after E3D V6 upgrade, over extrusion problems

You may want to look at what the extrusion widths are set at in your slice profile.  Perimeters should be set at .48 but you can tweak the width for infill separately.  This is the tool path width so if over extruding on infill try increasing its width of the tool path.  This will space the path further apart.

oh no, please - not another one using that guide...

There is some good info in there, please dont get me wrong, but I have to take exception to the extrusion calibration info it contains. Too much "fudging" going on...

take a read through this post - Jagowilson does an excellent job of discussing why you don't want to "fudge" with extrusion if you can help it...
http://www.soliforum.com/post/83278/#p83278

Well, I wasn't really suggesting I'd leave the E steps at adjusted value, more like it's a value one can adjust live while getting visual feedback. And thinking a linear relationship between E steps and multiplier, so I can translate the 'measured' 5% excess to the multiplier. Well, I stand corrected.

Thanks for linking the awesome writeup of jagowilson. Certainly helped me clear the extrusion calculation. What I do next I guess is:

Turn E steps/mm back to measured 690 (400 steps motor, drv8825 driver and hobb-goblin)
Re-measure my filament. Might have been a bit sloppy there.
Find the lowest multiplier wher wall can be said to be 0.48mm.
Adjust any exess over extrusion with extrusion width settings on infill.

Does that sound about right?

Thanks for help, really.

veehoo

i never change the "default" extrusion width - unless I were to change the nozzle size of course...

the others I do play with to get good fill on a given model - things like text or small areas almost have to be played with to get them to come out well.

also never need to mess with the E steps/mm once it is set correctly.

what does get changed with each spool of filament is the average diameter and the extrusion multiplier to get that particular filament printing at the desired width.

I currently have 37 filament profiles saved in Slic3r - one for each brand/color/type of filament I have (and I have a few that aren't in there yet - haven't used them yet, or been a long time) Swapping spools is as easy as selecting the correct profile and I get consistent prints with every spool i have, every time.
If I run out on one spool and have to start a new one, the new spool gets calibrated all over again, even if it is the same brand, type & color. I may reuse the profile name, but the actual calibrations get done with each new spool to make sure it prints just as well as the previous one did.

It isn't mandatory to upgrade the motor or driver, but it would help.
I have an SD2 with a Printrboard (1/16) and a stock motor with the Goblin and it runs great.

Upgrading the driver, if possible, is a great work around to a 1.8 motor. If it is not possible, upgrading to a .9 motor is a great work around for the inability to work on integrated drivers. Regardless, the Goblin upgrade is a solid step in the right direction.

Can I ask what version of Slic3r and what version of Marlin ?

Slic3r is version 1.2.9, how would I check Marlin version?

For filament when I get roll open I check first couple feet at multiple points.

The biggest concern about all this is that when I printed at 200 C with old Solidoodle head there was no issue at all, perfect extrusion, then just switching to E3D and 260 C printing overextrusion.

If I only print the walls and no solid infill it is still good, solid infill is the problem area for too much plastic.

260C is a bit hot for ABS - should be more in the 235-245 range - 260 is more of a nylon temp...

for Marlin version, toggle the log on in RH when you first connect to the printer it should show.
Will look something like this:
07:40:28.073 : FIRMWARE_NAME:Marlin V1; Sprinter/grbl mashup for gen6 FIRMWARE_URL:https://github.com/ErikZalm/Marlin/ PROTOCOL_VERSION:1.0 MACHINE_TYPE:Rumba EXTRUDER_COUNT:1 UUID:00000000-0000-0000-0000-000000000000

For the E3D change I had my friend to the software and new power supply, I installed the new head, carriage, and ran the wires, friend just did his SD3 and could do it quick.

I use 260 C for ABS because at colder temps my prints grow cracks in 1-2 weeks. When I acetone smooth it is very evident of multiple white cracks. My friend goes up to 270 C with ABS for better adhesion, he read this somewhere else in this forum site.

As for first layer here are some items
- with old solidoodle head I had a nearly perfect flat first layer, no weeds or anything, no plastic bulges or bumps. Now with E3D it is a mess. I only use Hatchbox, only variable that changed was printer head and temp of printing.
- build plate is level, I have a zebra plate attached with binder clips and I have sanded it many items so it has created a little waviness that I live with.
- For my first layer I jam the plastic into my build plate so I get perfect adhesion and no lifting during print. I know that causes some sputtering or irregularity in some areas when combine with my worn zebra plate.

How can I make solid infill lines of plastic be spaced out while keeping spacing for everything else?

I am sorry, but this image says you do not have a level bed...

I have never had parts crack several days later, either... if they are going to crack/split, it is usually during printing, or right after taking it off the print bed from cooling too quickly (impatience) .. days later? nope.

FWIW, printing too hot can cause just as many issues as printing too cold. you need to find that sweet spot that is just right.

I am thinking that the cracks you are experiencing are from using the incorrect layer height - using 0.3 as a layer height will cause rounding errors in the steps on a Solidoodle with the stock 5/16" threaded rod which will cause problems in your layers not adhering together well enough.

Wardjr is not just making up numbers, either - using the RepRap Calculator gives the same results as what he is suggesting...

Try it for yourself: http://prusaprinters.org/calculator/#stepspermmlead

Yep and if you could count the layers you'd most likely find a pattern.  Cracks every X number of layers.