It works! Sorta...

Leveling the print bed, adjusting the z-level, and getting high print bed temperature will all lead to better prints.  Without an enclosure but in a room with a good ambient temperature, it takes my printer about 10-15 minutes to reach 95.  With paeltz's enclosure, it takes 5-10 minutes (http://www.soliforum.com/topic/1588/my-easy-and-relativley-cheap-sd3-enclosure/ ). 

Beyond that, I have had trouble with my SD3 and sticking and warping.  It is improved by wiping the print bed with alcohol or acetone.  If still having trouble, setting the speed of the 1st layer to be 1/2 or even 1/4 that of normal printing may help.  Printing with a brim is a last result.  For me, I am now using the enclosure, I'm heating to 100, and I'm using Cubestick from Cubify, but I'm doing all that because I'm printing on a glass plate.

I made really short thumbscrews - not pretty but functional - so I can keep storing stuff underneath without worry about the thumbscrews coming down too far.

Update:

First off, thanks for the helpful advice and .stls offered.  However, I am hoping to print this version as I want to figure out how to do overhangs.  I plan to print some stuff with overhangs and this seems like a good piece to learn on.

Secondly, I fiddled a bit, and got some stuff working.  I also discovered an odd motor control issue, but I rebooted and it seems to be working.  But what I did was:

1. Calibrated my Z-axis such that 0.00 mm Z was actually on the board (it was .2 mm high), and also set my first layer extrusion to be a bit higher, thus creating a nicely smooshed (but not smashed) first layer.

2.  Reduced speed on my first layer

3. Added a Skirt to make sure that material was flowing properly and ready at the tip of the extruder so I would get a good first layer

4. Added a Brim to give my part extra grip.  This particular model has a very small base relative to the length of it's final shape.  As such, there was a lot of moment when it caught.  This can easily be snipped off at the end (I actually have a pair of offset diagonal cutters perfect for the task).

5.  Set the temperature lower to 195 C extrusion (which supposedly helps).

All that got me the results seen attached:

This is my best one to date before it got caught and ripped off.  As you can see in picture #4, the 'ears' of the piece do indeed bow up.  This bowing gets worse and worse as the layers are built until they catch the print head and rip off (thus you can see the blob on the corner where it caught).  I assume this upwards bowing is due to laying down a hot piece on a colder layer below.  The part sticks, then contracts, thus pulling the top inwards and causing the upwards bowing (like the string on a hunting bow pulling it into a curve).

Ideally this would be solved by having no thermal gradient.  So, if the print head, the build plate, and the surrounding air were all the same, we would be fine.  Unfortunately, the ABS needs to go from liquidus to solidus, which necessitates a temperature shift.  So, what would be the best thing here?

Perhaps raising the bed temp, lowering the extruder temp, and going as quickly as possible in order to reduce the thermal gradient?

Perhaps slowing the whole things down and allowing the layers to completely cool one at a time so as to provide a stable surface that resists the top layer's contraction?

Changing the layers to be lower, thus reducing the contracting force?

Adding solid layers to the infill to act as buttressing?

support structures?

I am probably not the first person with this problem and am curious what other people have done about this issue.