Electrician Talk banner

1 - 19 of 19 Posts

Registered
Instrumentation and Control Technician
Joined
2 Posts
Discussion Starter · #1 ·
I have worked with people on installing and setting up VFDs, and I have had multiple people say "never run a VFD without a load connected to it or you will fry it". I had no reason to doubt them so that's what I went with, until now. I tried googling it to see why exactly it is bad, but I cannot find a single article saying it is bad to run a VFD without a load.

Is this just a common misconception about VFDs?
 

Band Member
DIYer Extrodinaire
Joined
7,474 Posts
It won't fry it, but depending on how the drive is programmed/set up, it may trip on loss of feedback.

Ran them without load (back in the day) without issue ever. Sometimes we had to for first level commissioning .
 
  • Like
Reactions: John Valdes

Registered
Scada Supervisor
Joined
3,134 Posts
Depends on the drive. I am going to power and run one today with no motor
 

Super Moderator
Retired
Joined
17,172 Posts
Its not true. I have run drives dry. No motor connected to test and set up parameters. Like Wayne said though, should there be a feedback device it very well may fault on loss of the feedback signal.
In this case just program the feedback out.
 

Registered
Scada Supervisor
Joined
3,134 Posts
This is from one of our resident drive experts @JRaef

May 17, 2017 (Edited)

In what's called "V/Hz mode", a VFD doesn't technically even know if the motor is connected or not, other than it will be monitoring current for the purposes of OL protection, so it will think "Hey, no OL going on here! Proceed as commanded.".

But now with Sensorless Vector Control, the VFD is looking for a current feedback from the motor, it uses that to determine the proper vector calcs. So no current, nothing to calculate from and the drive will trip on "Load Loss". On the PF700s, V/Hz mode was a true V/Hz mode, no feedback at all, so as long as you are using it in V/Hz, it's going to be fine with not having a load connected. If you are using it in SVC mode or Flux Vector Control mode (i.e. encoder feedback), it will not run; no more than a few seconds in SVC, about 1 second in FVC before it trips. Newer generations of VFDs, like the PF750, still use the current feedback loop for better motor performance even if in V/Hz mode, so they will trip off too now.

Is it weird how saying sentences backwards creates backwards sentences saying how weird it is?
 
  • Like
Reactions: emtnut

Registered
Instrumentation and Control Technician
Joined
2 Posts
Discussion Starter · #7 ·
Thanks Everybody all makes sense now, I believe I was just misunderstanding what I was being told. I believe what they meant was that if you run it with no load and then flip the disconnect to the motor while the VFD is running, you are going to damage things.
 

Registered
Scada Supervisor
Joined
3,134 Posts
Thanks Everybody all makes sense now, I believe I was just misunderstanding what I was being told. I believe what they meant was that if you run it with no load and then flip the disconnect to the motor while the VFD is running, you are going to damage things.
Yes drives don't like it both ways flipping it on or off.
 

Tool Fetish
Joined
813 Posts
The Schneider drives that we have installed come with a run/test switch that just shuts off the output/isolation contactor so you can program or test the drive without the load connected
 

Registered
industrial E,I&C
Joined
3,339 Posts
Thanks Everybody all makes sense now, I believe I was just misunderstanding what I was being told. I believe what they meant was that if you run it with no load and then flip the disconnect to the motor while the VFD is running, you are going to damage things.
99.9% of the time its a bad idea and you may damage the drive. (normally shorten the life expectancy rather than nuke the drive)

The 0.1 is due to engineered systems where multiply motors are running off one drive.
 

Bilge Rat
motors and controls.........
Joined
7,948 Posts
Some of the larger VFDs I've installed have a parameter that can be set for 'small motor' so you can connect a 1/2 HP motor to a 200 HP VFD and it won't trip out on load loss.
 

Bilge Rat
motors and controls.........
Joined
7,948 Posts
99.9% of the time its a bad idea and you may damage the drive. (normally shorten the life expectancy rather than nuke the drive)

The 0.1 is due to engineered systems where multiply motors are running off one drive.
It's even worse to close the disconnect with the VFD running no-load and the motor stopped.

The magnetizing current is pretty rough on the output transistors.
 

Registered
industrial E,I&C
Joined
3,339 Posts
It's even worse to close the disconnect with the VFD running no-load and the motor stopped.

The magnetizing current is pretty rough on the output transistors.
We did that for over 25 years. Up to 10 motors per vfd all connected to starters. Vfd ran at a set speed and we simply closed the starters. Danfoss originals were replaced when they could no longer get parts so they went to AB powerflex (approved buy the factory) and i think they only had one out of 6 drives fail in about 10 years.

The point is that you may come across odd stuff in the field that was engineered to do things that we have been told can not be done.
 

Chief Flunky
Field Service Engineer
Joined
2,078 Posts
It's even worse to close the disconnect with the VFD running no-load and the motor stopped.

The magnetizing current is pretty rough on the output transistors.
Ok I鈥檒l bite. Input side you鈥檝e got to deal with the precharge circuit and some of those are not very robust. Seems like the larger the drive the poorer the precharge.

Output side opening can be an inductive kickback issue. But closing? The VFD itself is already current limiting by nature. If fluxing up was an issue it would be just as big of an issue if it were starting up normally as it is if a contactor closes.
 

Bilge Rat
motors and controls.........
Joined
7,948 Posts
Ok I鈥檒l bite. Input side you鈥檝e got to deal with the precharge circuit and some of those are not very robust. Seems like the larger the drive the poorer the precharge.

Output side opening can be an inductive kickback issue. But closing? The VFD itself is already current limiting by nature. If fluxing up was an issue it would be just as big of an issue if it were starting up normally as it is if a contactor closes.
There are basically 3 currents to a motor;

1) When it's at a standstill, the first current is the magnetizing current. This is pretty much a bolted fault for the first 1/60th of a second or so.

2) Locked-rotor current, this is the current after counter EMF has been established. Usually 6X full-load current but higher with energy-efficient motors.

3) Full-load current. Self-explanatory.

When a motor is closed in to an operating VFD, you have a DC capacitors then the output transistors then the impedance of the wire from the VFD to the motor then the impedance of the motor itself. Basically, the capacitors discharge into the motor via the transistors, some are designed to handle this current, some are not.

Closing a switch in between a stopped motor and an operating VFD is a lot like closing a switch in to the line side of a VFD with no pre-charge circuit.

So far, I've seen maybe 2 or 3 VFDs wrecked by opening a switch while the motor was running and about a half-sozen by closing the switch between a running VFD and a stopped motor.
 

Chief Flunky
Field Service Engineer
Joined
2,078 Posts
There are basically 3 currents to a motor;

1) When it's at a standstill, the first current is the magnetizing current. This is pretty much a bolted fault for the first 1/60th of a second or so.

2) Locked-rotor current, this is the current after counter EMF has been established. Usually 6X full-load current but higher with energy-efficient motors.

3) Full-load current. Self-explanatory.

When a motor is closed in to an operating VFD, you have a DC capacitors then the output transistors then the impedance of the wire from the VFD to the motor then the impedance of the motor itself. Basically, the capacitors discharge into the motor via the transistors, some are designed to handle this current, some are not.

Closing a switch in between a stopped motor and an operating VFD is a lot like closing a switch in to the line side of a VFD with no pre-charge circuit.

So far, I've seen maybe 2 or 3 VFDs wrecked by opening a switch while the motor was running and about a half-sozen by closing the switch between a running VFD and a stopped motor.
As to your three currents magnetizing current varies but is limited to about 3 times the locked rotor current. LRC can be quite high but the most I ever realistically see is about 10x FLA. The highest magnetizing current I鈥檝e seen is a 300 HP Toshiba that hit around 23x FLA. now these are clear in 鈥渟hort circuit鈥 range but it lasts for 1-2 cycles..

But let鈥檚 be clear here. Unless you current limit it, magnetizing inrush as well as LRC are independent of the load. As you stated the magnetizing inrush occurs when the motor first fluxes up. But you cannot avoid it. Magnetizing inrush is going to happen if you energize the motor. You can鈥檛 avoid it which is my point.

I鈥檝e worked with dozens of drive models and designs and although I鈥檝e seen some differences in precharge circuits as well as the occasional output contactor but NEVER any circuit topology specifically to limit magnetizing inrush beyond the natural tendency of the circuit itself.
 

Bilge Rat
motors and controls.........
Joined
7,948 Posts
As to your three currents magnetizing current varies but is limited to about 3 times the locked rotor current. LRC can be quite high but the most I ever realistically see is about 10x FLA. The highest magnetizing current I鈥檝e seen is a 300 HP Toshiba that hit around 23x FLA. now these are clear in 鈥渟hort circuit鈥 range but it lasts for 1-2 cycles..

But let鈥檚 be clear here. Unless you current limit it, magnetizing inrush as well as LRC are independent of the load. As you stated the magnetizing inrush occurs when the motor first fluxes up. But you cannot avoid it. Magnetizing inrush is going to happen if you energize the motor. You can鈥檛 avoid it which is my point.

I鈥檝e worked with dozens of drive models and designs and although I鈥檝e seen some differences in precharge circuits as well as the occasional output contactor but NEVER any circuit topology specifically to limit magnetizing inrush beyond the natural tendency of the circuit itself.
I guess I don't type very well......lol.

What I meant was that a VFD starting a motor from a standstill can handle the magnetizing current much better than being slammed into magnetizing current after it's up to speed.
 

Chief Flunky
Field Service Engineer
Joined
2,078 Posts
I guess I don't type very well......lol.

What I meant was that a VFD starting a motor from a standstill can handle the magnetizing current much better than being slammed into magnetizing current after it's up to speed.
Again why? Nothing in the software algorithm or hardware would be any different except it might trip on overcurrent.
 

Registered
Joined
5,482 Posts
Magnetizing current can rise way too fast for a microprocessor to see and react to it. Some drives have a dedicated IC on the output transistors to detect the dI/dt and block the base emitters fast enough to prevent damage to the transistors, some do not and even with those that do, it's not 100% effective 100% of the time, especially if they are forced to act repeatedly (incremental damage can occur). So not knowing if the drive has that IC or if it has been damaged by doing its job too much already, it's just a good practice to NOT close a switch into a motor if the VFD is already modulating (in a Run state).
 
1 - 19 of 19 Posts
Top