UM2 doesn't remember certain setting

[simon 1]

Just a quick question. Every time I turn on my UM2 do I need to use advance maintenance settings to reduce the acceleration from 5000 to 3000, if 3000 is what I'd like to use all the time. Material setting can be saved but it seems the movement settings can't. Or I am missing something?

 

[Daid 306]

Can you try changing the acceleration, opening the LED settings and closing that. After that reboot the machine to see if it has sticked. If that is the case then there is missing a "save settings" somewhere (I know the LED settings force all the settings to be saved)

 

[simon 1]

Thanks Daid, I'll try this after I finish my current print. While you're on the line, can you tell me what the other parameters in that menu are controlling i.e. x,y,z and Jerk? I thought that the head speed was controlled by Cura.

 

[Daid 306]

Cura controls the requested speed. What is happening is that the firmware needs to accelerate and slow down depending on those speeds. And it needs to slow down for cornering for example.

The Max speed limits the speed always to this value. So even if you ask for 1000mm/s the firmware can limit this to for example 300mm/s. Especially for the Z this is important, as Cura requests that the platform is moved at the travel speed (150mm/s) but the actual platform cannot move much faster then 40-50mm/s.

Jerk is a sudden acceleration change, when cornering, the printer does not come to a complete stop, but goes to jerk/2 speed and then leaves the corner at jerk/2 speed, so it has a sudden change of speed. Increasing the jerk will print faster, but a too-high jerk can shake your prints apart or even caused missed steps.

 

[simon 1]

Thanks for the explanation

 

[gr5 2,265]

If you are printing a circle with 10 segments, jerk allows the printer to not slow down much at each vertex so it can print the circle fast. If you are printing a square it slows down more at the corners (vertexes). This is controlled by "jerk" which is a Marlin specific term and not the same as what a physicist would call "jerk".

 

[simon 1]

Ha ha, got it, thanks. Perhaps there's something else you could answer for me. I want to reduced the speed of the top layer, or at least the first 2 layers of the top layer. Its the bit specifically where it's bridging over the open box infill pattern directly underneath. Is that part of the infill speed or the print speed?

 

[illuminarti 18]

I think there may be a 'save changes' missing on the retraction settings too - seems like I ended up having to change that several times, and then suddenly it stuck one time.

If you change things in the 'tune' menu during a print, are those only changed for the duration of the print, or do they persist too?

 

[gr5 2,265]

Is that part of the infill speed or the print speed?

 

print speed. Infill is this cross pattern with mostly open space.

 

[illuminarti 18]

Actually, no, I just sliced a cube to check.

The solid infill on the top layers of the cube is printed at the 'infill' speed. The mesh part in the middle is 'sparse' infill. The solid layers on the top and bottom are 'solid' infill. But both types are controlled by the infill speed.

Actually what the bottom does is more complicated. On the first layer, the loops around the outside are printed at the 'first layer speed'. The solid infill on the first layer is printed at the first layer speed multiplied by the ratio of infill speed to normal print speed.

For instance, if first layer speed is 10, normal print speed is 100 and infill speed is 200...

The the perimeter at the bottom is printed at 10 mm/s. The solid infill on the bottom is printed at 10 x (200/100) = 20mm/s. The perimeter and infill speeds on the bottom layer then look to be increased in equal steps over layers 2, 3 and 4, so that by layer 5 the perimeter and infill are printing at the full speeds set in Cura.

(All the above assuming that minimum layer time isn't a factor in any of those layers - if it is, then the speeds will be reduced even more to slow the print time for each layer down to the minimum layer time.)

 

[gr5 2,265]

So on a cube the very top layer is printed at infill speed? That's messed up! Isn't the whole point to print faster on the "hidden" parts?

 

[illuminarti 18]

So on a cube the very top layer is printed at infill speed? That's messed up! Isn't the whole point to print faster on the "hidden" parts?

 

'Yes', 'Possibly', and 'I can see why you'd think that'. In that order.

 

[illuminarti 18]

I think that you could also argue that the bottom layer speed should be treated as a maximum and/or constant speed for all parts of the layer, rather than printing the infill faster than that speed, if the infill is normally printed faster than the print speed.

 

[simon 1]

I'm not sure the UM2 print head gets quite hot enough to make the process of printing in ABS as easy as it could be. If it went up to 265 C it would be better, and this is why (IMHO). With the Afinia, the print head does go up to 265 and it creates fine strings in ABS better than the UM2. When you're bridging over the sparse infill with the UM2, the plastic isn't quite hot enough to allow the you to print over 30mm/s ish. If you increase it to 45mm/s the strand breaks before it reaches the next plastic rib of the sparse infill. This short broken strand then sticks up above current layer, possibility because until it broke it was under tension, and when it breaks, it releaves the stress by sticking up. It takes a long time and many layers of solid infill to seal this area over and the end result looks like a quilted blanket.

I think that because PLA can be heated well above its melting point and perhaps because of its visco dynamic properties, it creates a really fine string easily, which makes bridging a breeze. The UM2 is max'd out at 260 so the plastic isn't quite free flowing enough to produce this high speed, fine strand.

Printing the sparse infill is easier at higher speeds because its always pushing down onto another piece of plastic, but the first layers of solid infill need to be printed more slowly because the plastic isn't quite hot enough. Therefore if we can't increase the print head temperature by another 5 degrees, it would be beneficial to independently control the temperature of sparse and the first 2 layers of solid infill. Then you reduce the first 2 layers of solid infill to say 30mm/s and the others could be nearer 40 or 50mm/s.

 

[gadgetfreak 7]

We have FW 13.12 and when we change the retraction settings to 5.5mm and 35mm/s, they do not get saved on power cycle.

We also tried changing the LED settings and the LED setting we're saved but the retraction setting cannot be saved no matter what we try...?

 

[gr5 2,265]

Yeah - you can do it - but it's confusing. Try selecting PLA, then saving, then selecting PLA again? It is very confusing how you save material settings. It took me many tries to get it to work but you can do it if you just push the right menu values in the right order. I think this should be made more clear.

For example maybe it should say at some point "select material to save settings to:" instead of "select material".

 

[illuminarti 18]

Yeah, I currently change the retraction settings every time in tune. I did getit to save at one point, but it seems to have forgotten it again.

Now, I can't get the LED settings to save correctly, lol.

Its weird... it seems that when I power it on, the printer comes on at whatever it was set to, two changes ago. And then when I go into the LED menu, it immediately changes to the previous setting, without me doing anything. But then when I go out again, it goes back to the previous setting.

Except then later when I power on, it goes to a different setting. Sometimes.

Maybe.

I don't know. I just know I don't seem to be able to get the damn lights to stay at the brightness that I want them at; they're always at the brightness I wanted a day or two ago... :-)