Peter,
This is a bit generic, but should get you started in the right direction.
There was a really good posting about PHD on the Y! Stark group the other day (May 20). It explained lots of the brain settings. The post you want is
Neil Heacock's reply in the "Help Guiding is crazy !" thread.
Try to use a guide star which works with exposures of around 1 sec. Long exposures lead to any tracking errors causing elliptic guide stars.
The setting I'd start with would be the Calibration step (ms). You want to get about 10-15 steps in each direction for PHD to get the desired pixel movement. Too small a number takes too many steps. Too few loses accuracy. For the C14 something of the order of 200 will get you in the ball park. It is also affected somewhat by the pixel size of the guide cam. Bigger pixels probably need more millisecs.
If you have a 1.12 version with "Star Mass" make sure it is at least 1.12.4. Until you are really comfortable with PHD turn that off by setting it to 1.0.
A lot of the other settings relate to how your mount reacts.