Separate names with a comma.
Discussion in 'Getting Started' started by Spoon Unit, Aug 22, 2019.
I'm currently trying to talk KISSlicer into playing nicely with the toolchanger.
Turns out it's trying to do some needless gcode optimization by omitting superfluous M106 commands.
Or what it thinks is superfluous.
So if, e.g., the last command was M106 S128, and the next one is supposed to be also M106 S128, then it will omit that one.
Which completely messes up things on a toolchanger, because if the first M106 S128 was while T0 was active, and the second one was *supposed* to be during T1, that second one will never execute, and second tool will start with its fan off. Or in extreme cases the fan will never even turn on.
That's very far from KISS.
I'm finding that M106 Sn is setting all tool fans to the same speed. This is actually not a huge problem, but it's a bit noisier than it needs to be.
Did you apply the changes i've posted in this thread:
I thought I had, but will need to double check again.
Check the tool changing scripts too (post/pre) - by default they turn on the fan when you change tools.
Already changed that for sure.
I definitely do have the Tools set appropriately. I notice that you set the fans before the tools, whereas I do that the other way round, but what I see from the fan setup is that it doesn't really do anything that should affect the tool setup. If anything, each tool should be set a little more holistically than it is.
Also, I don't know if you have the same, but every time I try to change the Heater from active to standby using DWC, the tool gets collected. I think this could be because the heaters have the same name as the tools. Will test changing that at some point.
That's strange, it works fine for me - if i've got T0 active, M106 Sn will control fan on T0, if i've got T1 active, it controls M106 Sn etc.
In fact, from the documentation, the only way to get them all to work in unison would be to define tools like this:
M563 P0 S"T0" D0 H1 F2:4:6:8
But they are set up "holistically".
Look at the code above, for each tool, you attach the heater, the extruder, and the fan(s) in one go.
I do, yeah. That's expected behaviour AFAIK.
I don't think the names matter but i might be wrong.
M563 P0 S"T0" D0 H1 F2:4:6:8
I created macros to do that with the Gcode you recommended earlier M104 T0 S0
I didn't mean that way, I meant that whole single tool with fan setup could be pulled together in config.g instead of split up. It's just tidy up though, and not actually a functional change.
Well, i think the idea was that you set up the fans.
Then the heaters and thermistors.
Then the extruders.
And then you tie all 3 together into a tool.
The reason it's done that way is because 1 tool can technically have multiple devices, or, multiple tools can share 1 device.
E.g. you could have 1 fan shared between all tools, or have 1 heater but 2 tools, or have 2 fans on one tool etc.
That makes sense ... technically. Ultimately it's all just code to be understood and trimmed up.
Having some real trouble getting purge/brushing to work...
Haven't found a suitable silicone brush or wiper yet, so i'm using a paperclip and a very very soft brass brush for now.
What happens is that for some reason, the filament refuses to stick to the paperclip.
So when i purge and wipe across it, the paperclip just folds the string back onto the nozzle, then it drags the whole thing inside the brush.
The brushing action then detaches it, but it remains in the brush.
So after about a dozen layers, there's a tangle of filament inside the brush, that the tools pick up and drag into the print.
It's a right mess...
I've got no idea what the deal with the paperclip is, but the filament sticks to it LESS than onto the Nozzle X.
In fact, the nozzle is positively sticky compared to it.
Think i have a leftover pad of hard silicone that i've used for various vibration dampeners, gonna try and make some kind of a wiper out of it.
Maybe i should try lightly sanding the paperclip...
And also use two, one above the other. The one acts more like a wire-bending peg in a wire-bending CNC would.
Edit: another thought. The old original BigBox wiper and bucket worked perfectly. I think i still have it *somewhere* in some drawer. If i find it, i might adapt it to TC.
Sanding the paperclip doesn't do anything.
What is this thing made of?! Teflon?
Which brush were you using?
Very soft (thin strands) brass brush.
But the brush isn't the problem, it's the paperclip i've set up in front of it.
The idea was that i purge, wipe across the paperclip (straightened into a wire) to get the purged filament off the nozzle, then wipe across the brass brush quickly. The brass brush is doing its job, the problem is, the paperclip doesn't - when the nozzle goes across it, it doesn't get the strand of extrudate off it, rather, it just folds it over the nozzle.
So after a while (a dozen layers), i end up with a bush of filament on the brass nozzle.
Another issue i've noticed is the default print cooling fan duct design seems to produce a very thin and flat (good), but also very directional stream of air (bad). The reason it's a problem is you get very little cooling on overhangs that are on the opposite side of the duct, and increasing the fan speed just causes extrusion issues (filament freezing before it's deposited and adheres completely.
I'll need to figure out some kind of a bi-directional design, perhaps from the sides.
The Duetwifi is not exactly a wifi powerhouse. A tiny antenna buried within an enclosure and rf noise will trip it up easy. Also it has issues with routers that have 2 lans on different frequencies but with the same name. I use a repeater and keep it on the simplest wifi LAN.