[Solved] Surface Quality Issue with E3D Cyclops

[626Pilot]

Neither of them seems to have any slop. If I rotate either one of them, I can feel the transmission turning. If backlash causes a problem, it's probably hidden by your slicer doing loops from the inside to the outside.

[bot]

Yeah, it's annoying knowing there is backlash... I'll see how much it affects anything after some more extensive testing. Thanks for the input.

[626Pilot]

Printing the dragon thing has failed over and over again. At some point, KISSlicer commands the printer to print with the wrong filament and shut off the heater. It's not at the first tool change, either. It gets through a bunch of tool changes, and then suddenly it just screws up like this.

It makes no sense to me.

[gestalt73]

Hey 626Pilot,

My guess is that KISSlicer is generating gcode to take the inactive extruder+hotend to standby temp, and after about 4 layers it doesn't have enough heat left to extrude after it keeps cycling between standby and print temps?

With the Cyclops, all that gcode stuff needs to be commented out of the appropriate tabs in KISSlicer.

If you're using v1.5 beta 2.20, look on the Printer G-Code tab, and comment out/remove everything in the following tabs:
- Select New Ext & Warm
- Warm Same Ext
- Cool Same Ext
- Cool and Reture Old Ext

Probably kinda hacky, but it will definitely work.

Don't touch Prefix (I think) you want your gcode to verify print temps before starting, and make sure you either turn off your hotend or take it to standby in the Postfix tab.

[626Pilot]

Thanks for the advice. I commented all that stuff out (except T<EXT+n>) but it's still doing exactly the same thing.

[JFettig]

Printing the dragon thing has failed over and over again. At some point, KISSlicer commands the printer to print with the wrong filament and shut off the heater. It's not at the first tool change, either. It gets through a bunch of tool changes, and then suddenly it just screws up like this.

It makes no sense to me.

You aren't using repetier server are you? I had the same problem then gave up on it

[626Pilot]

No, I'm running straight off the SD card.

[Polygonhell]

Printing the dragon thing has failed over and over again. At some point, KISSlicer commands the printer to print with the wrong filament and shut off the heater. It's not at the first tool change, either. It gets through a bunch of tool changes, and then suddenly it just screws up like this.

It makes no sense to me.

I think you'll find the issue is with repetier firmware, not KissSlicer.
I saw exactly that when I had a none zero offset in the firmware settings for the nozzle, it would just occasional ignore the swap command. I moved the offsets into the Slicer settings instead and the issue went away. There was a supposed fix in later firmware, but I tried pulling that version and had the same issues.

[626Pilot]

Printing the dragon thing has failed over and over again. At some point, KISSlicer commands the printer to print with the wrong filament and shut off the heater. It's not at the first tool change, either. It gets through a bunch of tool changes, and then suddenly it just screws up like this.

It makes no sense to me.

I think you'll find the issue is with repetier firmware, not KissSlicer.
I saw exactly that when I had a none zero offset in the firmware settings for the nozzle, it would just occasional ignore the swap command. I moved the offsets into the Slicer settings instead and the issue went away. There was a supposed fix in later firmware, but I tried pulling that version and had the same issues.

I'm using Smoothie, not Repetier.

I tried printing directly from Rep Host so that I could see which layer it was on when it failed, but RH decided to screw up the g-code and it stopped tool-changing entirely. (Which I find crazy, why would it mess with the G-code instead of just sending it?)

[Polygonhell]

Hmm... Maybe it was a repetier host issue then.
Makes me wonder what it's doing, all it's supposed to do is strip comments and add a line number and checksum.

[626Pilot]

It's printing now and seems to have made it higher than it ever has before. This is what I put into Select New Ext & Warm:

Code: Select all

; Select new extruder
T<EXT+n>

; Warm it up (don't wait)
M104 S<TEMP>

The other extruder change tabs are all commented out.

In theory, it could be that something in Smoothie is getting confused when the toolchange (T0/T1) code is sent that causes it to forget what temperature it's supposed to be at. I suppose it could also be that I pasted something in there that confused the parser (maybe it had \r at the end of the line and Smoothie expects to see \n or \r\n). "Forgetting" the temperature and no longer selecting the right tool sounds a lot like a parser error to me.

[626Pilot]

First two prints. For material settings on the 2nd one, I have temp=220C, prime=3, suck=3, wipe=5, and some G-code to help reduce color bleeding:

Select New Ext & Warm

Code: Select all

; Select new extruder
T<EXT+n>

; Warm it up (don't wait)
M104 S<TEMP>

; Prime filament
G92 E0
G1 E1.5 F200
G92 E0

Cool & Retire Old ext

Code: Select all

; Retract filament
G92 E0
G1 E-1.5 F1000
G92 E0

I tried retracting 3mm, but that caused jamming, so I backed it off to 1.5.

To prime the extruder on color changes, I used "Wall (all layers)". The short wall option didn't work as well - lead to the small layers on top getting too gooey.

There is more stringing than I saw before, probably because my initial test prints were with PLA and HIPS. I guess HIPS doesn't have the same thermal expansion? Maybe? The 1st model was printed with a wipe setting of 10mm, the second with 5mm. I think that helped a lot.

The weird thing is that even though I'm having it do a full-length wall, there's still substantial color bleeding. It makes the output model look kind of weird. Is there anything I can do to reduce that effect?