Calibration: Bridging the Hollow Cube

I'm trying to solve a problem with creating bridges. When trying to create bridges, the filament either appears to be too thin, inconsistently stretches across the gap, or holes appear as the hot end drags over the bridge. I tried to print the 20mm hollow calibration cube and got the results in the attachment.

I am using 1.75mm ABS filament and here are a few of my printer settings:

Layer height .2mm
Edge Width over Height Ratio/Infill Width over Thickness = 2.50
Temperature= 225C
Bridge Feed Rate Multiplier (ratio) = 1.0
Bridge Flow Rate Multiplier (ratio) = 1.0
Feed Rate = Flow Rate = 48.0

For the first test (on the left) I used the default parameters. The print proceeded normally until it was attempting to print the bridge. The filament closest to the corners (I am using a diagonal fill), seemed to bridge the gap well, however the bridge was inconsistent in quality (very thin in appearance). The center diagonal did not show up at all. The bridge started appearing better after about 3 layers, however as you can see in the picture holes appeared in the center.

For the second test (in the middle), I followed the Calibration instructions and increased the Bridge Feed Rate multiplier ratio to 1.1. This however, as you can see resulted in a worse quality print.

For the final test (on the right), I added an extra fill layer to allow for the bridge to better form, but it resulted in the same quality as the first test.

When I was previously using .4mm layer heights, the bridging problems were not as bad; however the print quality was noticeably worse.

Now my question is, what are the settings that affect bridging the most?

I plan on experimenting with raising extrusion temperature and increasing the feed rate over the bridge to about 1.6 to see how things turn out.

Would it be useful to add external cooling?

Any help is appreciated.

Edited 1 time(s). Last edit at 02/22/2013 06:55PM by aznoohwee.

Try feeding less plastic during the bridging. I had the same problem with this particular test, with results that looked much the same. Part of my problem was my filament was of poor quality and changed a lot in thickness, and I was feeding a thicker diameter plastic than what I had originally measured. I found that I got much better bridging results when I 'stretched' the plastic over the bridge. Letting it flow at the normal feed rate caused sagging, beading, and broken bridges.

Question : I too am experimenting with the same hollow cube test piece.
im using Sic3r. is the first bridge layer considered the bridge and the following layers just normal fill ??

seems to be a fine adjustment between flow and speed. can anyone share their settings so i can use them as a starting point
Thanks