1 Topic by w_smith

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Topic: First 4(?)mm of object doesn't print right?

So I'm printing final(?) Lawsy Carriage parts one by one, and on the first one (Left side) I'm getting what looks like shrinkage of the first few layers, up to about 4mm height:

http://geekho.com/Push/20150322_212132607.jpg

It's almost like the part wants to peel up, but since I'm printing on glass with hairspray and finally got the bed temperature up to 100C and print with a 1mm brim, they can't peel up, but the stresses that would cause them to peel up are still there.

This is with Octave filament and feed and flow calibration, printing with 100% infill.

Does this look like anything obvious to anyone?

Thanks!

2 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

What filament temp and hotend are you using?
Do you have your printer enclosed?
That is curling off the bed so your adhesion isn't as good as it could be.
I would argue that your bed temps should be higher also.
I have printed literally hundreds of those and even with 12 stacked on the plate never have I had curling.  I run my bed at 108 using octave filament at 245 on an e3d and a full enclosure.

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

3 Reply by w_smith (edited by w_smith 2015-03-23 14:18:03)

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Re: First 4(?)mm of object doesn't print right?

wardjr wrote:

What filament temp and hotend are you using?
Do you have your printer enclosed?
That is curling off the bed so your adhesion isn't as good as it could be.
I would argue that your bed temps should be higher also.
I have printed literally hundreds of those and even with 12 stacked on the plate never have I had curling.  I run my bed at 108 using octave filament at 245 on an e3d and a full enclosure.

Stock hot-end (trying to print the parts to convert to E3D)
Hot-end Temperature 215C (I set it to 210 but RH bumps up to 215 at the beginning of the print)
SD4, so it's enclosed.
I've got the bed temp set to 125 so it'll actually reach 100 (my glass is pretty thick).  It won't go any higher, I can set it to 150 but it never gets much past 125, and the bed never gets much past 100.

Using Octave filament, just finished the extruder and flow calibration.

It looks like it peeled up, but it didn't, it's just showing the stresses of trying to peel up.

4 Reply by jagowilson

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: First 4(?)mm of object doesn't print right?

Try a thinner sheet of glass, it can't hurt. I use cheap $2 glass at the hardware store. Your bed should reach 105. Maybe check the plugs for a good connection and take up the aluminum to make sure there isn't an air pocket around the bed thermistor.

5 Reply by w_smith

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Re: First 4(?)mm of object doesn't print right?

jagowilson wrote:

Try a thinner sheet of glass, it can't hurt. I use cheap $2 glass at the hardware store. Your bed should reach 105. Maybe check the plugs for a good connection and take up the aluminum to make sure there isn't an air pocket around the bed thermistor.

I've got some thinner glass on order, so that might help.

But that kind of distortion doesn't immediately say <something> specific?  Everyone's thinking bed temperature?

6 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

That is a very close up image so it is difficult to judge things.  You may be slightly over extruding but other than that and a few blobs there is no smoking gun here.  That area of the print will not affect fit or performance so it is usable as is.
That doesn't make it correct but it sounds like you're doing everything right.  It could just be that particular roll of filament.
And as Jago said, go get a cheap thin piece of hardware store glass cut to size for a few bucks.
Are you using .2963 for your layer height?

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

7 Reply by w_smith

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Re: First 4(?)mm of object doesn't print right?

wardjr wrote:

That is a very close up image so it is difficult to judge things.  You may be slightly over extruding but other than that and a few blobs there is no smoking gun here.  That area of the print will not affect fit or performance so it is usable as is.
That doesn't make it correct but it sounds like you're doing everything right.  It could just be that particular roll of filament.
And as Jago said, go get a cheap thin piece of hardware store glass cut to size for a few bucks.
Are you using .2963 for your layer height?

Yeah, my kid's cheap USB microscope starts at 10X, and my iPhone doesn't do well on closeups.

My glass is on order.  8*)

Layer Height is 0.1mm
First Layer is 0.35mm

8 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

Yeah, for a part like that you want to use a .3 layer there is no benefit to a .1
You want to use the proper z height for those as well.  The SD comes with a 5/16" Z rod so you need values that actaully match up to the metric nature of the rest of the system.  This is very important and is often over looked.
.3=.2963
.2=.1976
.1=.09878

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

9 Reply by w_smith

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Re: First 4(?)mm of object doesn't print right?

wardjr wrote:

Yeah, for a part like that you want to use a .3 layer there is no benefit to a .1
You want to use the proper z height for those as well.  The SD comes with a 5/16" Z rod so you need values that actaully match up to the metric nature of the rest of the system.  This is very important and is often over looked.
.3=.2963
.2=.1976
.1=.09878

OK, I'll switch to 0.3mm layer height when the current print is done.  I assume I need to recalibrate flow rate?

Should I just use those layer heights, or should I calibrate the Z axis and change the M92 constant in the firmware so it's really moving 1.0000mm when I tell it to move 1.0000mm?

The Z constant is currently 2268.0 steps per unit, if that means anything

Does the 'modern' (got it in Nov) SD4 have the same problem, or do I not know without calibrating it?  What's the best way to calibrate, tell it to move 100mm and see what it actually does, or use a dial indicator on a 1mm movement?

Thanks!

10 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

Just use those numbers as your layer height instead of .3 or .2 or .1
all the hard work has been done for you.

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

11 Reply by TickTock

  • TickTock
  • TickTock
  • Member
  • Offline
  • Registered: 2014-05-28
  • Posts: 263

Re: First 4(?)mm of object doesn't print right?

Don't touch the Z constant.  It *is* moving 1mm when you say 1mm.  You are changing the layer height so it moves .2963 *instead* of 0.3000.  This is to ensure that the stepper stops at an integer step since the fractional steps are prone to non-linearity. You can see variation from layer to layer if it stops at a different fractional step each layer so using the numbers wardjr provided avoids this error.

12 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

TickTock wrote:

Don't touch the Z constant.  It *is* moving 1mm when you say 1mm.  You are changing the layer height so it moves .2963 *instead* of 0.3000.  This is to ensure that the stepper stops at an integer step since the fractional steps are prone to non-linearity. You can see variation from layer to layer if it stops at a different fractional step each layer so using the numbers wardjr provided avoids this error.

Thank You for the explanation I'm on the phone ordering computers sooo....

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

13 Reply by w_smith

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Re: First 4(?)mm of object doesn't print right?

TickTock wrote:

Don't touch the Z constant.  It *is* moving 1mm when you say 1mm.  You are changing the layer height so it moves .2963 *instead* of 0.3000.  This is to ensure that the stepper stops at an integer step since the fractional steps are prone to non-linearity. You can see variation from layer to layer if it stops at a different fractional step each layer so using the numbers wardjr provided avoids this error.

Ah, now I get it, thanks!

And do I need to recalibrate the flow rate?

14 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

w_smith wrote:
TickTock wrote:

Don't touch the Z constant.  It *is* moving 1mm when you say 1mm.  You are changing the layer height so it moves .2963 *instead* of 0.3000.  This is to ensure that the stepper stops at an integer step since the fractional steps are prone to non-linearity. You can see variation from layer to layer if it stops at a different fractional step each layer so using the numbers wardjr provided avoids this error.

Ah, now I get it, thanks!

And do I need to recalibrate the flow rate?

NO

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

15 Reply by Goshdarnit

  • Goshdarnit
  • Member
  • Offline
  • Registered: 2015-03-05
  • Posts: 162

Re: First 4(?)mm of object doesn't print right?

Are those numbers also correct for the Press?

16 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

I don't know what the Press uses for the Z-rod if it is 5/16 - 18TPI then it would be the same.  Given that it is made in China I would guess that it is a metric rod.

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

17 Reply by Goshdarnit

  • Goshdarnit
  • Member
  • Offline
  • Registered: 2015-03-05
  • Posts: 162

Re: First 4(?)mm of object doesn't print right?

Ah OK. Thanks for the rapid reply.

18 Reply by w_smith

  • w_smith
  • Member
  • Offline
  • Registered: 2014-11-23
  • Posts: 97

Re: First 4(?)mm of object doesn't print right?

wardjr wrote:

Yeah, for a part like that you want to use a .3 layer there is no benefit to a .1
You want to use the proper z height for those as well.  The SD comes with a 5/16" Z rod so you need values that actaully match up to the metric nature of the rest of the system.  This is very important and is often over looked.
.3=.2963
.2=.1976
.1=.09878

What should the first layer height be for each of those layer heights?

Thanks!

19 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,624

Re: First 4(?)mm of object doesn't print right?

w_smith wrote:
wardjr wrote:

Yeah, for a part like that you want to use a .3 layer there is no benefit to a .1
You want to use the proper z height for those as well.  The SD comes with a 5/16" Z rod so you need values that actaully match up to the metric nature of the rest of the system.  This is very important and is often over looked.
.3=.2963
.2=.1976
.1=.09878

What should the first layer height be for each of those layer heights?

Thanks!

That all depends on preference, I like to run my first layer height at the same height to keep things simple.  I then adjust my z end stop to achieve the proper amount of squish on that first layer.  I have found that by decreasing the extrusion width of the first layer to something like .36 will greatly improve adhesion.  I know others that use a higher first layer height for the same reason but I haven't found that to be as reliable.

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website