By "drifting" I assume (yes I know the 'a' word) the spindle speed is changing while running. What type of feedback is being used for the spindle and what drive also? Could be a speed feedback problem or a mis-adjustment of the P & I pots (or program variables). Some more info on the drive, etc. would be helpful. If I read this wrong and its creeping at standstill it obviously does not have an isolation contactor in the output, and could be indicative of a SCR in the output bridge shorting.
2 of them have the same problem. The way I understand it is, when the spindle is in stop position, the spindle "creeps" back and forth. The tolerances of the machine are not holding true.
Hardinge says it is power problem, or noise in system. Engineer at plant doesn't think so. I am wondering if it is a ground issue, or noise on the ground. But I would be lost on how to find out if this is true. Hardinge wants ground rods for the machines, but instead we ran insulated grounds back to the panels. Ground rods provide a ground loop, do they not?
Any ideas. Is there noise limited equipment out there? Some type of filters to put on the electrical system?
I would be insisting the machine-tool manufacturer reasonably demonstrate to you and your client that power quality is unambiguously the cause. There was lot of fingerpointing up to about 15 years ago with major machine-tool producers, and most fixed their designs to work properly on the average industrial power system. Get the machine vendor to find the exact problem, at their expense if they are so sure of themselves, unless prior to purchase they provided a clear disclaimer that any problems were not theirs to deal with,
There are special-purpose, high-harmonic drive-isolation transformers rated 230∆—230Y/133, that would allow establishing a separately-derived system with local system grounding for each machine, but are pricey with no guarantee as to effectiveness. Smaller sizes have ±5% taps for minor voltage adjustment—~219V on primary. would give ~230V on the secondary. www.acmepowerdist.com/pdf/Page%2046.pdf and www.acmepowerdist.com/pdf/Page%2048.pdf
Not being familiar with your equipment I'll just throw out some ideas, in the past on my CNC's I've had a D/A output card develop an unstable zero, more of a problem on our carriages and tool slides, as our spindles had (have) either a 3 phase input contactor or a DC output contactor to isolate the spindle when stopped. What type of command signal do you have? (0-10V, serial, etc) The D/A card problem I referred to were all +/- 10VDC outputs. I would probably be worthwhile to scope the command inputs if they are analog to look for irregularities.