Talking to my wife today after doing internet searches she sai that every time she would come out when it froze, i was sanding and it was at a slow speed. So after dinner tonight I went out and fired the lathe up at a high speed (1200) Then i would wait for 10 seconds, and lower the speed by 200 rpm. I finally found after several speed lowering and speed ups hat at 200 rpm and LOWER, the TV/Computer signal would die. I tried it several times. The TV would die, i would turn the lathe off, wait until the TV restarted (1 minute or so) then restart the lathe and quickly sped it up. After several times, it was evident tonight at least, at 200 RPM or lower, the TV/computer dies. At 225 and above, its ok.
You didn't say what lathe you have, but I will assume that you are using a Powermatic 3520 with a Delta inverter. At that slow speed, the "synthesized" three-phase power to the motor is probably around 5 Hz or less and is very "trashy" at such slow speeds. Part of the reason for such a trashy signal is that the inverter is trying to compute load torque by sensing motor current. Even under optimal circumstances motor current is a noisy parameter, but things get even worse when this noisy data is used to regulate speed by modifying an already noisy signal with more noise. Perhaps the only reason that it even works as well as it does is probably due to the mechanical inertia of the motor and drive train.
If you are interested, I may do a search for a drawing that shows what the waveform looks like from the inverter to the motor versus normal three phase sinusoidal power.
Back during the Cold War, the Soviets would have killed for such an effective broadband jammer to use against the VOA and BBC ... and now you can buy them for just a few hundred dollars to power your lathe.
In using a line choke, are they sized for voltage, or just diameter of the power cord. Is it he cord from the wall, or the cord from the inverter into the motor?
There are all sorts of permutations of power line filters with varying degree of effectiveness and as you might guess, the cheap ones probably won't satisfy your needs -- an expensive one may not either depending on circumstances.
Sizing the type of three-phase line reactors used with inverters is not a simple task. It is based on line impedance and typically sized at 5% impedance for approximately 35% reduction of total harmonic distortion.
Generally speaking load reactors between the inverter and motor are not used except in industrial environment where the distance separating the two is more than 100 feet, in which case they are needed to reduce what is known as
δv/δt transients which can shorten the life of the motor and the inverter. These transients may also cause radiated interference. For our application where the distance between te inverter and the motor is just a few feet, this is not an issue although using them has been known to give "warm fuzzies" to the lathe owner.
Regarding your question about where to install a line choke, as far as I know, they are not the appropriate devices for suppressing interference from an inverter. However, the line and load reactors used with inverters do a reasonably good job of filtering conducted noise on the power line and fair job of reducing radiated interference. I would first install a line reactor and if necessary a load reactor. Seeing as these are industrial devices, they come with industrial strength price tags and it is possible that they may not satisfactorily solve your problem. From your description of problems that you are having, I get the impression that the interference is radiated rather than conducted. In that case, it may require some serious shielding to minimize the interference. A suitable NEMA equipment enclose for the inverter might be a good idea. The Robust AB lathe has a built-in enclosure for the inverter and while it is not a NEMA style enclosure, it seems to be close enough for all practical purposes. If you buy or make an enclosure, it needs to have adequate ventilation while still maintaining shielding. Beyond that, braided shielding on the leads from the inverter to the motor and possibly also to the external controls may be necessary. This type of shielding is normally grounded only at the source end and left floating at the other end. Hams sometimes use either aluminum foil or metal tape as a cheap alternative to braided shielding.
Even if the interference is conducted over the power line, chokes or filters on the power cord of affected devices seems to me to be marginally effective at best. The reason, as Ralph indicated, is probably because immunity from interference probably was never part of the equipment deign.
BTW, I found an image, while not exactly what I was looking for, illustrates what I was saying about the power to the motor being "trashy".
