Hey Frank,
EMI can have so many causes. It is both emitted and received. The most important thing would be to make sure if you use a spindle, that you have the spindle motor cable shielded and terminated properly, like described here:
…and that your VFD is enclosed in a grounded VFD control cabinet, like described here. The VFD and the spindle cable can be a major source for EMI. That is because in contrast to the mains power from your wall, the current that comes out of a VFD and drives the spindle through the spindle cable is by no means consisting of nice sine waves. Because of the switching nature of the VFD and the use of internal high carrier frequencies, the current for the spindle is very “dirty”, which means, it looks rather like aggressive square waves. But the carbon brush commutators of a hand router like the Makita RT070x emits a lot of EMI, too, both through the cable and mainly over the air because of the sparks, and there is no possibility to shield them, except in special motors for explosive environments. But rather you would use a spindle in the industry, which is an induction motor, which has no carbon-brush commutators at all, and no coils on the rotor, only on the stator. But with spindle/VFD, the EMI comes from the “dirty” switched power they produce. So anyway such machines are heavy sources of EMI, and usually you shield and ground everything the best possible.
On the other hand, EMI is received on any cable and can lead to deterioration of signals, which means it can make errors in the machine operation. Therefore, on a CNC machine, all cables should be shielded and grounded. Obviously this is not the case on the Onefinity, and yet the Onefinity manufacturer did not change that. But I think it would be the most important countermeasure against the known EMI problems. Therefore I provided you with a link in my previous post for replacing the stepper cables by cables that are shielded and grounded. This is the first thing I would try, besides grounding the machine, and using a shielded spindle cable and a grounded VFD cabinet.
Finally you still have the unreliable Amphenol/Molex connectors. They are not made for sockets on the outside of a device, but only for internal use, so they are wrong here. They tolerate neither permanent movement nor to be often re-connected. They are known from the internal ATX power supply in computers, where you usually plug them into the board exactly once and you never touch them again. In the industry, power connectors on CNC machines have gold-plated contacts and reliable strain relief (you could try to retrofit strain relief on the moving X axis to the older Onefinity and replace the curly Z cable, that can break internally), but these Molex/Amphenol connectors only have a thin tin plating and only a small one-per-contact strain relief. Once the tin plating is removed by frequent re-connecting them and the copper of the contacts is exposed to air, the corrosion comes even faster, making bad contacts. But replacing them is not an easy thing, as described here:
That is a good question, but it’s not me who can answer this. You got to ask the Onefinity manufacturer.
That is really understandable!
I would not put the machine in stock configuration into production, since I know what to alter on the machine first. I however knew that when I bought the machine. I plan to make my modifications first. Wood is much too expensive to be wasted.