Extruder Pressure Dropping During Moves

I searched the forum but couldn't find this question.

My extruder seems to be loosing pressure when the head is moving but not extruding. Then when it starts extruding again I get very thin or skippy lines for a bit until pressure builds back up. After that, the plastic flows nicely. I've tried bumping up the heat, bumping up the flow rate, bumping down the feed rate but I can't seem to get it to stop. Do I need to bump things up or down some more? Is there some skeinforge setting I'm missing?

I have Techzone electronics and am running Skeinforge 31.

Thanks,
Brian

Hi

Are you using Skeinforge 5D code (Dimension activated)? If so you can alter the retract settings. Restart extra distance sets the extruder to deliver more plastic than it retracts.

[fabmetheus.crsndoo.com]

---

Regards

Ian

The restart distance should be exactly the same as the retraction otherwise you will have an excess of plastic at the start of a line. I.e. a blob.

If you are losing pressure do you mean the plastic is oozing during the move or is the motor slipping backwards, or are you retracting and not fast forwarding during the restart?

---

[www.hydraraptor.blogspot.com]

Perhaps it back-flows while it is not extruding meaning that you would need to restart further than you retracted, but your flow rate would be increased for the same stepper rate until the back-flow plug reached equilibrium again. Is your barrel diameter much bigger than your filament?

Edited 1 time(s). Last edit at 12/02/2010 06:44AM by Greg Frost.

Well if you continually restart further than you retract you must extrude too much material as it can't accumulate in the extruder forever.

That is why I simply fast forward the same distance I retract, about half a milimeter plus the gear backlash. I do this at about 10 times the extrusion rate and follow with about 30ms delay to give it chance to come out and bond to the bed.

---

[www.hydraraptor.blogspot.com]

The barrel is oozing during a move. Not a lot but some and it takes a little bit to flow again. It doesn't appear that the motor is turning at all during the move. I'll look into this restart setting. I wasn't aware of that one.

Just took at look at the Skeinforge settings and both Retract and Restart are set to 0.0

nophead, How do you set that delay so it has a chance to bond?

Does the retracting help with the oozing?

I see where there is a extruder retraction speed. Mine is set to 13.3. Should that be set higher?

Retraction completely fixes oozing. Once retracted the plastic will ooze very slowly due to gravity, but as long a the head moves are fast enough there is negligible ooze at the end of a long move, so there are no strings between my objects or blobs.

I don't do it via SF though, I use my own software, so the retraction, fast forward and delay is in my Python extruder class, not the g-code.

---

[www.hydraraptor.blogspot.com]

nophead: I think you get better results with retraction than most people. Are you using the hotend with the tapered barrel? That may be key. I find if I retract too much I get jams (heat being pulled up the barrel? perhaps my barrel is not smooth enough at the transition zone). If I retract as much as I can to still have a reliable extruder it still oozes a little (although it is much better than with no retraction).

Even when it doesnt ooze it still seems to take longer to start again, presumably because molten filament has been pulled up the barrel where previously there was air. The tapered nozzle may make this less of an issue.

Edited 3 time(s). Last edit at 12/02/2010 08:52AM by Greg Frost.

HydraRaptor has a tapered SS barrel, but my Mendel has a straight brass barrel with the transition zone in PTFE (constrained in a copper pipe). The same technique also works with Brian's hybrid, which has no taper.

I only retract about 0.75mm, so I don't have any problem restarting. It only needs to be far enough to release the pressure in the melt zone, not far enough to suck air in. I find if I retract too far it sucks the filament off the bed at the end of the run leaving the end sticking up. That can can cause the outline to lift.

It doesn't stop the plastic dead as it will still ooze a little due to gravity but that is a couple of orders of magnitude slower. I only notice it when the extruder is left stationary for a while. If the retract distance is too short then it oozes during long moves and the symptoms are blobs deposited on the outside of objects as the extruder moves over them. These come off easily with a fingernail, but a bit more retraction gets rid of them. As the gears wear the backlash increases and I need to retract a little further or move the motor closer.

I fast forward the same distance before moving off so the plastic should return to where it was. Since it do that quickly I find a small delay is required on Mendel to get the start to stick to the bed. I don't need one on HydraRaptor, but the melt zone is much smaller on that extruder. Presumably it is the time it takes for the plastic to flow from the nozzle to meet the bed. Too long and you obviously get a blob, too short and the end does not stick down.

---

[www.hydraraptor.blogspot.com]

You really are the master of simple comprehensive explanations nophead! Thanks.

Changing the restart setting seems to get my flow going better for a short move but I still have a lot of drool. Pulling it back by .5. Still on along move, like when it orbits, it doesn't start right away.

If you've got excessive drool, are you sure you have the right temperature, and haven't got things over-temp slightly?

If it's working like you describe, I'd suggest dropping the temp a degree or two downwards and printing again. If it's still doing the same thing then do it again till it improves. If it starts to jam and doesn't flow, then you are too low.

Basically my guess is that the plastic is just over-temp. I'm guessing that it's just enough over-temp to run faster (more liquid-like) than you want, leading to drool. Cutting the temp downwards (in small increments) until it improves should (in theory) be the right move.

Yes 10C can make a big difference with PLA. I have it hotter on the first layer to get it to stick to blue tape and then drop the temperature.

I have no idea why skienforge does this orbiting thing but it seems a really bad idea. Can you turn it off? You don't want any pauses once you have started building the object. If it gets too hot then slow down, make multiple objects or use a fan (you must insulate the extruder if you use a fan). If you leave a hot extruder full of PLA for more than a second or so then it will ooze significant amounts and to restart you need to run it a while and wipe off the excess with a brush.

---

[www.hydraraptor.blogspot.com]

I haven't been able to find where to turn the orbiting off. I can edit the Gcode though. I edit the Gcode anyway because the last move often puts the head into my object. I just move the head up above the last Z setting. Cool is turned off.

I'm using ABS but I would assume the same principle applies. The wire for my thermocouple came loose this weekend so I had to attach it again. I must have it in a much better place now because it seems to be running hotter at a given temp than before. I've reduced the temp by 5C a couple times but I think I still need to dial it in a bit. I think I'm getting close. I'm still printing and drooling even though my temp setting is at 215. I'm going to keep dropping it and see what happens.

On another note, the first layer of my vertex still has some tiny gaps in the infill. I've tried turning up the first layer flow rate but my second layer seems to have too much infill. Do I need to continue dropping over all flow and upping first layer flow or is there a better way?

Odd that you get much drool with ABS, as it is nowhere near as runny as PLA.

If the first layer is too sparse, but the second layer OK it usually means you are starting with the nozzle a bit too high.

---

[www.hydraraptor.blogspot.com]

Even with a raft? The nozzle being too high I mean?

No, with a raft it should have sorted itself out by the first layer. I can't however think why the second layer would have more plastic than the first.

It could just be a cumulative effect. If there is slight too much plastic then it gets worse each layer until it switches to sparse infill.

Edited 1 time(s). Last edit at 12/06/2010 06:58PM by nophead.

---

[www.hydraraptor.blogspot.com]

here is a picture that actually shows both my problems. You can see where on the first layer there are some tiny gaps in some of the infill and where on the second layer it appears to have too much plastic. There is a first layer multiplier so I may just need to work with that.

Also the object is not printing in the center of the raft. I've got that going in a different thread here in the forum.

I think the reason the first layer is a little sparse because the raft is sparse, so it can sag between the raft lines. You can see the pattern of the raft on the top of the first layer so that must be the case. Also the second layer doesn't look to have too much plastic on the photo. Does the nozzle end up ploughing through it?

I wouldn't worry about the small gaps in the first layer at this stage. If you get the amount of plastic right on the second layer it will probably be OK.

I haven't used rafts for a long time but when I did I used three layers and made the top layer lines almost touching. That gave a flat base to my objects but was difficult to remove. I also cooled the raft to room temp before starting the first layer and also did the first layer at a lower temperature.

The rest of the world seems to use much sparser rafts, which must be easier to remove, but the first layer will not be squashed as much as the other layers of the object, so the filament on the first layer will be more cylindrical and have less contact area horizontally.

Here is the work I did on rafts:
[hydraraptor.blogspot.com]
[hydraraptor.blogspot.com]

I have no idea why your raft is offset. Does it display like that in the g-code viewer?

---

[www.hydraraptor.blogspot.com]

Hi

I used to get orbiting but I cannot remember what setting I changed to stop it. I think it occurs if you have different temperature settings for different layers. I have all my temperatures at the same setting.

You can fix the head ploughing by using end.gcode in the Skeinforge Alterations directory.

Mine is:-

M104 S20
F50
G91
replace_with_g1 Z10.0
F3000
replace_with_g1 Y40.0 F1500
G90

You also have to have this line in replace.csv

replace_with_g G

NOTE it must be a TAB character between _g and G

This is to get round a problem where Skeinforge was dropping a character and is a fix Enrique implemented.

---

Regards

Ian

Wouldn't that always make the z axis go to 10 and y axis go to 40 or with it add 10 to the ending Z height, etc.?

Hi

The G91 sets incremental.

So the movements are from the last position generated by Skeinforge.

I get a 10mm move up from the top of the job and a Y axis move of 40mm from the corner of the job.

---

Regards

Ian