Layer changes

I seem to be still getting some ooze during layer changes. I already use retract and have it set to
Retraction distance 3
Restart extra distance 0

when I tried a negative number for the restart I ended up with gaps in some places so wonder if there is a setting to make the layer change faster?

Thanks,

I know more stupid questions but I'm learning

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

It's odd that a lot of people report this problem. How long does a layer change take? On my Mendel IIRC my z- axis moves at 5mm/s with acceleration over 256 steps. Moving say 0.4mm takes no time at all compared to an XY move from one corner of the bed to the opposite corner, so oozing on a z-lift isn't an issue.

How much ooze depends on how fast you retract as well as how far. I think it may also depend on how dry the PLA is.

---

[www.hydraraptor.blogspot.com]

Are you using a bowden cable?

I guess Nelson is using a Bowden cable. I have the same issue, but it's either strings or blobs with a Bowden, it seems.
I think retraction moves are generated independently in Skeinforge, so always happen without movement, which could be part of the issue - though maybe this is due to being able to specify the retraction speed, which needs to be set differently to any other feed rate? Another issue is that it does it on all hot end movements, even if it's only a move of a couple of mm. I saw someone wrote a python script to remove these, but it would be better if there was a little setting in SK - "Don't retract if move is less than..." kind of thing.
I'm using the 'lower left' setting for choosing where the next layer starts on layer changes, and I tend to get a ridge on the outside of the part in that position. But it's usually pretty neat, and easy to cut off if it's in the way.

I was never able to get rid of the blobs when I was using a bowden at 40+mm/s. There is just too much elasticity in the filament, and not enough stepper speed to forward/reverse the feed fast enough.

Until somebody gets a nice mathematical model to accurately predict the nozzle response to the stepper, I think It will just have to be something you are forced to live with.

Or you can always switch to the predictable and accurate direct drive extruder.

I'm actually running an Adrian's extruder on a Mendel.

I have extruder retraction speed set to 75mm/s. I might try and set it a bit higher for my next round of prints.

I have speed and flow set in speed to 30 each and I'm thinking I might lower these now too

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

If you're on SF 40 or later, then 75 mm/s of feed is way too fast! I have it set to 15 mm/s. The electronics can't even pulse the steppers that fast.

Slowing down the feed rate from 30 mm/s won't really affect the ooze or blob between layer changes. Perhaps your nozzle is too hot.

Maybe you're running the heater too hot, making the filament too liquid? I've reduced the temperature on mine a bit, from a reported 205 degrees (for PLA) down to 190. Unscientifically, it seems to ooze slightly less, and pushes through the extruder with about the same pressure. I wish I had one of those laser thermometers to know exactly what the temperature is. I'm building a Prusa, so should be going direct drive in time.

Well I'm running at 175C which is flowing nicely

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

If you mean the IR thermometers with laser guide, they're not good for measuring the nozzle temperature. The sensitive region is too big. The better way is to insert a thermocouple into the nozzle where the feed goes.

If you have a Huxley, why not build Adrian's extruder for 1.75 mm filaments. I thought that's small enough to fit on the carriage.

Try retracting only .6mm at 20mm/s. I get small blobs if the retract distance is too high.

Since you had it set to 75mm/s my guess is you were skipping steps, too.

Hmm not sure, I originally set it to retract at 50mm then was told on the Mendel forums that might be too slow.

So can anybody confirm or not whether a mendel with Adrian extruder running Gen6 can retract at those speeds.

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

It depends on what version of Skeinforge you're using. For version 40 and later, that's too fast since the speed is based on the feed. If it's 39 and earlier, then it's too slow since the speed is based on the extruded filament. For an Adrian extruder 15 mm/s is a good value for version 40 and later and 540 mm/s (~36x) for earlier versions.

Ah yep that makes sense I'm using SF41, will give it a go and see what happens

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

I think the Gen6 board is speed limited by the firmware. I have one, and played around with speeds in repsnapper, but it topped out - sorry, can't remember off-hand at what speed. Higher speeds didn't seem to make it skip steps, just go at max speed. As retractions are carried out as single commands, is it likely to skip steps?
On mendel-parts, Camiel is experimenting with printing at 250mm/s with Gen6, so there should be some legs in it!

brnrd Wrote:
-------------------------------------------------------
> If you mean the IR thermometers with laser guide, they're not good for measuring the nozzle temperature. The sensitive region is too big. The better way is to insert a thermocouple into the nozzle where the feed goes.

Thanks, I didn't realise they weren't very accurate. Obviously I've got a thermistor in my heater block, but I'm considering going the thermocouple route for subsequent hot ends.

> If you have a Huxley, why not build Adrian's extruder for 1.75 mm filaments. I thought that's small enough to fit on the carriage.

My repstrap is a MrKim's 1X2, which is loosely based on the Huxley I suppose. However, it has the X axis smooth rods stacked vertically rather than horizontal, making is awkward to fit an extruder on top of the hot end without severely limiting the Z height. But it prints well enough for a repstrap. I'll go direct drive on the Prusa I'm building.

Andrew Diehl Wrote:
-------------------------------------------------------
> Try retracting only .6mm at 20mm/s. I get small
> blobs if the retract distance is too high.
>
> Since you had it set to 75mm/s my guess is you
> were skipping steps, too.


Wow that is much better thanks, I didn't know that if it was too fast it would not work properly. The blobs are just about gone although I am getting a bit of stringing across big openings now so might up it a wee bit from 0.6 when I do the next prints.

Now all I have to work out is why my carriage stopped halfway across the last layer of a print, kept on extruding leaving a very large blob then dragging it all the way over once it thought the print was finished. I had to turn the power off and back on to get it working again which it has been all afternoon, maybe I looked at it funny or it didn't like my tone of voice

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

NelsonRap Wrote:
-------------------------------------------------------
> Andrew Diehl Wrote:
> --------------------------------------------------
> -----
> > Try retracting only .6mm at 20mm/s. I get small
> > blobs if the retract distance is too high.
> >
> > Since you had it set to 75mm/s my guess is you
> > were skipping steps, too.
>
>
> Wow that is much better thanks, I didn't know that
> if it was too fast it would not work properly. The
> blobs are just about gone although I am getting a
> bit of stringing across big openings now so might
> up it a wee bit from 0.6 when I do the next
> prints.
>
> Now all I have to work out is why my carriage
> stopped halfway across the last layer of a print,
> kept on extruding leaving a very large blob then
> dragging it all the way over once it thought the
> print was finished. I had to turn the power off
> and back on to get it working again which it has
> been all afternoon, maybe I looked at it funny or
> it didn't like my tone of voice

Awesome. The other thing to check is if there is any backlash in the extruder gears. Reducing that to practically zero also helps a lot.

Is that what the lash tab is for? I haven't played with that before

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

The lash tab is for x/y backlash. I don't think there is any way to compensate for extruder backlash. Best try to eliminate it mechanically with good gear/gear spacing.