Isolator between VFD and Motor

What does VFD manual says? Or may be NEC?

As has been stated, you must not isolate the motor while under load if a VFD is in use.

I don't know if the rules have changed recently, but it was acceptable for the safety isolator to operate in the control circuit of the motor starter.

Therefore a simple SPST switch wired in series with the stop button would be sufficient to stop the VFD and so provide safe isolation of the motor for maintenance purposes, and running a couple of extra control wires can often be more economical than the cost of a complicated (and often very large) motor isolator.

The switch must be lockable if not adjacent to the motor.

JRAEF would be in a better position to advise if this is now aceptable, and local rules will obviously have to be observed.

You wouldn't get away with using "control" as a method of isolation in the EU countries. The Machine Control Directive 2006/42/EC will have it covered somewhere.

An auxiliary early break / late make as JR suggested is the accepted "norm"

Depending on the drive and HP (and local codes) - you may be interested in "Safe Torque OFF" (or similar wording) that may be able to provide machine operators with the required safety levels at a fraction of the cost.

Having said that See JR post - excellent.

(I am a traditionalist (obsolete stick in the mud?) from back in the 70's - I like positive lockouts that PLC's and people poking at relays can't bypass.)

<...advisable...>? In some countries' electrical codes, it is a requirement. Therefore the <...Merits and Demerits...> issue doesn't arise.

If in doubt, consult a qualified Electrician.

"In some countries' electrical codes, it is a requirement."

If memory serves, VFD's are designed to operate as 'current sources, ' not 'voltage sources.

Unless the rules o0f electricity have changed, motors are considered Inductive loads.

Putting a make/break switch between a constant current source and an inductive load sounds like a Bad Thing? How bad? "Try to imagine all life as you know it stopping instantaneously and every molecule in your body exploding at the speed of light⁽¹⁾." Okay, not crossing-the streams bad, but it still sounds like it'll end up killing someone and/or causing expensive property damage.

Notes:

1. Thanks to and in memory of Harold Ramis. Egon but not forgotten.
   

Both types exist, current source, and PWM "voltage" waveforms.

I believe the majority are PWM, and can run with V/Hz or in vector, torque control modes.

JR can elaborate.

Thank you for the information. Since I don't build or design VFDs, the ones we have at work I tent to think of in 'black box' terms: "This is the speed controller for the motor, these are the pots you adjust for minimum/maximum speeds, these are the pots for ramp-up/ramp-down durations, these are the run/stop signals. The power feed connects here, and the motor connects there. How does it function internally? I don't need to know that today, I just need to make the speed of THIS belt motor adjustable, and this VFD is described as the right type for that class of motor."

Yeah, I tend to start falling into that class also.

At one point I had all the chip numbers, device numbers etc at my finger tips - but like computers (at one point I made some of my own pc boards for the Z80) I think of them more as a tool now - until one needs fixing, and even then it is usually not cost effective.

Yeah, I remember a project I was designing back when I was working on my Associate Degree; a 'memory / I/O expander' for 8 bit microprocessors. Using 4 address spaces, I created a 'gateway' to a block of 64K addresses. It was pretty clever for its time; read/write slots for the 'internal address, one read/write slot for the data, and one 'half read/half write' slot for control and status. Thanks to USB sticks, it's now completely useless, but it helped me better understand digital circuits, and even learn a few things about 'waveform shaping' using TTL components.

"My draftsman looks blank when I tell him the pad on the resistor need to be a half inch square ---- then you explain the resistor is 7 watts, and solder melts at 188c, and the resistor can run at 220c - the pad keeps the lead temperature down on the board. etc etc"

You bring up a good point there, and something that is very relavent to electronics design, specifically in the Console Games market.

Sony's had some heat-related issues with their consoles in the past: the PS2 needed to be vertical to dissipate heat properly, the PS3 needed to stay horizontal or the SMT components could slip when the system was running at 'full power,' I haven't heard much on which is the 'right' way to set up the PS4, however, I think that unit is more tolerant of positions because they're not using so many custom, designed-in-house chips what the designers were underestimating the heat dissipation on.

Notice that we never had so much 'component slippage' when we were using through-board mounting. and the 8085/8088 processors are still the only ones certified as 'space radiation tolerant' and listed as suitable for spacecraft that will be carrying a human crew. Sometimes the 'old ways,' while not the fastest/flashiest, are the most 'durable' in the long run.

The ideal thing to do is to configure it so that the emergency stop button cable runs through it, and the contact for the emergency stop circuit opens before the phase contacts open.

Pretty much what I would think:

Stop button signals VFD to stop.

E-stop de-energizes VFD as quickly as possible without causing a secondary hazard. (a 'safe' VDF would be designed so that if it lost supply power, it would 'bleed out' any residual reductive/capacitive charge trough an internal resistor load instead of 'dumping' the power through an external contact. The system should stay safe even if some doofus trips over the cord and yanks it out of the wall socket.)

Nobody I am ware of still sells a Low Voltage (690 or less) Current Source inverter, they are all Voltage Source now.

In Medium Voltage, Rockwell still sells a Current Source inverter and if you find the right person, GE/Converteam can. But other than those, I believe the rest are all VSI now too. But MV is a whole different ball game when it comes to circuit safety.

The damage risk from opening of a switch downstream of a running VFD comes from the fact that in an inductive load supplied by a PWM VSI the inductive kickback potential in the motor can act to make the arc persist longer than it should. As the arc expands when the contacts separate, the PWM DC pulses begin to make the gap act like a capacitive circuit, INCREASING the voltage rapidly. That DV/Dt (Delta in Voltage over Delta in time) can cause transistors to self commutate, meaning they will turn on when not being told to turn on, and that leads to extreme stress on the transistor substrate, followed by (potentially) explosive failure. It might not happen the first time, or the second, or the 73rd, but sooner than later it happens and there is no way to predict when, other than to say there is a POTENTIAL for it to happen the very first time. SOME drives have additional protection built in to decrease the likelihood of a first time failure, but that is not universal and it adds cost, meaning the more "competitively priced" the VFD is, the less likely it is to have these extra steps taken.