Printing Quality

That was my initial thought, but it also seems to happen in the middle of the part, and it also seems to happen in areas that have tight geometry:

I'd called Da Vinci Support and they said to use '100'  as the calibration target. 
This was printed with layer set to .2 . . . so based on other posts . . .

So here is the real question:

I know that (in my present case) very short lengths of extruded filament tend to 'bump up'

Could this 'roughness' be based on not having the proper 'calibration height'?  Da Vinci doesn't explain what 'Calibration' is, and I've been spoiled with using 'fire and forget' printers.  I'm using a 2.0 Duo, and when it was shipped, the leveling nuts were so loose, they actually fell off after the first few prints.

I'm thinking of designing some small models that will print in a short period of time to experiment with changing the 'Calibration' number (assuming that the cal number is the number in parenthesis).

------

A little background- I bought the printer to prototype a product I'm developing, I needed to iterate a design and examine an aspect of it.  I've now gotten through the R&D phase and I can spend spare time playing with the printer to get a better understanding about what it's really capable of if it's tuned right.

I'm thinking of taking the geometry of this part, but squeezing it to something about 1.5" long and making it about .1" high and playing with different bed mounting (blue painters tape, kapton tape, dissolving glue in water then spreading the glue with a knife blade and letting the water evaporate, etc).  Then experimenting with the layer and fill settings in the stock software.

I've been nervous about letting the printer run completely unattended, and that sucks when a print run is 9 hours long.  So I feel like I need to understand this printer a lot better.

-----------------

Here's another detailed look at a part that I think does not have 'warping issues'.  It has areas where the filament has to fill tight geometries.  I haven't played with the 'shell' setting either, but I'm thinking of doing that  to see what effect all the settings have on the surface of the 'as printed' part.

Carl, Thanks for pointing me towards the other thread.  Is there a thread that explores the gap between the head and the bed/part?

In a call to Support at Da Vinci in the first week I had the Duo printer, I was told to calibrate the printer to achieve 100.

I suspect that print quality has everything to do with the layer thickness and gap.  I had so many problems early on and I didn't have the time to explore it, that I just started using .2 as the layer thickness because that seemed to print the best . . . or at least 'good enough' for my early purpose.

So now I want to explore the relationship between gap and layer thickness (and shell thickness) as it relates to top layer quality.

Carl said:

I don't have feeler gauges, but I am planning to print a part with dimensions that I might be able to use to confirm.

I plan to create a triangular part that is 2.5 mm high, whose corners are above the leveling studs.  I'll print this using a variety of settings and I'll record the values (including the 4th 'cal' value reported by the printer).  I've got a microscope, so I can count the layers. I'm hoping that this exercise will give me the insight about the printer I'm looking for. 

I know that the members of the open source community have all this figured out for the printers they've used, but I'm a beginner and this printer is . . . what it is.

Unless this object is an integral part of some important proprietary design, why not post the STL file. I'd be interested in seeing what kind of results I get with my printer

I printed the object on a essentially stock daVinci 1.0 (only modifications are: soldering in new extruder heater clip, placing electricians' tape over the filament sensor in the extruder (to allow for the printing of clear filament), using a glue gun to secure the z axis bearings and the use of a filament resetter (for this sample print I used daVinci filament with a reset cartridge in the printer and a spool of daVinci filament hanging off the top).
Sorry, I did not notice until after I started the print that you use a daVinci 2.0... I have one of those as well, but I find the results pretty much the same on both machines. If you don't have any success with my remedy I'll print it off on the 2.0 just to make sure.
I like you am interested in seeing what one can expect from a stock daVinci, so I'm really not interested in flashing firmware and god knows what else. This sample was printed with mac XYZware 1.1.1 (using little snitch to block all communication with XYZ) and 1.2.3 firmware.
As you can see from the first picture (taken immediately after the print head moved off the object) I experienced no warping. Picture 2 shows the object after the daVinci says it has cooled and lowered the bed... you can see it beginning to lift now.
Picture 3 is a magnification of the back corner.
Picture 4 is a magnification of the center hole from the bed side and
Picture 5 a magnification of the center hole from the top side.

All in all I'm pleased with the print, and the only preparation I made was making sure I had a good coating of glue on the bed. After thoroughly cleaning the bed I applied the glue left to right using moderate pressure, then front to back, then angled from left front to right back, the from right front to left back then front to back again and finally left to right again.
The only problem I can see is the term "moderate pressure" because its a subjective term. What I aim for is a coating of glue on each pass that is as wide as the glue stick (I used the larger diameter uhu) and sufficient enough to completely coat the path, but not so much as to have it clump or deform the glue in the holder so that it starts bulging at the side.

I matched your printing specs re density. layer height and speed.
for calibration i just use the XYZ firmware's calibration function. I've never particularly concerned myself with the numbers and i'll go with anything the firmware accepts.
although I did discover that the 1.0 is happiest with numbers in the 200 range, but that the 2.0 has to be closer to 100 or the extruder bumps into the glass of the bed.

that is due to same old issue. upper layers cooling too fast causing plastic to contract which 'peels' uncured threads loose across gaps like inside hollow curves or in this case between the higher sides looking like hairs of piano/guitar strings.

this is much worse on larger, taller objects like a vase for example. some will stretch until cured then break like they did inside my vacuum Cyclone separator inside making for very difficult cleanup. this is not just a type/brand of printer issue. any extruder type 3D printer can have this issue with some materials like ABS that are fussy about shrinkage with more than slight temp differences in layers.

n2ri.... as has been pointed out to you in other threads, this is a daVinci forum. Not a solidoodle forum... your comments are not really helpful here.

Thanks, I think pictures are unnecessary, it's pretty clear from your previous pictures that my setup and yours has some unexplained differences.  I'm working on a modeling project tonight, so I'm going to do another attempt and use 4 layers of glue and a 'stock' gap (I don't know what else to call it) of around 150, with the bed leveling as flat as I can make it:

156, 154,159 'stock' gap: 154 which I think means the closest gap between the nozzle(?) and the bed(?), which is this case is the right front corner for a 2.0 Duo.

------------------------------

It printed well and I used my 'ball peen' hammer and plastic scraper' approach to removing it (it was stuck hard!)   

Results are . . . Good?  My ultimate goal is to figure out how to use the stock software and firmware and cal procedure to get results that are good enough to include as components in fixture systems that I design and build for my clients.  My clients are high tech companies in Silicon Valley (mostly), so they expect professional grade fixtures with professional grade components.

'Good' is ok for R&D work, so I'm going to continue - the '4 layer' glue technique, coupled with a 'stock' gap of 154 for the 2.0 Duo got me to good.  So my next print is to try the same glue technique, with a 'stock gap' of 200.