4-20 Noise / Oscillations

[celichi]

Hello Everyone,

I hope we are all well.

I have an L36ERM with a 1769-OF8C/a driving a 5HP Power Flex 523 speed reference 4-20 mA.

With the pump/VFD running, the 4-20 mA command signal is oscillating, causing the VFD to oscillate between 5-10% while the PLC has the 4-20 mA command steady state.

Shields are grounded at one end only.

I have 3 pumps in total, and this is the only one with the oscillating reference.

I think I am picking up noise on the 4-20 mA signal somewhere between the wiring arm and the VFD, all in the same control panel.

What we have tried...

We swapped to a new OF8C card - problem remained.
We ran a new shielded cable from the card to the VFD outside the duct - problem remained.
We swapped to a spare Analog Output on a different slot (OF4C) - problem went away.
Other Analog Output devices are not having any oscillations.

We are going to try swapping wiring arms on the card.

Does anyone have any other suggestions, or has anyone see this or similar before?

Thanks in advance.

 

[ndzied1]

We swapped to a spare Analog Output on a different slot (OF4C) - problem went away.

Was that an OF4C or an OF4CI? I don't see an OF4C output module (not an AB guy so I could be wrong).

But, on the OF8C, all the Anlg Com points are connected internally. On the OF4CI, all inputs are independent. Maybe you have a ground loop?

 

[ArcReactorKC]

If you swapped to another analog output and the problem went away then I would really be eyeing that card for failure. Do you have a process meter? I would inject a 4-20ma signal on the wiring and see if I could repeat the problem that way, as well as metering the output signal from the PLC to see if it fluctuates.

 

[celichi]

Was that an OF4C or an OF4CI? I don't see an OF4C output module (not an AB guy so I could be wrong).

But, on the OF8C, all the Anlg Com points are connected internally. On the OF4CI, all inputs are independent. Maybe you have a ground loop?

My apologies, it is an OF4.

I think you are correct, I think I do have a ground loop.

Just how to find it...remove all other loops on that card, and add them back one at a time?

 

[celichi]

If you swapped to another analog output and the problem went away then I would really be eyeing that card for failure. Do you have a process meter? I would inject a 4-20ma signal on the wiring and see if I could repeat the problem that way, as well as metering the output signal from the PLC to see if it fluctuates.

We swapped the Analog Output Card, and the problem remained, so that led me to believe the Analog Output card is okay, and that the channel is okay?

Yes, we have process meters, that is a good idea, we can inject at the card.

We measured at the VFD terminal in series with our Fluke, and the signal was oscillating.

I did not measure at the card, but will do that today.

 

[celichi]

Well, I am happy and embarrassed to say that I had two PAx objects writing to the same Analog Output Channel.

One PAx object was writing 4000 every scan to the Analog Output.

After doing this for 30 years, you can still be smacked in the face with assumptions.

 

[ArcReactorKC]

I literally was caught up in the same way earlier this week. Although it was a scanport writing to a powerflex 70. I had two separate routines attempting to adjust speed.