Bizarre DualStrusion failure!

[DaveGadgeteer]

I'm trying to make a fast-to-print pattern for checking extruder alignment, which is clearly off when printing a traffic cone, but that's too time consuming to use as a test pattern.
So I made a trivial OpenSCAD program, and exported the STL separately for the two pieces, then imported them into Simplify3D and connected them to the two extruders, and prepared to print. The preview for printing looks like it's only a kind of skirt or more accurately an oversize outline, and it shows the two pieces as overlapping by half, which doesn't seem to make any sense.
Anyway, the preview is correct! That's exactly what printed.
Evidently I did something wrong (new user here).
Can anyone steer me aright?
Here's the OpenSCAD program:

cube([.4,10,.5],center=true);
//translate([0,10.5,0]) cube([.4,10,.5],center=true);

(Comment the first line instead to get the STL for the other piece.)

I'll attach the factory file.

[DaveGadgeteer]

I found the problem, but couldn't post for 4 days because of waiting for moderation!
The problem is that Simplify3D won't print features that are not larger than the nozzle. I was trying to print lines that were equal to the nozzle diameter.
So I switched strategies, printing things with large features that could be measured with calipers.
I'll attach my file.
Oops, .scad and .thing are not allowed in this forum, so I'll have to post on Thingiverse instead. I can post the .gcode here, though.
I switched to MakerWare for printing, because only that can print with the wipe walls that are helpful in keeping the prints clean, important for this job.
I used ReplicatorG to observe and set the extruder offsets that I was trying to get correct.
Turned out I was making occasional sign errors, forgetting which convention I was using for giving the sign of the error, and that greatly adds to the difficulty of getting convergence!
I found that plotting the position error versus setting on graph paper was a big help for discovering such errors in time to avoid calculating with them.
On my FlashForge Creator, the X printed position for the left extruder moves about -1.5 times the setting movement, and Y moves about -1 times. Wonder why they're different.