VFD interference

[mhw]

I swapped out a T1H-DM1E PLC rack with a BRX. All of the controlled equipment stayed the same. Several of the AC inputs on the BRX turn on when any of the VFDs are outputting. This was never a problem with the T1K-16NA-1 inputs. I tried sinking resistors with no luck. I have put Roxburgh filters on the VFD inputs and this has fixed the problem.
My question is why is the BRX more susceptible to noise and can this be fixed in the BRX without the expense of the filters.

[BobO]

I swapped out a T1H-DM1E PLC rack with a BRX. All of the controlled equipment stayed the same. Several of the AC inputs on the BRX turn on when any of the VFDs are outputting. This was never a problem with the T1K-16NA-1 inputs. I tried sinking resistors with no luck. I have put Roxburgh filters on the VFD inputs and this has fixed the problem.
My question is why is the BRX more susceptible to noise and can this be fixed in the BRX without the expense of the filters.

Everything is faster in BRX, 3.3v vs 5v parts, etc.

If it is onboard inputs, you can increase the filter time in the system configuration. If it is a module, you should be able to use the DEBOUNCE instruction.

[mhw]

Everything is faster in BRX, 3.3v vs 5v parts, etc.

If it is onboard inputs, you can increase the filter time in the system configuration. If it is a module, you should be able to use the DEBOUNCE instruction.

Thanks for the response but no joy. I tried it all the way to 8000ms.

[BobO]

Everything is faster in BRX, 3.3v vs 5v parts, etc.

If it is onboard inputs, you can increase the filter time in the system configuration. If it is a module, you should be able to use the DEBOUNCE instruction.

Thanks for the response but no joy. I tried it all the way to 8000ms.

You're seeing inputs turn on with an 8 second filter? Are they staying on? Or turning on and off?

[mhw]

You're seeing inputs turn on with an 8 second filter? Are they staying on? Or turning on and off?

The physical input shows on in data view when the VFD is on. The output bit of the DEBOUNCE comes on after 8 seconds and stays on until the VFD goes off.

[BobO]

You're seeing inputs turn on with an 8 second filter? Are they staying on? Or turning on and off?

The physical input shows on in data view when the VFD is on. The output bit of the DEBOUNCE comes on after 8 seconds and stays on until the VFD goes off.

I'm not sure what to make of that. Noise is generally symmetrical, but this would have to be walking to one side with a directional bias. Makes me wonder about grounding.

Now let me qualify that...while I have a degree in electronics, I have written software for the entirety of my professional life, so I may be spouting nonsense.

[mhw]

I'm not sure what to make of that. Noise is generally symmetrical, but this would have to be walking to one side with a directional bias. Makes me wonder about grounding.

Now let me qualify that...while I have a degree in electronics, I have written software for the entirety of my professional life, so I may be spouting nonsense.

That grounding term is something that a good support tech uses when they can't solve the problem. 
I can assure you that everything is bonded to a common plain or plane, I not sure which, and then connected to an earthing rod.
My question is why did this only appear after moving from the T1 to the BRX platform.

[BobO]

I was thinking about it wrong...AC input. Any signal above the on threshold in either direction enables the input. The likely difference is speed of the input transistor or lower capacitance in the input circuit. Because the only way for an AC input to turn off is for it to rest below the on threshold, filtering won't help for fast moving AC noise...can't tell the difference between signal and noise.

[pbw]

Just curious, did you try lowering the Carrier Frequency on the VFD?
Wondering if it had any effect at all?

[mhw]

Just curious, did you try lowering the Carrier Frequency on the VFD?
Wondering if it had any effect at all?

I have done that in the past with GS drives and had limited success. On this job there are two controllers that are using KB open frame inverters. There is no way I know of to change the carrier frequency on them.

[BobO]

I was thinking about it wrong...AC input. Any signal above the on threshold in either direction enables the input. The likely difference is speed of the input transistor or lower capacitance in the input circuit. Because the only way for an AC input to turn off is for it to rest below the on threshold, filtering won't help for fast moving AC noise...can't tell the difference between signal and noise.

And just to clarify, what I mean is the way the filtering is implemented in the FPGA and controller. A hardware level filter would work, but since the PLC is on the other side of a diode bridge, by the time the PLC sees the signal the damage is already done.

[rlp122]

VFD's create harmonics out the wazoo.   Also VFD's/motors must be grounded per the manufacturers specifications.  For GS drives, this means the motor ground must go to the drive chassis.  Bonding the motor to a common ground will cause noise issues.  In addition the longer the drive leads to the motor, the more noise you will get due to the increased reflections of the carrier wave.  Plus you have the whole impedance matching of the supply transformer to the drive to consider.

VFD's are complicated beasts and everything matters when you are trying to mitigate noise issues with them.  Adding EMI filters and Line reactors is often just a part of a more complex solution.

BobO has it right with the speed of the input transistor or lower capacitance in the input circuit.  Both of these items are something that is different in BRX.  When we designed these input circuits we wanted to mitigate the super high inrush current that the older AC input designs had.  We also wanted a high enough speed that it would match up with the fastest AC sensors that ADC currently sold.  Tradeoffs...

[Controls Guy]

In practice I think most AC inputs are on slower functions.  Maybe  a fast and slow version module?  Always the option for interposing relays too.

[mhw]

I am on a job now with two ABB 100HP (ACS880) drives. There are two wires coming from limit switches that will turn the inputs on when the drives are running. There are 2 other limit switches in the same conduit that are not affected. There is only 1.8vac showing up at the PLC input but the light on the PLC input is on. These are BX-16NA input cards
I have moved the affected limits to different inputs, rerouted the wires in the cabinet, and installed voltage dropping resistors at the PLC input terminals. None of these actions helped with the problem.
I then took a 500' spool of 12 ga wire and connected it in series with the input. This has solved the problem.
Maybe you can add an inductor to your AC inputs to help with this problem in the future.