Questions about calibration

Hello
I'm working through the calibration wiki. I'm getting reasonable prints but am having trouble with bridging and am wondering if I have'nt done some of the previous steps right.

I'm using slic3r 0.7.1 and pronterface. Printing with 3mm Yellow PLA through a .5mm JHead IV I think. I have a 120mm Fan blowing lightly after first layer which improved .5mm thin wall dramatically.

First I don't get how bedleveling.stl is supposed to help me. When I load it up it prints nothing.

To get .5mm thin wall looks best with layer height at .3mm and I think it's slightly better with extrusion multiplier set to .9. Does this mean I should be reducing e-steps?

I'm doing Infill nowfor a third time. It gets halfway through then the print comes away from my glass heatbed. My heatbed cant keep up with the fan blowing and manages only 50 degrees. Will this effect how it holds to the glass?

Temerature controll 50mm tower: Is the idea to set the temperature as low as possible or as high as possible without drooping? I'm using 165 acording to ramps

Perimeter-wt.stl doesn't work for me? Do other tests depend on this?

I've set bridge feed rate multiplier to 1.1 but box seems to just fill up with little bridging. Is it supposed to make random piles that allow top to rest on?

My 20mm hollow box

I dont print with PLA. ABS exclusive so take the following with a grain of salt.

Quote

First I don't get how bedleveling.stl is supposed to help me. When I load it up it prints nothing.

check to make sure it's scaled to fit your build area and set your layer height to .4, also might try adding infill, perimeters, brim or whatever you can with the version of Slic3r you have to make it print something. the idea is that it prints 4 thin squares 1 layer high on each corner of your print bed so you can see if your print bed is perfectly level and even at every corner. if they come out squished looking on one corner and barely touching the print surface on the oposite corner then you have some leveling to do.

Quote

To get .5mm thin wall looks best with layer height at .3mm and I think it's slightly better with extrusion multiplier set to .9. Does this mean I should be reducing e-steps?

Yep. I reduced my multiplier from 1.0 to 0.60! Thats after calibrating the extruder. You can just leave the multiplier setting like it is for every print if you dont feel like re-uploading your firmware.

Quote

I'm doing Infill nowfor a third time. It gets halfway through then the print comes away from my glass heatbed. My heatbed cant keep up with the fan blowing and manages only 50 degrees. Will this effect how it holds to the glass?

50C is plenty for PLA. The problem isnt the bed temp. Your not printing on plain glass are you? throw down some blue painters tape and wipe down the tape with some acetone to clean it. Your prints will stick much better.

Quote

Temerature controll 50mm tower: Is the idea to set the temperature as low as possible or as high as possible without drooping? I'm using 165 acording to ramps

As low as possible! The tower shouldnt be a big deal for PLA (I think) because PLA isnt very sensetive to curling with temp changes. Instead it shrinks and you get layer seperation, but PLA is meant to be heated barely enough for it to become maleable and extruded.

Quote

Perimeter-wt.stl doesn't work for me? Do other tests depend on this?

I have no idea what that is, can you link it?

Quote

I've set bridge feed rate multiplier to 1.1 but box seems to just fill up with little bridging. Is it supposed to make random piles that allow top to rest on?

When bridging it's supposed to just create a bridge with nothing underneath it. Try increasing your bridging speed.

Hope this helps.

---

-Tom

I've limited my Y axis to 195 because thats all I can get with my opto end-stops mounted the

way they are. I am not sure how to select scaling or anything like that. But I'm not too

fussed about bed leveling as I check this pretty accurately.

.85 extrusion multiplier left gaps in layers. .88 was the sweet spot. I just printed a .5mm

thin wall with .1mm layers and it's awsome.

The tape is great. I was trying to avoid using it because I have read you don't need to and I

had pictured a mess at the end of each print. I now need to print a little hammer to help

get the prints off.

perimiter-wt.stl is a model used in the calibration wiki.http://www.thingiverse.com/download:17277]perimiter-wt.stl[/url]

I'm still working on bridges but its getting better

The .5 thin walled object is meant mainly to check your extrusion rate, have a look at the code you generate for it, in the header will be listed single perimeter width, this is what you should be checking on this object, mease the wall width of the last extrusion and calibrate either steps or multiplier till the average of the sides matches that.

The solid block then checks that 100% fill doesn't overfill, overfilling may be the cause of your block being dislodges from the bed

Thanks I'll check this now before I print the box for the 9th time today. In desperation I thought I'd try slic3r 0.9.3. Just tried to slice the 20mm-hollow-box and I get an error message "illegal division by zero at ......slic3r/gcode.pm line 177" Now when I run the old 0.7.1 I get windows telling me slic3r has encountered an error..

reboot time.

So I measure .53 .5 .53 .53 for an average of .5225
A look at the header shows single wall width=.53
At least I feel like I'm on the right track.

Edit: It seems when it starts the bridge it is supposed to lay an internal lip around the inside edge of the box, but one or more sides usually drop straight into the box. Then when the diagonal briging starts it just falls in.

Edited 1 time(s). Last edit at 10/26/2012 12:16PM by Gibbedy.

By "one or more sides" do you mean one or more strands of filament? That happens to mine sometimes. Try increasing the Solid Layers. Usually I like to do 3, but always at least 2.

Your single wall width seems ok. Remember the part will shrink a tiny bit as it cools.

---

-Tom

What I mean is just before starting the bridge it extrudes around the internal edge of the box. I presume this is meant to stick to the edge of the box but at least two sides always fall in. I've tried bridge rate from .8 to 1.6 in .1 increments and they all look pretty poor.

Thanks.
Gavin.

Strange. I'm using 0.8.9 atm and it just starts a bridge then adds a rim followed by another layer on the bridge. You might want to look into later versions of Slic3r for that problem.

---

-Tom

I'll try and sort out my illegal division by zero problem with 0.9.3 tonight and try again,
edit: just printed 20mm hollow box with slic3r 0.9.3 and got same result. I'll start increasing the Bridge flow ratio and post a good video of what is happenning if I still have the problem.

Edited 1 time(s). Last edit at 10/27/2012 12:00PM by Gibbedy.

Here is a video (of still images with text) showing my results at different bridge ratios.
I'm starting to think there is something else wrong as the longest diagonal seems to hold well and I would have thought that would be the toughest to do if bridge flow was wrong.

Here is a video of the process

Edited 4 time(s). Last edit at 10/28/2012 01:41AM by Gibbedy.

You won't be able to print good bridges or sparse infills at 0.3 mm layer thickness and 0.5 mm width with a 0,5 mm nozzle because you're stretching the extruded filament too much. Let slic3r pick the extruded width automatically and you would have better success. If you really want to print at those thickness and width, then get a smaller nozzle like 0.4 or 0.35 mm.

Sorry I should have mentioned. All of the prints shown were done with .4mm layer thickness. Should it be set to this?
slic3r is deciding on layer width. The only values I have set are layer thickness, extrusion multiplier and bridge flow ratio. The only other setting I have changed is the retraction speed from 30mm/s to 15mm/s because It was completly skipping some retractions at the default speed. I still haven't gotten back to what that was about.

when I'm watching it closely I dont see the extruded plastic stretching and breaking. I see it not sticking to the side in the first place, like the position is 1 layer out when making a bridge.

Would it be possible to get a lower bridge test stl. It just takes a bit of time to build up the box just to get to the bridge part.

Thanks for the help brnrd.

Sorry, I misread your message.

The problem is that you're setting for a 0.5 mm extruded width but I'm sure that slic3r determined a much wider extruded width than that. You should look in the gcode file to see what extrusion width slic3r has calculated. You can do this after loading the file in Pronterface and using the edit menu option. Then you should compare that width with the measured width of your single-wall object. Then adjust the extrusion multiplier as follows:

new extrusion multiplier = (current extrusion multiplier) * (expected width)/(measured width)

You may still need to tweak this value a bit to get the correct extruded width.

Once you get the correct extruded width on the single-walled object, you can work on bridges. Set the bridge parameters to the default settings beforehand.

I use this simple bridge test piece from Caru: [www.thingiverse.com].

Edited 1 time(s). Last edit at 10/28/2012 06:43PM by brnrd.

Gibbedy Wrote:
-------------------------------------------------------
> So I measure .53 .5 .53 .53 for an average of
> .5225
> A look at the header shows single wall width=.53
> At least I feel like I'm on the right track. ....

I measure what the header says it should be. I have also tried +-.01 on extrusion multiplier and I still get the same bridge issues. So are you saying slic3r's calculated layer width is different from the actual width extruded and therefor the head of the extruder is starting its bridge and missing the wall? This is exactly what I am seeing but the numbers on my vernier and in my Gcode match.

I can't fault my extruder slipping at the motor or shaft/gear. Is it a possibility that my jhead mkIV is slightly undersized as I have read somewhere?

My .4mm layer height .5mm thin walls come out great with the occasional layer slightly out of alignment. .3,.2,.1 are all in line as much as I can see.
Thanks. Will experiment more tonight.

OK. I'm still catching up.

I didn't realize that you're trying to close the top of a single-walled cube. I don't think this would work. When it prints the layer to close the top, you can see the printer laying down the the perimeter 3 times. One loop goes on top of the previous perimeters. But the other two loops goes on top of air and mostly falls down. When it closes the top with the bridge, the threads have nothing to stick to and most of them falls except where the inner perimeters are still close enough to support them.

Try printing Caru's test piece that I suggested.

You are right that if the nozzle size is actually smaller than 0.5 mm, then bridges will fail since slic3r calculated the extruder steps based on that assumption and you end up extruding too little plastic. You can decrease the nozzle size in slic3r to see if this improves the bridges if the 0.5 mm setting doesn't work with Caru's test piece.

Caru's test piece printed great.
But if I print a part the bridges fail in the same way (the bridge starts in air).

Caru's piece printed without infill. You should change the number of perimeters to 2 instead of 3 in slic3r. Also, in the advanced tab (slic3r 0.9.3), make sure that the extrusion width for the first layer is set to 100% or it won't fill that layer, For this test, you want the bridge to only one have layer or stop the print after the bridge layer. You're want to see all the threads make it across the bridge.

what is the other part that you're printing?

I changed perimters from 2 to three 3 to 2 which made caru's print solid. First layer of bridge is perfect. I'm printing this piece for a filament spool. with infill set to .15 fill density it is lik a wire mesh but allot of the mesh doesn't join. I was watching it when it tried to bridge on the ouside edge of the part about halfway throught the print and it missed again. I'll try the 20mm hollow box again now.

Thanks.

Edited 1 time(s). Last edit at 10/29/2012 12:22PM by Gibbedy.

changing perimeters from 3 to 2 fixed my 20mm hollow box. I paused the print after the first layer of bridge. If this look ok to everyone then I'm off to print some parts.

You should be able to print that piece for the filament spool at 0.15 infill now with your settings.

Thanks. Part printed well

Trying to Calibrate our Printer with a series of tests. The "thin wall" test seems to be producing a pretty peculiar malfunction. One wall prints quite smoothly, while the other 3 seem to experience quite a lot of "jitter". I attached a video. Listen closely because the sound is not that great.

https://vimeo.com/67898963

Any suggestions on what this problem could be?

Andrew.