How Do I Calibrate My Titan Extruder and Test Print?
-
Where is the config for retraction settings, and what's the basic science behind over/under/tuned retraction?
If you use "normal" retraction, then that is done in the slicer. If you use firmware retraction then the parameters are set using M207 in config.g but you must "tell" the slicer to use firmware retraction so that it will put the relevant G10 / G11 commands into the gcode file.
The basic science of retraction is that FDM printers are essentially hot melt glue guns - albeit accurately positioned. At the end of a print move, there is a blob of filament at the nozzle tip so when you do a non-print move, that blog will either be left behind or pulled into a string. Filament can also ooze of it's own volition due to pressure in the nozzle after a print move making that blob bigger. The idea of retracting filament is to pull that blob back up into the nozzle at the end of a print move and before a non-print move. Retraction should be set to the minimum that is required to alleviate leaving any blobs on the print and prevent stringing. Too much retraction can lead to under extrusion at the start of the next print move due to the amount of filament that needs to be "unretracted", excessive wear of the filament as it passes over the hobbed bolt in the extruder and in extreme circumstances, can lead to molten filament being drawn up inside the heat break where it will solidify and cause a blockage.
-
The basic science of retraction is that FDM printers are essentially hot melt glue guns - albeit accurately positioned. At the end of a print move, there is a blob of filament at the nozzle tip so when you do a non-print move, that blog will either be left behind or pulled into a string. Filament can also ooze of it's own volition due to pressure in the nozzle after a print move making that blob bigger. The idea of retracting filament is to pull that blob back up into the nozzle at the end of a print move and before a non-print move. Retraction should be set to the minimum that is required to alleviate leaving any blobs on the print and prevent stringing. Too much retraction can lead to under extrusion at the start of the next print move due to the amount of filament that needs to be "unretracted", excessive wear of the filament as it passes over the hobbed bolt in the extruder and in extreme circumstances, can lead to molten filament being drawn up inside the heat break where it will solidify and cause a blockage.
Well said, and well understood. Thanks for that breakdown. It makes more sense, particularly having the nozzle block up twice on a couple of small prints. There's a lot of blobbing and stringing when the nozzle heats up.
I used to have a nozzle fan on, which cooled any oozing filament prior to print, and blew it off and to the corner of the bed. Fancy. I should fix the retraction settings though. I'm quite happy to do it in the slicer at this stage, seeing that it's a consistently used process with print iterations.
-
Mhh, after testing and testing i'm now using the German RepRap X1000 PLA-Profile for SimpliFy3D which can be downloaded at Germam Reprap. This Profile is available for PLA / PETG and also for Dual-Nozzle and uses the 0.8mm Nozzle.
That Settings seems to fit to my machine almost perfectly.
-
…................ There's a lot of blobbing and stringing when the nozzle heats up..................
Yes that's normal too. When you first start heating the hot end, it's full of cold solid filament. When it gets heated, it expands and has to have somewhere to go so it oozes out of the nozzle tip. Some hot ends suffer more than others and my theory is that it is proportional to the volume of the melt chamber. My Diamond hot ends ooze a lot when first heated which I believe is down to the fact that they have 3 or 5 largish melt chambers. My start gcode moves the head to the very front of the bed so that I can get hold of the oozing filament with a pair of needle nosed pliers just prior to the first move. I'm in the process of doing a redesign so that the nozzle will be clear of the bed with a "bucket" under it and a strip of silicone rubber that will wipe the nozzle when it moves to the first print position.
Edit. Most people use the slicer to do the retraction. The only real advantage of using firmware retraction is if you have a mixing hot end and need to retract multiple filaments simultaneously. Depending on your slicer, firmware retraction may give you a bit more control like being able to unretract a different amount to the retract, and at a different speed.
-
Another advantage of firmware retraction is that you can experiment with different retraction settings during a single test print.
-
Another advantage of firmware retraction is that you can experiment with different retraction settings during a single test print.
Good point - and it's very useful too.
-
By default, it looks like Cura was retracting at 6.5mm, which appeared to be a lot when watching it work.
The filament is drying and blocking the nozzle, or the extruder isn't quite returning the filament to the previous position to continue extruding consistently, and the nozzle cooks the small amount of filament left.
I'm going to shorten the retraction to 1.5mm, and see how that goes. The E3D kit has shipped today, which should be here later in the week I guess, or next. I don't trust this clone.
-
The E3D Volcano kit has been installed this week and operates beautifully. The quality is so much better than the close, as expected when you pay $190 more for what appears very much the same thing.
It still ended up jamming, and in the exact same place. So weird. I spoke to the guys on the official forum, and someone suggested that it may happen if there's no PTFE liner in the heatsink, or if the heatsink fan isn't working properly. Turns out that my fan had stopped working, and I have no idea why, so I swapped it out with a cheapo and that one works. It's currently printing.
The cooling issue doesn't make sense to me at all, but that aside I have another problem, yet again, and I can't figure it out. The problem I have now is a shifting print; as in the printer shifts on the X and Y a little as it's going up, giving me a slightly sloped print.
I replaced the factory X and Y belts with reinforced belts, which have teeth anyway, but have been optimally tightened. The bed is also tight. It's supported by tightened spring and a glass top that's taped down and levelled manually. So there's zero permanent movement in the belt and bed.
The only other thing I can think of is stepper current inconsistency, but my steppers are well under the upper limits of the drivers. I haven't done any tuning in the area of steppers, etc, so maybe that's a thing I have to do, of which I have no understand at present.
-
I've fixed a working fan to the heatsink and had better luck with the printer working consistently.
The quality of the print is still shaky, showing waves in the walls along the Y direction. Just had a look at the extrusion calibration again, and found that it's only extruding 80% of a measured amount. I removed the hotend and did the test again and found the filament was moving through at 110% of the 100mm length.
Adjusting the extrusion setting, it comes to E390 at 101%. The issue is that putting the hotend back on will mean it will still fall short 20% of the extrusion, because while the extruder can correctly measure the distance, it can't force the same amount of filament through the hotend in that time.
I guess this means I have to adjust the speed of the extrusion, or something in the movement section. Has anyone been through this process before? I have a feeling it's not as simple as reducing the maximum speed by 20%, or is it?
-
Take a look at this thread https://www.duet3d.com/forum/thread.php?id=3693.
-
I then went and did a M307 H0, and took the results and updated the heater details in the config.g.
On my machine I thought H1 was the hot end and H0 the bed?
-
….............The cooling issue doesn't make sense to me at all,..................
I've been down this road, not with a volcano but with a 5 colour Diamond that uses 5 (modified) E3D heat sinks and also with the 3 colour Diamond which uses 3 E3D Lite heat sinks. In simple terms you need a hot zone where the filament melts and cold zone where it stays rigid (so that the extruder can push it into the hot zone). Necessarily there is a transitional zone between the two and this is where the heat break comes in. It is designed to keep that transition as short as possible so one side is hot and the other cold. If heat creeps up towards what should be the cold end, then you get partially melted filament which, when pressed by the incoming cold rigid filament, will get squashed out sideways as if it swells. This will then cause a blockage (partial or complete). You need to keep the cold end cold to prevent this heat creep and the associated "swelling" of the filament in the heat break region. Hence the need for cooling. In theory, a heat sink on it's own might dissipate enough heat by transferring it to the air around it but if the air is static, it will itself warm up and make that transfer less efficient. Hence the need for a fan to replace the warmed up air with fresh cold air.
I've seen a lot of posts from people with E3D hot ends who experience blockages and many more from other people who don't. It seems to me that people fit any old fan to the heat sink without any thought to the flow rate capability of that fan. I can't help wondering if those who have problems are just using a lower flow rate fan than these who dont't have problems. Or perhaps users who have problems are in a part of the world where the ambient temperature is higher. I don't have an E3D hot end to test but certainly the cure for the heat creep issues on the 5 colour Diamond (with 5 E3D heat sinks) was to use a bigger fan.
The other thing that can cause a blockage is too much retraction. What happens here is that molten filament gets pulled up into the the cold zone where it will then cool and solidify.
HTH
-
Just wanted to say thanks for a great write-up on the hot end issue. It's been this long since I've touched my printer because I had the whole Christmas break, back into work, then moved house, and frankly I just needed a break from the printer after calibrating for a whole year straight.
Every time I introduce a better part, it takes a bit of fiddling around and incurs a few new issues. Introducing the Duet was a big change, followed by a proper hot end. The funny thing about the clone was how well it worked in any condition, but it gave me very little know-how about the science you mentioned.
For now, I just need to go in for another calibration and see if I can get it working again. It's been heating the bed up to scorching temperatures on start up, with no input, which is a new issue. And I never fixed the issue with the printer sliding across the print early in the peace, then straightening up after 50 layers or so. There's no explanation how it can slip consistently at the beginning, and go straight up later. It's not a mechanical issue, so I'm stumped.
-
Error since updating to Firmware 1.21:
G28
Error: G0/G1: insufficient axes homed -
Have you already seen the release notes about adding "S2" to moves before an axis is homed?
-
Error since updating to Firmware 1.21:
G28
Error: G0/G1: insufficient axes homedYou're probably trying to move an axis before it's homed. Probably raising the Z axis.
From the 1.21 upgrade notes.
On Cartesian and CoreXY printers, normal G0 and G1 moves are no longer allowed before the corresponding axes have been homed. In particular, if your homex.g, homey.g and homeall.g files raise Z a little at the start and lower it at the end, you will need to add the S2 parameter to those G1 Z moves. Otherwise the G1 Z move will be refused unless Z has already been homed and the homing macro will be terminated.
-
Have you already seen the release notes about adding "S2" to moves before an axis is homed?
I did catch the notes a few days ago, and have tried implementing it, but I'm formally an animator, not a programmer, so I have no idea precisely which file and what lines of code should include 'S2'.
I threw it in here (under Lift Z):
; homeall.g
; called to home all axes
;
; generated by RepRapFirmware Configuration Tool on Sat Mar 04 2017 15:46:00 GMT+0100 (W. Europe Standard Time); Relative positioning
G91; Lift Z
S2 G1 Z5 F6000; Course home X and Y
G1 X-235 Y-215 F4000 S1; Move away from the endstops
G1 X5 Y5 F5000; Fine home X and Y
G1 X-235 Y-215 F250 S1; Absolute positioning
G90; Go to first bed probe point and home Z
T0 ; select first hot end
G91 ; Relative coordinates
G1 Z3 F200 ; Lower bed to avoid hotend dragging in bed.
G90 ; Absoulte coordinates
G1 X170 Y125 F3000 ; Go to first bed probe point and home the Z axis
M401 ; Deploy probe
G30 ; Probe
M402 ; Retract probe
G1 Z10 ; Move nozzle up 10mmIt homes X and Y, then throws this error:
G28
Error: Bad command: S2 G1 Z5 F6000
Error: G0/G1: insufficient axes homed -
Additionally, since 1.21, it's heating my bed up on startup, over a 110degs-c, AND won't start the WiFi module, so I need to run M552 S1 in Pronterface with a USB connection every single restart.
1. WiFi module won't start when switched on.
2. Bed heats up with full power.
3. Can't Home All axis. -
Move the S2 to the end of the line:
G1 Z5 F6000 S2
…rather than:
S2 G1 Z5 F6000
-
G1 S2 Z5 F6000 will also work. The important thing is that the actual command (G1 in this case) comes first.
