krouviere
Posts: 18
Joined: Wed Aug 27, 2014 6:21 pm

Re: Raft separation distance and next layer height problem

Fri Oct 16, 2015 3:56 am

It's not just changing the representation in the preview. It's changing the gcode that controls the machine.

For instance, here are sections of gcode from 2 different raft separation settings. The first is with .14mm and the 2nd is with 1mm.
In the first example, layer 7 (the first layer after the raft) is at Z = 2.2473 and layer 8 is at Z = 2.3073. Layer 8 is above layer 7.

In the second example, layer 7 is at Z = 3.1073 and layer 8 is at Z = 2.3073. Layer 8 is below layer 7.

Notice layer 8 Z setting did not change regardless of the separation. Layer separation only affected one layer instead of shifting the whole model.

So layer 7 is the only layer that got the separation applied, and then the nozzle dives back down to the same height for layer 8 regardless of the setting you applied that is supposed to elevate the entire model above the raft. You can effectively never get more separation except for the first layer of the model (layer 7 in this case) and at best, layer 7 didn't stick to the raft quite as badly however the nozzle will basically push layer 7 down to the raft when it does layer 8. This is just really ugly.

Whether or not there is nothing but air under the nozzle allowing layer 7 to fall to the raft is beside the point. It's simply doing the wrong thing in the gcode. The preview is working like a CNC backplot function to show what the generated code will do. With the powerful multi-process and multi model capability in S3D you can construct complex print jobs or do other things to create something other than air between layer 7 and the raft.

GCode below has only layers 7 and 8 for simplification:
-----------------------------------
First with .14mm separation.

; layer 7, Z = 2.2473
; tool H0.314 W0.400
; inner perimeter
; outer perimeter
G1 X17.182 Y17.199 F6000
G1 Z2.247 F600
G1 E1.0000 F1200
G1 X0.818 Y17.199 E0.8549 F3600
G1 X0.818 Y0.835 E0.8549
G1 X17.182 Y0.835 E0.8549
G1 X17.182 Y17.199 E0.8549
; layer 8, Z = 2.3073
; tool H0.200 W0.400
; inner perimeter
; outer perimeter
G1 X0.818 Y-0.835 F6000
G1 Z2.307 F600
G1 E1.0000 F1200
G1 X0.818 Y-17.199 E0.5443 F1800
G1 X17.182 Y-17.199 E0.5443
G1 X17.182 Y-0.835 E0.5443
G1 X0.818 Y-0.835 E0.5443


Next with 1mm separation.

; layer 7, Z = 3.1073
; tool H0.314 W0.400
; inner perimeter
; outer perimeter
G1 X17.182 Y17.199 F6000
G1 Z3.107 F600
G1 E1.0000 F1200
G1 X0.818 Y17.199 E0.8549 F3600
G1 X0.818 Y0.835 E0.8549
G1 X17.182 Y0.835 E0.8549
G1 X17.182 Y17.199 E0.8549
; layer 8, Z = 2.3073
; tool H0.200 W0.400
; inner perimeter
; outer perimeter
G1 X0.818 Y-0.835 F6000
G1 Z2.307 F600
G1 E1.0000 F1200
G1 X0.818 Y-17.199 E0.5443 F1800
G1 X17.182 Y-17.199 E0.5443
G1 X17.182 Y-0.835 E0.5443
G1 X0.818 Y-0.835 E0.5443

krouviere
Posts: 18
Joined: Wed Aug 27, 2014 6:21 pm

Re: Raft separation distance and next layer height problem

Fri Oct 16, 2015 4:09 am

"Both can be said to be either right or wrong....but how can this be correctly represented to satisfy everyone??"

Only one of them is right. :D
It doesn't need to satisfy everyone, it just needs to be right :D :D :D

V3.x is doing the wrong thing.

I don't mean to sound like a jerk but this is clearly a bug or a bad decision to purposely change it away from how it worked in 2.x.

User avatar
dkightley
Posts: 2164
Joined: Tue Mar 10, 2015 4:09 pm

Re: Raft separation distance and next layer height problem

Fri Oct 16, 2015 8:00 am

In a bid to help move this on, I've done a comparison between v2.2.0 and v3.0.2:

Run 1- a 10cm cube run on a fresh install of v2.2.0 using the machine default profile (0.2mm layer)...modded only to add a raft with a separation distance of 0.5mm

Run2 - the same cube run on v3.0.2 using the above default profile exported from v2.2.0 and imported into v3.0.2

Run 3 - as Run 2 but first layer separation of zero.

The z location values for layers 1 to 6 were identical at 0.45, 0.95, 1.32647, 1.54869, 1.77092, and 1.99314. This is the raft.

Layers 7, 8, 9 and 10 for each run:

Run1: 2.6073, 2.8073, 3.0073, 3.2073
Run2: 2.6073, 2.5073, 2.7073, 2.9073
Run3: 2.3073, 2.5073, 2.7073, 2.9073

So, the theory put forward that I supported does not prove to be correctly implemented in the code....and there does seem to be some more work required in this area to get a working solution.

The v2.2.0 code did (surprisingly) print the first layer of the model in fresh air, but then printed subsequent layers on the basis that the first layer didn't drop. This obviously worked for small separation distances, but using a silly distance like 2.5mm probably resulted in a totally messed print as the second layer would not have been printed snugly to the first layer.

In a change that I assume was to try and handle oversized separation distances, I assume a code change was made which looks as if it ended up with the second layer onwards being printed as if there was no separation distance...and not at a calculated distance taking the height of the non-amalgamated first layer into account.

So...to conclude....based on the evidence I have laid out above, I now also believe there is a bug in v3...insofar that more work is needed on achieving the desired effect of easier raft separation.
Doug Kightley
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net

dunginhawk
Posts: 39
Joined: Mon Jan 19, 2015 10:15 am

Re: Raft separation distance and next layer height problem

Fri Oct 16, 2015 11:29 am

so with all this being said and verified, is there any comment/input from the S3d team on this? Rafts are pretty important, and if i cant use them, I dont see why i want to use s3d at all for prints with rafts. I love the software otherwise.
thanks

jomireyn3
Posts: 22
Joined: Wed Aug 05, 2015 11:08 am

Re: Raft separation distance and next layer height problem

Fri Oct 16, 2015 12:39 pm

dkightley wrote:
In version 2.2, this worked correctly.
For example, setting to a .014mm gap yields the result in the first picture below.
Setting to 1mm gap yields the result in the 2nd picture below.

This is exactly what I would expect. I have control over what the machine is doing. I can also easily see the result of changing a value in the user interface and validate that the control is changing the parameter I want to change. And as odd as it may seem, I may actually want this if I am doing something creative with multiple processes and overlapping models.

I don't see any downside at all to making it work like it used to. In fact the way it is now, I don't think there is good control between 1st layer after separation and the layer after that. The way it worked prior to 3.0, the 2nd layer after separation would have a fixed relationship with the 1st layer after separation.

Sure, I can goof up and set the gap too large, but that's my fault. I want the control.
I'm at a loss as to how changing how this is represented in the preview will have an effect on the printed model. The v2,x preview showed the second and subsequent layers printing above where the first layer was printed....showing the model "levitating" above the raft. This graphical representation looks as if has been changed in v3 to show the first layer printed at the elevated height and then the second and subsequent layers being printed at approximately the height where they actually is printed in real life. The effect looks equally strange. The model no longer "hovers", but looks as its been mis-drawn. Both can be said to be either right or wrong....but how can this be correctly represented to satisfy everyone??

Surely the real issue is the first layer of the model is for some reason sticking too hard to the raft after being printed in fresh air and dropping under the effect of gravity. To me, this is probably caused by other factors...and not how high the layer is printed! Or how it is represented in the preview!


Well, my assumption was that the print preview was an illustration of what the printer would do. So if the gcode instructs the print head to be a certain z height, then that's what's illustrated in print preview. If I see from the illustration that the z height will below layer 7 at layer 8, I assume that that's what the print head will do too. So it's the same thing. If the preview doesn't show what the gcode instructs the printer to do, then illustrating the separation distance (or anything else for that matter) is pointless.

I don't know if the gcode generation is different or just the illustration in print preview is different, all I know is that I can't use this software to do what I bought it for, and other people are having the same problem.

Also, I think the definition of the separation distance IS how high the layer is printed, and the separation distance is Simplify3D's stated tool for adjusting the adhesion of the raft.

krouviere
Posts: 18
Joined: Wed Aug 27, 2014 6:21 pm

Re: Raft separation distance and next layer height problem

Sat Oct 17, 2015 1:00 am

FWIW, the response I got from support was that it "was" an intentional change. Support forwarded the issue to the development team to take another look. I hope they really do, because I hate to see it broken. It worries me more that they don't seem to think that it's broken. It's a departure from the "total control" approach that I like which is why I bought it to begin with.

I think the more voices they hear about it, the more likely it will get fixed. It may not be a top priority for them but as long as it makes it onto the bug list it should eventually get fixed.

The large separations illustrated in the examples posted in this thread point out the problem. I don't think anyone here is really suggesting they would use huge gaps (unless doing something tricky on purpose with multiple processes and models). It's important to be able to specify what you want it to do though.

Typically I print with ABS and find that the .14mm default gap is too much for my Makerbot 2X. I usually reduce it to .07mm for ABS. I always have more trouble with PLA when it comes to separating raft and support from the model.

Maybe they would let you downgrade until it gets fixed

PPLDC3
Posts: 12
Joined: Wed Jun 24, 2015 4:14 pm

Re: Raft separation distance and next layer height problem

Fri Nov 13, 2015 5:22 pm

Adding my voice to this thread. The second layer of the print is just squishing down all the material from first layer after the raft. It's impossible to remove the print from the raft, and makes it near-impossible to use our Airwolf printer.

jomireyn3
Posts: 22
Joined: Wed Aug 05, 2015 11:08 am

Re: Raft separation distance and next layer height problem

Fri Nov 13, 2015 11:13 pm

PPLDC3 wrote:Adding my voice to this thread. The second layer of the print is just squishing down all the material from first layer after the raft. It's impossible to remove the print from the raft, and makes it near-impossible to use our Airwolf printer.

I'm the OP on this. I have given up on using this software at this point with our Makerbot 5th. I contacted support and got no help, other than - to paraphrase - "It's supposed to do that. It's not doing what it looks like it's doing in the print preview. Have you tried (all the things that I already told them that I had tried.)?" Thanks.

jomireyn3
Posts: 22
Joined: Wed Aug 05, 2015 11:08 am

Re: Raft separation distance and next layer height problem

Fri Nov 13, 2015 11:17 pm

dkightley wrote:In a bid to help move this on, I've done a comparison between v2.2.0 and v3.0.2:

Run 1- a 10cm cube run on a fresh install of v2.2.0 using the machine default profile (0.2mm layer)...modded only to add a raft with a separation distance of 0.5mm

Run2 - the same cube run on v3.0.2 using the above default profile exported from v2.2.0 and imported into v3.0.2

Run 3 - as Run 2 but first layer separation of zero.

The z location values for layers 1 to 6 were identical at 0.45, 0.95, 1.32647, 1.54869, 1.77092, and 1.99314. This is the raft.

Layers 7, 8, 9 and 10 for each run:

Run1: 2.6073, 2.8073, 3.0073, 3.2073
Run2: 2.6073, 2.5073, 2.7073, 2.9073
Run3: 2.3073, 2.5073, 2.7073, 2.9073

So, the theory put forward that I supported does not prove to be correctly implemented in the code....and there does seem to be some more work required in this area to get a working solution.

The v2.2.0 code did (surprisingly) print the first layer of the model in fresh air, but then printed subsequent layers on the basis that the first layer didn't drop. This obviously worked for small separation distances, but using a silly distance like 2.5mm probably resulted in a totally messed print as the second layer would not have been printed snugly to the first layer.

In a change that I assume was to try and handle oversized separation distances, I assume a code change was made which looks as if it ended up with the second layer onwards being printed as if there was no separation distance...and not at a calculated distance taking the height of the non-amalgamated first layer into account.

So...to conclude....based on the evidence I have laid out above, I now also believe there is a bug in v3...insofar that more work is needed on achieving the desired effect of easier raft separation.

Thanks for your efforts on this, Doug. That seems to confirm what I'm seeing in the print previews, despite the support team telling me that the preview doesn't represent what actually happens

User avatar
dkightley
Posts: 2164
Joined: Tue Mar 10, 2015 4:09 pm

Re: Raft separation distance and next layer height problem

Sat Nov 14, 2015 5:07 am

Thanks for your efforts on this, Doug. That seems to confirm what I'm seeing in the print previews, despite the support team telling me that the preview doesn't represent what actually happens
To reiterate what I said......the preview DOESN'T represent the end result of the extrusion process, quite correctly as the S3D support staff say. It does, however represent EXACTLY what the slicer code is telling the hardware to do.

What is in question is the reasoning behind extruding the first layer in fresh air a certain distance above the raft....the possible theory being to allow the layer to "rest" on top of the raft - and thus not adhering strongly,....and then ramming the layer down into the raft by extruding the second layer of the model as if the first layer had been extruded normally.

Using common sense, I would work on the principle that the first layer having been "rested" on the raft, then the height of the nozzle for extruding the second ( and subsequent layers) would be very slightly higher....to take into account the non-squished first layer. The alternative would not to bother at all with trying to allow any separation at all!
Doug Kightley
Volunteer at the National Tramway Museum http://www.tramway.co.uk
Railway modeller and webmaster at http://www.talkingtgauge.net

Return to “Troubleshooting and Bug Reports”