Yeah I'd also like to cast the vote that you should forget the dream of using a guide scope on an SCT like the EdgeHD, SE, etc., without having distorted stars (and therefore, signal).
This is not because of the focal length; when I first started guiding, I got a 70mm guide scope to piggy back and it was a phenomenal piece of equipment, and the guide star was more than adequately sampled, even when imaging at prime focus (2032mm). The problem is that the guide scope will fail to account for the biggest sources of error in the system: differential flexure and mirror movements ("drift" and "flop" ) . I guided very well with that setup but never achieved round enough (I'm picky) stars because of these problems, and they change throughout the night because of the angular differences as the mount tracks. In some instances, I had football-shaped stars with excellent guiding. This is of course because the guider is keeping the star centered according to what the guide camera sees while piggybacked on the telescope, however it can't correct for movement of that system with respect to the imaging scope, nor can it correct for the movement of the mirror in the SCT. I was confident there wasn't TOO much differential flexure (it seemed very solid), however to this day I don't know which problem dominated. Maybe they both did equally. Neither was fixable, however. I had to return the lovely guide scope and move right into off-axis guiding. I use 5-minute subs now, and this would be impossible with the old setup. I keep almost all of them (seeing hurts me more than guiding).
Off-axis guiding is challenging. You have a high focal length at the guide camera (few, and dim, guide stars in the field of view, as well as distortion of the stars, being at the edge of the light cone). However it picks up every single problem that the imaging sensor sees and I keep most of my 5-minute subs with razor-sharp, circular stars.
In my opinion, the only proper way to even think about off-axis guiding is to use stellarium (free), and program in your specific optics. As mentioned before me above, you can see the focal length and FOV there, but you can also program in the OAG and use the program to plan exactly what angle and center coordinates you'll need for a specific target to guarantee you have a BRIGHT guide star in the field of view of the guide camera. With extreme brevity, the steps are determine ALL equipment in the optical train that you're using (OAG, spacers, rotator, etc.) and get that finalized, get the pick-off prism as far into the train as you can without casting a shadow in your images (this is hard to get optimized- too far out, and your guide stars are noodles...), find your field of view and get that into stellarium, determine the FOV of the guide camera and program that in, determine the distance between the guider and imaging FOVs and program that in (by placing a planet in the guider, taking an image through the imaging camera, plate solving the latter, and moving to the center coordinates in stellarium- then set the distance such that the guider FOV ends up over the planet in stellarium), and finally calibrate and mark the 0deg angle on the scope using plate solving and stellarium again. Yes, this takes some nights of setup.
However, I can now choose a target, go to stellarium and plan out the angle and composition for a target that gives me a bright guide star in the pick-off prism, and move the rotator to the correct angle, all during the day. At night, it's as simple as polar aligning, then using my sequence software's "center" feature to center to the coordinates determined in stellarium. My target will be centered, and the guider will see the planned, bright guide star immediately. No fuss.
It's a steep learning curve, but truly the only way to guide at high focal length properly, and actually incredibly easy to use once set up. I plan to have a detailed step-by-step for the above process on
my website within a week.