I agree. I think you can get it broken further. Have you ever tried bead beating?

I am thinking just using 0.1mm glass beads and a vortex to shear it. Or if you have a high powered bead beater like a FastPrep, you could shear the DNA down pretty quickly.

What about using 4 base cutting restriction enzymes?

The sonicator is probably the lease expensive way but the bead beating is easier for multiple samples.

My cursory google search for ChIP and bead beating turns up a couple of protocols that use it for ChIP in organisms with cell walls.  It's used as a mechanism to lyse the cells, not to shear the DNA.  All of the protocols I found have a sonication or restriction digest step for DNA shearing.

 Good luck with your project Brian. ChIP was the bane of my lab life! It worked great for one transcription, but not at all for the other two. However, all my problems were related to protein stability, so as long as you can do regular ChIP ok ChIP-seq should work for you. Regards the size, I think you just have to play with your protocol until you get it right (just like you are doing).

To shear our DNA we used a nebulization method. Worked like a charm if you can get your hands on a nebulization cup and a tank of nitrogen. The protocol I link to says 700-1300bp in size, but you can get it lower by altering viscosity and neb times.

http://cshprotocols.cshlp.org/cgi/content/abstract/2006/4/pdb.prot4539

The "device" itself is very simple. A nitrogen tank with tubing and a nebulization cup. The cups can usually be purchased from any medical store supplier. The nebulizer cups we used to use were from IPI Medical Products, based out of Chicago (part #4101). It needs to be modified slightly, but the protocol can be found here (http://www.genome.ou.edu/protocol_book/protocol_partII.html). I used one cup for my entire grad school career ... cost me around $7 (sans nitrogen tank and nalgene tubing).