Well, the new Gecko's are installed and man are they nice!! I cannot believe how smooth the motors both sound and move. There is no perceivable resonance anywhere in the range of speeds from dead slow to 120 inches/minute. Also, the motors seem to have far more torque than with my home brew design. The head is traversing easily at 60 IPM (no counterweight) and the table is humming along happily at 120 IPM. I will have to fiddle with the timing settings in EMC2 however, as now it is reporting real time latency problems when it wasn't before. My drives were running at 20 micro steps, whereas the Geckos are fixed at 10 micro steps. So, my actual step rate has been cut in half, but EMC2 is still complaining.
On a side note, it does crack me up how uptight some folks seem to be with "real time" computing being necessary for CNC operation. Especially in a hobby setting, using Chinese machinery and home-made components, does nanosecond accuracy really matter? Clearly, MACH3, the Windows based competitor (which is not real time) seems to handle accurate machining just fine without all the "real time" voodoo. Sometimes I wonder if geeking out on precision timing is more an exercise in how-geek-one-can-be rather than practicality. Far be it for me, however, to question the efforts of the great folks that wrote EMC as I would have to shell out cash for different software if it weren't available. I realize EMC has to fit many applications and that, perhaps, some of those need that real time precision. On the other hand, I wish there were a "hobby mode" it would run in where it would be a little less sensitive!
With regards to wire routing, I have managed to organize everything into a three-layer design. The signal wires from the computer for step and direction are routed against the mounting panel. They are held there with tie wraps and 3M Command Strips. The power supply wires are rigid, so I've bent those into place so that they are on the same plane as the drives. At the top of the photo they run back toward the panel and are also held in place with tie wraps. That gives approx 1.25" of clearance between the power supply wires and the signal wires in the areas where they overlap. Finally, the feeds to the steppers come out of the drives and towards the camera in the photo. They then exit through the top of the cabinet. This gives about 1" of clearance to the power supply wires and 2.25" to the signal wires.
In addition to the step and direction signal wires, I've also wired the "DISABLE" input on the Geckos to the "Amp Enable" output of EMC (optically isolated, of course). This is nice because the drives do not move until enabled by EMC. In other words, they don't jitter while the computer is booting and they immediately cease operation if there is a software following error or E-STOP is activated. It has also been nice during testing as, if I close out of EMC, the drives are powered down so they aren't consuming power, putting out heat or causing the motors to "sing". Although this doesn't replicate a true "E-STOP" where it removes power from the drives, I believe it is sufficiently safe for my needs.
Overall I couldn't be more pleased with these little buggers. I wish I hadn't screwed around with rolling my own design in the first place, but that is part of the value of these sorts of projects - learning the hard way!!
Here's a quick video of it in action. Sorry, but my digital camera only allows for 30 second movies - I'll upgrade in the future! The program running is the "EMC2 AXIS" logo program that comes with EMC. Unfortunately the time limitation only let's it get through the "EM" but you get the idea. Who would have thought a Sharpie would fit so well in a drill chuck?!