UPS and Power Line Noise

You can try and filter the supply to the UPS (voltage stabilisers etc e.g. https://www.claudelyons.co.uk/products/voltage-stabilisers) but at the end of the day most folks would be better to fix the problem at source and filter the VFDs. Always difficult - either throw solutions at it until one sticks or investigate properly and find the actual source of the problem.

Bear in mind that if your local UPS can't cope then you are also sending those same harmonics/dips etc back up the supply chain and may be affecting other gear.

It sounds as though the bottleneck is the cables. Have a qualified local Electrical Engineer assess the cable sizing according to local electrical codes, and recommend improvements.

You have harmonics generated by non linear loads.

These harmonics can generate excessive currents in the neutral conductor.

The NEC recommends doubling the size of the neutral in such cases to prevent damage to wiring and equipment.

There are many ways to mitigate the harmonics as well.

These issues are covered very thoroughly in this link:

https://www.researchgate.net/publication/3169578_Improving_power_factor_in_the_presence_of_harmonics_using_low-voltage_tuned_filters

Thanks to all,

The cables are good and sized correctly. The load lines to the motor are only 15'. Voltage drops are negligible or non-existant. Of course getting the noise at the source would be best but some someone above my pay grade decided to remove them and now we have to sort out the issues

• So why does the UPS alarm on low voltage at higher loads, if <...The cables are good and sized correctly...>?
• Has anyone oversized the cables yet, particularly the neutral conductor, by way of experiment?
• Has thermal imaging been carried out on the cables, particularly the neutral conductor, at higher loads?

Way to go.

As suggested,view the power with an oscilloscope,or rent a line monitor and it will give a recording of the upsets on the line(voltage).

A clamp-on amp meter will not detect harmonic currents in the neutral,that is why the NEC recommends DOUBLING THE SIZE OF THE NEUTRAL to prevent overloading of the neutral and reducing harmonic frequencies.

There are special meters that will detect harmonic currents,but not your standard clamp on type.

There is the possibility of a loose connection at some point in the system that is exacerbated by high load.

An infrared thermal scan would beneficial to find the problem.

Where is the source of the 120Volts for the UPS?

Is it powered from the same source as the 3 phase for the inverter,through a transformer?

Lots of unknowns here.

Hi Dean,

Are the cables from the VFD to the motor screened and is the screen terminated at one end.

Normally screened cables are required to prevent the RFI from being transmitted but remember the screening should only be terminated at one end else an earth loop could be introduced to act as yet another source harmonics.

As others have said check the supply to the pump skid with a power scope to check for interference.

A suitable filter on the supply for to the UPS would be beneficial in any case or you could use a constant voltage transformer to supply the UPS.

Also placing the UPS and PLC in a single enclosed cabinet should help eliminate RFI and EMI.

Regards Stef

Are the motor leads in METAL conduit? If so, are the conduit(s)properly bonded at each end to the metal frame,using bonding lock nuts where applicable? Do the motor conduits share the same conduit at some point?Does each motor have a separate conduit?Are the phases to the motors separated at any point? If the motors share a common feeder conduit,are both motors run at the same speed all of the time?

All conductors of a 3 phase supply conductors must run together in the same conduit to prevent induced currents in the conduit and in the other conductors.

Here is a link to factory recommendations to eliminate noise in/from VFD drives.

I suggest that your problem is probably not unique,and the answer may be found in this link.

Good luck.

http://www.heanderson.com/VFD%20Whitepaper.pdf

If you had a working system that was butcherd by a higher pay grade, the only way is to get an even higher pay grade involved and go back to what worked. Engineerig should run engineering and not higher pay grades.

Can you put an oscilloscope on the equipment? A little insight into what is actually happening could be very useful. If the steady state voltages are good then filters might be very helpful. If the E=IR drop drags line voltage below the turn on threshold of the UPS then all the filters in the world won't help.

Yes we did test the voltage at 100% load and there was no voltage drop seen with a true RMS meter. I was using the term "noise" generically- classifying harmonic distortion in that category.

3% line reactors were used in the original configuration but someone decided to omit them in recent manufacturing, creating this situation.

And thank you for pointing out the sensitivity adjustment. That is something we missed.

One of the advantages of the 3% line reactors on the VFDs is that momentary power/voltage dips that are not long enough to drop out a typical FVNR motor starting contactor (63% one cycle) will also possibly not drop out your drive.

I found this out the hard way, where I left off the reactors because of space in an Allen-Bradley Centerline MCC retrofit. So the next time National Grid bumped the plant, the only two motors that stopped and shut down the process were the 2 new conversions I had just completed. This happened at least 2 more times before I made the space for the reactors. After they were installed, either we rode through the bumps, everyone, or the entire plant went down.

I agree.I have seen 1:1 transformers clean up a lot of harmonics;such as from a inductive cap sealer, VFD,elevator controls etc.

What's your power ratio Supply Capacity of the UPS vs. Load? If its close to 1 then probably you got scam by harmonics.

Line filters will do the job.

How close are the pumps and the controls?

Pumps and controls are about 15' away.

UPS load ratio is about 0.23

We are going to try new settings on the UPS on this weeks startup

"Stepped approximation to a sinewave"

I think any process running 2 @ 100hp pumps likely involves more value than that little UPS is worth. Last time I looked at APC's products, there wasn't much in that price range that had sine wave output, filtering, and voltage regulation.

Buy a better unit, one that provides your controls clean power. I have purchased CyberPower PFC models for a similar application in the past, because of cost & size considerations. I wondered why Schneider/APC doesn't sell a equivalent product.

Since you have not given the UPS mains input frequency, voltage or VA/current rating; it would be difficult for someone to suggest a filter between UPS line input and UPS!!

The usual line filter does not have much effect At 100 kHz, particularly not for line-line - which is what the UPS looks at.

As suggested in another post, a look at the UPS mains input with an oscilloscope would help a lot to find the nature of the interference.

Do you have a grounded supply to UPS?? Floating adds problems

I had a communication issue that was resolved by the addition of a filter to the electric function. In my case it was an LED FAA Marker light that messed up a 911 relay antennae on a water tower. I found the filter at the following link.

https://www.alliedelec.com/schaffner-fn2010-10-06/70027166/?mkwid=s9chY8ysI&pcrid=30980760979&pkw=&pmt=&gclid=EAIaIQobChMI2PaVxqPw3wIV1LfACh2EYAlCEAQYASABEgKAI_D_BwE