Problem Print with lots of Openings on base layer

I have not been able to figure out what to change to fix these kind of print problems. If I have large print area like this 165x145mm print with quite a few openings on the base layer, I have a hard time getting the 1st layer to adhere to the print bed. I am printing on an i3v 10" onto a buildtak surface using eSun PLA at 215C & heat bed set to 60C, wades extruder using a Hexagon 1.75 hotend. I used to print at 225C, but found 215C to work better. The bed is manually leveled to .2mm gap, which has worked well for other prints without all the openings. I might try .1mm or .3mm gap next. I am using the original firmware settings that came with this printer. This is a modified version of the sewing bobbin top piece http://www.thingiverse.com/thing:570849 to allow for more bobbins. I made it parametric, but have not uploaded it to thingiverse. I have probably over 500 hours of printing on this buildtak surface, so changing that might help, but since other prints are coming out well I am not inclined to do so. I changed the openings to be rounded on the ends instead of original rectangles & that helped some. The slicer I am using is 0.9.9 version of Slic3r with .2mm layer height & .2mm rectangular infill. I have printed this successfully before months ago, but was not using the buildtak surface then.

Does anyone have some other suggestions to use?

Thanks,
David
Attachment 7408Attachment 7409

Bed gap looks too high. On other prints the extrusion has time to "catch" on the bed and start printing, but on these holes it has to re-catch each hole, and it does not have time. So a smaller gap will reduce the time for the plastic to catch and start.

Also may help, don't do any retraction.

I've also found prints with a bottom detail to be tricky. A lot of things have to be "just right". Concurring with 3DPBuser, bed clearance is critical. Bed needs to be very level if ABL isn't being used. The skirt width seems to vary around your print, or is that just the photo focus? Are you confident in the current extrusion calibration? IIRC Slic3r 0.9.9 tended to under-extrude. How long has it been since any buildup of crud has been cleared out of the nozzle? I have no experience with buildtak, but 3M blue painters tape swabbed with 91% isopropyl alcohol gives me better first layer PLA adhesion than anything else I've tried. By far. I've broken some prints and one glass plate trying to get prints removed from it. Printing onto an unheated bed with the blue tape seems to also give me better bottom-side detail than printing on a heated bed.

I did the e step calibration which I had not done before & changed mine from 841 to 885. The calibration cube looks pretty good with 885 steps. That did not really fix this problem, but certainly should help in overall prints. Seems like the gap was off. I should have checked that again before posting this, but was good to get the reminder of doing the e step calibration. With the buildtak, I do have to use a little larger gap otherwise it will stick too much to the surface. I have not cleared the nozzle lately, but seems to be flowing pretty good thru the hotend.

Quote:

---

Originally Posted by 3DPBuser

Bed gap looks too high.

---

Agreed! You can see the plastic isn't getting pushed into the bed.

I've had the similar problems when printing the first layer too fast as well

Think I fixed my problem. The hotend had too much wobble in it. It has had a slight wobble in it for awhile, but never really bothered my prints, so I was not motivated to fix it. I printed a couple of shims in ABS, but they did not print well. I put a little kapton tape for a shim like I believe printbus mentioned in a post somewhere. What really made it a tighter fit was to put the aluminum plate upside down from what the build guide video showed, but looks like he has updated the way to do that now in a newer build guide. It leaves a little bit of aluminum overhanging on the left side but does not interfere with anything. I might rotate it 180 degrees next time I take it apart as per the new build guide. Here is a before and after photo of a print of a tool holder for a bosch palm grip router on my CNC machine. The after photo is on the left. The before photo weighs 1.56oz & the after part weighs 2.11oz. That sure made a big difference making that hotend a tight fit. I am printing the original problem part now & it is looking really good.
Attachment 7457

Yeah, use of printed shims, a flat washer, a shoulder washer, and my multiple layers of kapton tape have all been ways people have improved the fit of the hexagon hot end in the greg's wade extruder.

FWIW, in my cut at refining the wade's extruder (http://www.thingiverse.com/thing:812899), I added a definition for the hexagon to the other hot end types already in the legacy openSCAD file. That makes it pretty easy to customize the diameter and depth of the large hole for the hot end - I felt that's a better solution than just shimming it until it fits. I also added a customized u-channel plate that will print to the same dimensions as the greg's wade body. No more fretting over which is the top and bottom of the U-channel plate, and no more overhang.

I saw that a while back. Thanks for remixing that. Next time I have to reprint that, I will use your design.

Here is the photo of that problem print that started this thread. It came out quite well.Attachment 7467

How would a wobbly nozzle cause such under-extrusion?

It does seem peculiar, but that was the only thing I remember changing between the bad print & the good print. My guess is wobbling while it was printing was keeping the plastic from flowing at a constant rate somehow.