Arizonavideo, I am not sure how to go about measuring torque output and am operating on blind faith alone that the 2.5K resistors are limiting the current to 2.5 amps. I plan to order a 48vdc power supply today, and with connect it as soon as I receive it. I will report the results once that is done. If I am interpreting correctly what is being accomplished by using the center taps and one coil end only, is that only HALF the coil length is being energized. This did result in an amazing increase in speed, with fewer lost steps. Before trying this, the motors could not be relied upon to turn the screws for even a minute without stalling. The actual rating of these motors is 2a per coil. (So, I am pushing these about .5a above their rated current). I played last night until about 3:00am with the OPEN test file included with Vectric's V Carve Pro software, trying to carve a pocket. About halfway through the process, I lost steps, (and of course ruined the part). Thus my decision to go ahead and try your suggestion of the higher voltage supply. Still monitoring the temperature of the driver, and motors. Still only slightly warm to the touch. (Like a kid, I could'nt wait to play, so I still haven't mounted the Gecko to the heat sink). I currently have Mach 3 setup with the default 2500Mhtz. What changes will I need to make, to set the motors to 900 PPS? (My processor is a Pentium IV running at 1.8 Ghtz). The parallel port is set to EPP mode thru the BIOS (per Gecko's instructions). Are you referring to the pulse rate in Mach 3's Motor Tuning screen? What SHOULD the "inductive ring" sound like? Should the motors be free wheeling, or under load when this test is performed? The pocket cut I mentioned above was set up with a feed rate of 20 IPM. However, I set the Manual Overide to 60% from the MDI screen of Mach before starting the code running. (For an effective feed rate of about 12 IPM). And even at this snail's pace, it managed to lose steps halfway through the code. I do not have an automatic tool changer on my machine, so I stripped out all macros which were generated by the Postprocessor of V Carve Pro. I also do not yet have a touch sensor for setting the tool height, so I manually jogged the tool to the work, recorded this measurement with a back up device (pencil), and manually coded the Z movements to accommodate for the cut depths, and rapids (using notepad's search and replace function). This yielded EXCELLENT results up until the machine lost steps. I am very hopeful that the higher voltage supply will cure this problem. Thanks for all your help, I will be waiting for further instructions on the Gecko adjustments.