Well, im happy to Say, Art figured the problem out. It had to do with the debounce setting in config/logic .
He told me to set it to 25000 and that fixed the constant e-stop and limit switch requests. I can only assume there is noise interference some where, amd that noise was simulating the signal of limit and estop. The signal was so fast it barely lit the yellow leds in the diag screen. The debounce now requires the limit signal to be constant for a certain amount of time. So this makes the time from when the limit is actually tripped to the time when Mach3 responds a little longer. If you are using push button switches, you may run the risk of crushing the switch. The magnetic switches are unafected becasue there is no physical contact and you can simply relocate the magnet some where closer to stop at a predetermined spot. The magnets can also be stacked to double their effective range. The magnetic switches as well as having no external moving parts are also sealed from liquid penetration so they are good for coolant flood conditions, which is one reason I chose them. Plus, they are $5 each. I bought a dozen of them.
I now have it set up in the following way.
I have 3 radio shack NC magnetic alarm switches wired in series. Each is mounted in the center of its axis.
So there is a magnet on each side of the table and on the top and bottom of the head dove tail.
The switches on the X and Y are stationary while the magnets move back and fourth with the table.
The switch on the Z is mounted to the head and moves with it until it hits the magnets feild.
All pins are set at 15 which for me is the limit pin. So I cant see individual limit requests, but each axis homes and limits off the same switch and performs perfectly.
Ill take a video of it and post it later.