Retracts too long with 3.0.2?

[pbmax]

He said he doesn't see the issue with g-code produced using the factory file you sent him and asked if you still see the problem on your end. You said you didn't, at least that's how I understood it when I read the whole thread and obviously how he understood it as well. If you still see the problem in your g-code, reply to his email and provide him with a g-code file that exhibits the problem. They can't fix what they can't replicate on their side.

Attacking a perceived reading comprehension issue on the part of S3D support isn't terribly productive here. Sometimes you have to go around the problem a few times before both sides are on the same page.

Or you can try relative retracts as support suggested to me - that fixed it on my end.

[Bking1340]

He asked me "Are you seeing different results in your G-Code than what I'm seeing on my end?" (15 December 2015)
I replied "No also checked my G-code a few days ago, but did not see any ill effects." (15 December 2015)

So "a few days ago" on 15 December is not a few days ago on 15 December

And nowhere have I said that my problem has vanished - I am not a g-code expert and in all my prints that have this "too long retraction" issue, I can not see any problems - I only see 1.5mm retracts as are set in my settings.

I have sent him the factory file that shows no problem(In my stupid opinion - I`m no expert in G-code) on my system, but still the retracts are not always at 1.5mm - That is why I have asked support for help.

And by reading this thread, it seems like I am not the only person with this problem - Don`t know maybe my printer is buggered. I reloaded my custom marlin firmware. Tried again on S3D - same problem. Trying Repetier Host - NO problems on retracts - So don`t know, cant see a mechanical issue here.

[pbmax]

In Process settings under the G-Code tab in the "G-Code Options" box 2nd from the top is "Relative extrusion distances". Check that if it's not checked and see how it works for you.

If you post a known-faulty S3D-created .gcode file here I can take a look at it and see if S3D is screwing up the retractions.

[Bking1340]

Thanks man, yes maybe I was not reading or listening or typing correctly.

Will 1st try a print with Relative extrusion distances ticked and see what happens. The problem with my retraction happens basically on 95% of whatever different stl files I print.
Will post a factory file if my problem don`t creep away.

[kokop76]

I have changed to relative extrusion distances and still get strange retractions and completely consistent significant blobbing in the same areas of prints which does not happen when wipe while retract is turned off. When wipe while retract is on most of the print looks great except for those glaring errors. If I turn it off I get stationary retract blobs which are smaller. Happened to me both on the mac I used to generate gcode to be printed via SD or pronterface, and when I control the printer directly through Linux/simplify.

[pbmax]

Do you have any photos of the problem? Any g-code to post? Factory file?

[kokop76]

Little more work on the issue. I did use kdiff on a pair of g-code files where the only difference was the presence or absence of wipe on retract. The differences we exactly as expected and only were related to the wipe on retract motion. Some of the code isn't as clean as it should be in regards to tiny moves that a stepper will never see, but it's nothing that looks wrong in terms of retract amount, and I don't seem to have extra feed moves after as that would show in the diff output.
The two printed objects, which are the Benchy3d Boat both print well. I get minor artifacts from stationary retracts on the one without wipe on retract, which I'd rather get rid of but can live with. On the one with wipe on retract enabled there are at least two spots where the same thing happens every print. It looks as if there is an extra single outer perimeter laid down for an entire loop. A completely even band, for that whole loop, but not that whole layer. Nothing in the g code at these points suggests a difference from one with wipe on retract turned off.
I definitely had worse results before I enabled relative extrusion, so I am thinking that my remaining issue has more to do with the interpretation of the g code and not the code generated itself. Sorry to have muddied the waters any.

[pbmax]

Interesting. Which firmware? Good luck nailing it down!

[kokop76]

At the moment running smoothie ware. Going to try changing that to see if there is any effect on the same g code.
Another somewhat bizzare result that I had forgot to mention is that there was one other thing that showed up when I "diffed" the files. The total filament length was different by 300mm.
wipe with retract :
; Build Summary
; Build time: 1 hour 23 minutes
; Filament length: 3982.4 mm (3.98 m)
; Plastic volume: 9360.99 mm^3 (9.36 cc)
; Plastic weight: 11.70 g (0.03 lb)
; Material cost: 0.49

no wipe with retract
; Build Summary
; Build time: 1 hour 23 minutes
; Filament length: 3659.9 mm (3.66 m)
; Plastic volume: 8602.91 mm^3 (8.60 cc)
; Plastic weight: 10.75 g (0.02 lb)
; Material cost: 0.45

Perhaps this is more related to the cumulative keeping track of retracts, and the reason that turning on relative retracts helps. That the absolute filament length isn't being kept properly. I'm positivity I would see in the print that there was ~7.5% or so missing or added to one of the boats, and I don't have a scale that would be truly accurate in measuring the weight differences between those two predictions.

[pbmax]

My primary printer (a delta) runs smoothie as well. Thus far I haven't seen anything printed that didn't show up in S3D's g-code preview.