate mechanical issues I started to look at the programs. Looking at guides on the internet calibrating the extruder I noticed my printer extrudes 42mm of material when I ask for 50mm, showing the E steps are wrong. I updated my eprom to 705 steps/mm to get 50mm correctly.

But I then wasn’t sure what I should change in Slic3r to reset this. I started by altering the multiplier from 1.15 to 1 but my prints seemed worse, but I took the next step to calibrate wall thickness by printing a hollow cube 1 perimeter thick. Asking for 0.42mm would give me 0.51mm, adjusting the multiplier to 0.82 gave me a perfect sized cube with walls and overall sizes correct. However my layers would delaminate easily and running some common parts off resulted in awful prints, I was obviously not extruding enough material.

I was sure the multiplier of 0.82 should be left as close to 1 as possible and reset this, adjusting my perimeter % in Slic3r to try and get the same perfect sized cube. After playing for this for a few days I gave up and reverted back to the 600 steps in the eprom and default slic3r setting which work very well.

My question, has anyone else done this and got it to work?

I would still like to calibrate my printer close to perfection if I can.

Thanks
Andy

2

i’m interesting too by the answer, i would like to know why E step is at 600…

3

I run into the same issue. I set the E-Steps to 700 and with that I get the requested amount of filament extrusion. However, now my prints are nearly unusable, I get holes everywhere. I am still trying to track down the problem. I’ll let you know once I found something.
Can you post your slic3r config? Which version are you using?

~ Tectu

4

Tectu,

Please report back with any finding. I am now still back on 600 steps and using almost default setting in Slic3r, which work well.

My thoughts where the multiplier (1.10), filament diameter (3mm) and the steps (600) are connected. Increasing the steps to 705 means I could reduce the multiplier and go from there but like you this really made my solid layers wrong and I didn’t want to play too much with the advanced slic3r setting in extrusion widths as they are already at 170% and I would need to increase them more.

5

I was using Repetier host 0.90 and Slic3R 0.99?

Now Repetier 0.90 and Slice3r 1.0RC2

Default setting except I have altered the first layer speed to 50% to aid bed adhesion.

6

I am still doing testprints. I had to touch the advanced settings of slic3r. I am currently doing testprints with 200% extrusion widths… I don’t really like to go that far away from the default (as you said with the multiplier around 1 as well).
I’ll share everything I have as soon as I get usable results.

@Moorron: By ‘default settings’ you mean the velleman config from the download sections or what Slic3r 1.0.0RC1 gives you by default?

~ Tectu

7

Hi Tectu,

the extrusion width affects the amount of E steps generated over some length of printed line, but also the distance between this line and the next one. Therefore, changing it won’t result in more or less material in a certain layer, only the distribution of lines in this layer will be different.

When I counted the teeth correctly, the stock extruder has a 10:43 gear ratio. The hobbed bolt is M8, so the effective diameter (as the filament is pressed “into” the bolt till the bottom of the dents) is maybe 7.5 mm, the circumfence therefore 7.5 mm * pi = 23.6 mm. So, one full turn of the motor would result in (10/43) * 23.6 mm = 5.49 mm of filament transported. This would equal 3200 steps, so the theoretical steps per mm value would be 3200 / 5.49 = 583.
Apart from the guessed value for the bolt diameter (does anyone have a measured value for this?), this seems to indicate, that 600 would be a rather good value for the E steps per mm, already compensating for some percent of slip that might occur.

When you set your E steps to 700 (i.e., when your printer did not print enough material), did you check your filament for slip marks?

Cheers,
kuraasu

8

[quote=“kuraasu”]Hi Tectu,

When you set your E steps to 700 (i.e., when your printer did not print enough material), did you check your filament for slip marks?

Cheers,
kuraasu[/quote]
Hi kuraasu, I did have to set it to 730 to get the 100mm extrusion length right. i haven’t checked for slip marks but it didn’t behave suspicious and the extruder was 195 degrees. I will check for it next time.

Thanks,

bert

9

Some times ago I already posted similar subject, but no response …

[url]Extruder steps by mm Calibration]

10

This whole topic of why the default extrusion steps/mm is set to 600 is still a mystery to me.

It is very instructive that Velleman support is conveniently silent on this issue, and other issues that are inconvenient for them to discuss.

I changed the value in the configuration.h file and uploaded new firmware. Naturally, it screwed up all my prints because this is not compatible with the other Vellerman bogus parameters.

With an accurate (measured) extruder feed rate and filament diameter, one possible explanation is a grossly oversized hole in the nozzle.

HOW ABOUT IT VELLEMAN, isn’t it about time to throw some light on this topic???

disappointed,

Art

11

I think that I have an explanation for the 600 steps/mm that seems to work better than a correctly measured calibrated E Step value of about 700 …
The calibration process by measuring the real length of extruded material don’t take in consideration the material dilatation after fusion, I think that the volume of 50mm of filament at ambient temperature is not the same after the heating and cooling cycle, I think that the PLA loses less volume when cooling than the volume that he takes during heating , that explains why when we use a correct Estep value we need to use extrusion multiplier less than 1.0 to not have too much volume after extrusion.

It’s also why I prefer printers who use 1.75mm filament, because the fused volume is smaller and the extrusion is then far more accurate. The section of 1.75mm filament is 3 times smaller than the section of 3mm filament thus the volume of the same length of filament is 3 times smaller for 1.75mm that for 3mm filament and probably the error in volume is also 3 times less significative

12

OK, all you people who seem completely unable to get this to work, you are probably calibrating with the hot-end in place and extruding too cold or too fast.

Try it this way instead.

Set the filament diameter to the actual average diameter of your filament.

Set the feed multiplier to 1.

Create a test piece which is some kind of staircase with steps about 20mm square and a rise of a few mm per step. half a dozen steps should be plenty.

Set all extrusion widths to your nozzle diameter.

Slice at 95% rectilinear infill with a layer height of 0.2mm.

Print the part ignoring the first few layers because they’re sensitive to the initial height of the nozzle above the bed. Even so, if it’s grossly over-filling or under-filling, alter E steps (send an M92 Ennn command during the print) and re-start if necessary.

Once the layers have stabilised, observe the 95% infill. You should see slight gaps between the lines, reduce E steps by 0.5% every 2 layers until you can see the gaps. If the top infill has gaps then increase the E steps by a similar amount. You should be able to find a happy medium.

Once you have a good E-step calibration, the save it to EEPROM, code it into the firmware defaults, or determine the ratio to the default (600) steps and use that as your multiplier in Slic3er.

Do not at this point sit back and think that all is good. Do some temperature tests to get the cleanest surface finish and best layer adhesion.

You may now need to go back and recalibrate the extrusion rate!

Print speed, extrusion temperature and extrusion multiplier all interact.

13

Paul, I didn’t use exactly the procedure you suggest, but did something that I think is functionally equivalent. I ended up leaving my E steps at 600 after experimenting with changes to the firmware.

I ended up using parameters very close to the Velleman defaults even though I initially was very critical of the Velleman defaults. I do now try to leave the extrusion multiplier at a value of 1, and use a better value for filament width than the default 3 mm.

I’ve been getting good results except for the z-axis banding. Replacing the coupler and all the other suggested fixes has not eliminated the banding to my satisfaction. I am awaiting the shipment of my trapezoidal spindle, and will report back if this eliminates the banding. I will be disappointed if it does not because I cannot imagine that this will not fix the problem.

I believe the M8 threaded rods themselves do not have precision threads. Anyone who has a good threaded rod and is getting no banding is fortunate indeed.

14

hi all.
i also have had allot of issue with the filement extrusion.
and to make a long story short.
the main problem i noticed was that the PTFE tube is unreliable. the hole inside the tube around the 3.2mm got smaller at the threaded side. therfore the filement flow blocked and the fillement got stuck inside the tube.
so if i printed a object the quality of the object changed durring printing.
you can’t calibrate this.

after the fifth time the filement got stuck inside the tube.
i bought the E3D V6 extruder. (sorry velleman but i think your extruder $#(% )

the results and reliability are GREAT!!! the filement goes through the extruder realy NICE!!! (as the dutch say “kakken zonder douwen” translate “taking a crap without Deng”) :wink:
all my issues where gone.

15

Spirit18 is right, the PTFE barrel can cause some problems.

The manual just says “Screw on the white plastic barrel.” The problem is that the PTFE barrel is chamfered (cut off at an angle) on the end, presumably to seal against the bottom of the drilled hole on the brass barrel. If the PTFE barrel is tightened too much, it is squeezed smaller, making the hole too small for the filament. It even works as a non-return valve, allowing filament to be extruded but not withdrawn - I have noticed this a number of times when trying to pull broken filament out of the extruder with the hot end up to temperature.

The solution is to get a feel for how much to tighten the PTFE barrel.

Wally

16

I noticed that the Z axis banding was at the same pitch as the Z axis leadscrew, eg 1.25mm

I changed to twin trapizoidal leadscrews with 2 stepper motors and now have no banding at all.

The PTFE barrel problem i have had as well, it seemd to squash up when overtightened and also the pressure from the mounting bolts does not help

I added lock nuts to the mounting bolts to remove all pressure on the PTFE barrel

17

A single trapezoidal leadscrew eliminated the banding for me, but I never print with a layer thinner than 0.3mm.

18

I found that layer thickness was not a factor in the banding

19

Could someone that has eliminated z-axis banding please post a picture or two. I am very interested to see what is achievable before I invest in a new Z-axis leadscrew.

Many thanks
Peter

20

The bow below was printed with the (straightened) stock screw, a new motor holder and a flex coupler :