VFD Disconnects

[mkgrady]

I have a job where I am feeding four VFD motors. I submitted the motor disconnects and they were approved as noted. The note was to confirm that they are provided with auxiliary contacts for the VFD's. In doing some checking around it appears the auxiliary contacts are there to signal the VFD controller to get ready for the motors to be disconnected. I assume the auxiliary contacts open just before the power wires going to the motor. I think this prevents damage to the motor or the controller.

There is nothing in the contract that requires any kind of control wires from the VFD controller to the motor. It just requires the three power wires and an EGC. I don't see how auxiliary contacts are going to be of any use without a pair of control wires between the disconnect and the VFD controller. Do I have this right?

It looks like I will have to spend about $150 for each of the auxiliary contact kits. I'm installing GE 60 amp 3R disconnects. The contact kits cost more than the switch. None of this appears to be required by the contract. I'm assuming there is nothing in the code that would require this.

 

[iwire]

Nothing in the NEC requires this.

However you can blow up the drive by opening the disconnect while it is in the run mode.

You guessed correctly, that AUX switch would be used to signal the VFD to shutdown ahead of the disconnect blades opening.

Now that said it is not odd for us to provide this type of switch per contract for other contractors to wire.

Seems like a worthwhile RFI.

 

[__dan]

I'm sure you will get another response on this board.

The early opening contacts in the disco go in the drive operating control circuit to turn the drive output off. So yes, there's another pair of wires required.

It is especially or absolutely necessary when the disco is opening the drive output circuit, which can damage the drive when it happens under load.

When the disco is opening/closing the drive line side feed, there is an inrush to charge the caps that it's possible the early opening contacts and drive output off signal will do nothing for. That I don't know, it's a question for the manufacturer. repeatedly opening and closing power to the drive, as in a safety circuit, can shorten the life of the caps. I don't know what method may mitigate this. Early opening contacts in the disco are to protect the drives output circuit.

 

[mkgrady]

Nothing in the NEC requires this.

However you can blow up the drive by opening the disconnect while it is in the run mode.

You guessed correctly, that AUX switch would be used to signal the VFD to shutdown ahead of the disconnect blades opening.

Now that said it is not odd for us to provide this type of switch per contract for other contractors to wire.

Seems like a worthwhile RFI.

Thanks. I will send one.

 

[petersonra]

is this really a problem with modern VFDs?

in any case, one should not be opening the disconnect on a running motor.

 

[kwired]

is this really a problem with modern VFDs?

in any case, one should not be opening the disconnect on a running motor.

My thoughts as well. Opening the circuit isn't going to be a problem for the motor or the drive, closing the switch while drive has active output can be a problem, but most would still trip on overcurrent or other conditions without causing any damage.

If the drive is programmed to ramp the motor to a stop, you will never have enough delay between the time the aux switch gives a stop command and the time the main power contacts open the motor circuit. You however will never have a run command if the aux switch is open so you will never close the motor into an active drive output.

 

[jim dungar]

is this really a problem with modern VFDs?

The shut down time of a VFD is usually extremely longer than the opening time difference between aux contacts and main power contacts. For all intents standard contacts open at the same time, so the drive output will be open circuited while it is still ramping down.

I don't know of any 'safety switch' style disconnect with early breaker aux contacts, however I think they are available on the IEC style motor disconnects.

 

[Besoeker]

The shut down time of a VFD is usually extremely longer than the opening time difference between aux contacts and main power contacts. For all intents standard contacts open at the same time, so the drive output will be open circuited while it is still ramping down.

I don't know of any 'safety switch' style disconnect with early breaker aux contacts, however I think they are available on the IEC style motor disconnects.

In our case, I mean the company I worked for, we had contactor on the VFD output terminals. This would driven my a pilot relay - normally they would be too large to be driven directly anyway - and the electronic shut down would be initiated by aux contacts on the pilot relay. That way you had the inherrent delay of two coils being de-energised plus the inertia of the two armatures.

Not all that long maybe, but a VSD doesn't need very long for the VSD power circuit to be shut off. It is solid state after all. There is no need to wait for it to ramp down.

The other thing that was routinely done was to fit a "load make fault break" manually operated isolator at the driven motor. This also had early break contacts back to the VSD to allow the electronics to shut down so that the isolator main poles wouldn't break current. But the isolator was there mainly for safety reasons. It could be padlocked in the OFF position so that anyone working at or near the motor had line of sight of a positive disconnect.

 

[don_resqcapt19]

The shut down time of a VFD is usually extremely longer than the opening time difference between aux contacts and main power contacts. For all intents standard contacts open at the same time, so the drive output will be open circuited while it is still ramping down.

I don't know of any 'safety switch' style disconnect with early breaker aux contacts, however I think they are available on the IEC style motor disconnects.

A correctly wired and programmed AUX switch at the motor disconnect will put the drive in coast to stop mode, so there is no ramping down.
However, as you say, I am not sure there is enough difference in time between the opening of the AUX contact and the power contacts to make any difference.

As far as I know, early break AUX contact kits are available from all of the safety switch manufacturers.

 

[jim dungar]

As far as I know, early break AUX contact kits are available from all of the safety switch manufacturers.

I likely remembered wrong.:slaphead:

 

[Jraef]

What you are SUPPOSED to do is to wire the aux contacts to one of the drive's inputs programmed as a "base block" command*. This IMMEDIATELY kills the base emitters of all of the transistors at once, bypassing all microprocessor functions so that there is no delay in turning the transistors off. The issue at hand is that the opening of ANY device between the drive and motor draws an arc, and that arc is essentially a large capacitive circuit in which the PWM DC pulses are increasing the voltage well beyond the design of the transistors. The higher voltage and higher dV/dt (rate of change in voltage) of that arc will likely eventually damage the transistors. Yes, "modern" IIGBTs are now coming with beefier flyback diodes built-in to help prevent this damage, but nothing is 100%. I've had people with contactors on the output say their drives have been that way for years with no issues, I've seen some that blow the transistors the first time they do it. There are a lot of variables involved in just how much potential damage will exist, such as the size, carrier frequency, cable capacitance etc. etc., leading to the recommendation to users that they not tempt fate.

But more dangerous (to the drive) than that is the possibility that someone CLOSES that disconnect while the drive is still in a run state. The motor inrush current can immediately damage the drive. The aux. contacts to the base block also prevents that from happening.

All disconnects designed per NEMA KS-1 (knife switch) offer handle operated aux contacts, which are required to open ahead of the main contacts. The cheap little UL508 rated rotary IEC disconnects people like to use may or may not. You have to check carefully as often times they offer both, so you have to make sure to pick the right ones.

*"Base block" is also described by manufacturers in many different ways, such as an "Enable" circuit or now more recently, a "Safe Torque Off" or STO input. STO is the same thing, just certified as a "Safety Circuit".

Did the project specs call for VFD cable? The reason I ask is that there are numerous VFD cable suppliers that include a separate shielded pair of control conductors in the bundle for this purpose (or for separately powered motor brakes). I've seen it where the original specs called for a specific VFD cable using that feature, then someone opts for using steel conduit instead (valid replacement), but fails to notice that the VFD cable had those extra wires.

 

[mkgrady]

What you are SUPPOSED to do is to wire the aux contacts to one of the drive's inputs programmed as a "base block" command*. This IMMEDIATELY kills the base emitters of all of the transistors at once, bypassing all microprocessor functions so that there is no delay in turning the transistors off. The issue at hand is that the opening of ANY device between the drive and motor draws an arc, and that arc is essentially a large capacitive circuit in which the PWM DC pulses are increasing the voltage well beyond the design of the transistors. The higher voltage and higher dV/dt (rate of change in voltage) of that arc will likely eventually damage the transistors. Yes, "modern" IIGBTs are now coming with beefier flyback diodes built-in to help prevent this damage, but nothing is 100%. I've had people with contactors on the output say their drives have been that way for years with no issues, I've seen some that blow the transistors the first time they do it. There are a lot of variables involved in just how much potential damage will exist, such as the size, carrier frequency, cable capacitance etc. etc., leading to the recommendation to users that they not tempt fate.

But more dangerous (to the drive) than that is the possibility that someone CLOSES that disconnect while the drive is still in a run state. The motor inrush current can immediately damage the drive. The aux. contacts to the base block also prevents that from happening.

All disconnects designed per NEMA KS-1 (knife switch) offer handle operated aux contacts, which are required to open ahead of the main contacts. The cheap little UL508 rated rotary IEC disconnects people like to use may or may not. You have to check carefully as often times they offer both, so you have to make sure to pick the right ones.

*"Base block" is also described by manufacturers in many different ways, such as an "Enable" circuit or now more recently, a "Safe Torque Off" or STO input. STO is the same thing, just certified as a "Safety Circuit".

Did the project specs call for VFD cable? The reason I ask is that there are numerous VFD cable suppliers that include a separate shielded pair of control conductors in the bundle for this purpose (or for separately powered motor brakes). I've seen it where the original specs called for a specific VFD cable using that feature, then someone opts for using steel conduit instead (valid replacement), but fails to notice that the VFD cable had those extra wires.

The only wires run from the controller to the disconnect are three #10 plus a #10 ground. All thhn/thwn run in EMT. Same wires from disc to motor but in liquid tight.

 

[Jraef]

Well, if they weren't on the contract drawings but they want them now, you should get a change order. You should use shielded signal wire, twisted pair, ground the shield drain only at the VFD end. There is no energy to speak of on those wires.

 

[mkgrady]

Well, if they weren't on the contract drawings but they want them now, you should get a change order. You should use shielded signal wire, twisted pair, ground the shield drain only at the VFD end. There is no energy to speak of on those wires.

Can that signal wire go in the same conduit as the power wires? The system voltage to the vfd controller is 480

 

[Jraef]

Can that signal wire go in the same conduit as the power wires? The system voltage to the vfd controller is 480

If you use good quality shielded cable, yes.

 

[don_resqcapt19]

If you use good quality shielded cable, yes.

So the control circuit is an Article 725 Class 1 circuit? I kind of expected, based on available current and voltage of the drive control circuit, that it would be an Article 725 Class 2 circuit.

 

[Jraef]

So the control circuit is an Article 725 Class 1 circuit? I kind of expected, based on available current and voltage of the drive control circuit, that it would be an Article 725 Class 2 circuit.

Actually, that's a good point. I've done it, but the ones I did that on were 120V controls. Now that you bring that up, it will depend on what the drive has. One manual I just looked at says it is in fact Class 2 when using 24VDC, so would need to be barriered, meaning not in the same conduit. If the VFD has 120V control inputs they can technically be in the same conduit, although in that same document it says not to do that, it says to separate them by 3 ft.

Might be better to say it's of those "check with the supplier" kind of issues.

 

[GoldDigger]

...
... If the VFD has 120V control inputs they can technically be in the same conduit, although in that same document it says not to do that, it says to separate them by 3 ft.

I expect that would be mitigated by the combination of shielding and twisted pair(s) for the control wires. But no guarantee.

 

[mkgrady]

Manufacturer Answers and New Developement

Manufacturer Answers and New Developement

After sending the rfi, the hvac contractor purchasing the drives and motors has contacted the drive manufacturer Allen-Bradley to try to resolve the auxiliary contacts issue. They say the contacts in the switch are not necessary due to new designs and the disconnects are not needed because the controllers can be locked out. Really? More on this below.

As a seperate issue I have been struggling to find a route for the four runs of conduit that run from the controllers to the disconnects and motors. They are on the roof and the drives are about 24 feet below. We are replacing four old existing motors that are straight 480 three phase line voltage. ALL twelve of the existing conductors are in one conduit from the electric room to the roof. I'm told the manufacturer said that we could run all of the conductors from the drives to the motors in the same conduit. I'm no vfd expert but this surprises me. Does this sound right?

If we go by the manufactures suggestions we get rid of the motor disconnects, the auxiliary contacts (not in contract anyway), the four runs of conduit to the roof and maybe even the wire if it's the right size. I'm meeting the hvac contractor and the owner tomorrow. Any other comments are welcome and thanks to all for responding.

 

[iwire]

All I will say is lockouts are not acceptable unless the installation meets one of these conditions from 430.102(B).

(a) Where such a location of the disconnecting means for the motor is impracticable or introduces additional or increased hazards to persons or property

(b) In industrial installations, with written safety procedures, where conditions of maintenance and supervision ensure that only qualified persons service the equipment

Keep in mind 'impracticable' basically means impossible, it does not mean difficult.