Seems odd that if you already have a EQDIR connection to drive your mount that you would want to introduce another cable? I would also think that of the two, pulse guiding would offer better performance? See the following posted by Chris Shillito on SGL.
The mount is semi-intelligent. It knows how to driver stepper motors but knows very little about astronomy or astronomical calculations. So essentially what EQMOD can tell the mount to do is:
Move a motor a number of steps in a given direction
Set the mount moving a given rate/direction
When EQMOD tells the mount to track it does so using the latter command. The mount, if left alone, will then just keep moving at the specified rate. If the mount has been correctly polar aligned then there is no reason why it should ever wander off - the electronics and serial protocol are all optimally designed for initiating and maintaining sidereal rate (but other rates are less accurate). That said there is a roughly periodic wobble (it isn’t actually repeatable over any sensible time frame) due to the nature of the mechanics.
A guiding pulse simply consists of sending two rate change commands, one to move the mount at the guide rate and one to return to sidereal. The autoguider software specifies the interval between the commands and EQMOD does the rest. EQMOD’s PEC also works on a similar approach with the tracking rate be regularly adjusted to counter the periodic error movement (and any pulse guide movements will be additive to the PEC correction).
ST-4 guiding is exactly the same in principle except the rate changes are commanded via a hardware interface on the mount itself. EQMOD is unaware of ST-4 guiding and so cannot arbitrate between it and PEC as a result ST-4 pulses are likely to trash any currently active PE compensation so the value of PEC+ST-4 may be questionable.
Good luck anyway, hope it all works out.