Harmonic damage and repairs

[roger deas]

How many here have actually repaired known damage due to additive harmonic heating.

I know of occasions where burned neutrals were blamed on this in modular office furniture until identical units were opened up and the neutral(actually all of the conductors) were shown to be loose in virtually every unit. (poor quality control)

I have seen one 30 KVA transformer have problems, then was replaced with a K 4 and had no problems afterwards.

This was in an electronics bench testing area in a refurbished building where existing equipment was utilized. The electronics company actually said the problem would occure when they started using the facility. (obviously knew their products)

I know of one other case I have been involved with. This was (2) copiers on a 4 wire multiwire branch circuit.

A major portion of our work is in hospitals and I personally have not seen any cases in these facilities. (Note I said personally)

I was just curious to actual cases the people here have experienced

Roger

 

[bennie]

Re: Harmonic damage and repairs

Roger: Very good topic. I know that a lot of damage has been blamed directly on current from non-linear loads, when the actual cause is oxidized connections.

 

[bphgravity]

Re: Harmonic damage and repairs

This is a tricky topic because of the several cause and effect situations that involve harmonic conditions. Both of you have stated issues with loose conductor terminations causing overheating. The annealing effects of distorted wave forms can cause the terminals to loosen up resulting in greater heat problems. It may apear that the loose terminals were due to installation error, but this is not always so.

I have delt with many motors with greater winding losses on non-linear systems verses the same motors on "clean" systems. The final determination is increased I^R losses from skin effect issues. Isolating these motor systems increased motor life many load cycles with similar usage.

 

[roger]

Re: Harmonic damage and repairs

Bryan, you haven't given a known example as I mentioned in my first post.

Annealing effects can be present simply due to changes in ambient temperatures as well as purely resitive I^R losses. Your argument is speculation in it's best.

We specialize in motors and trouble shooting ( http://www.hayesandlunsford.com/motorrep/motorrep.html ) ( http://www.hayesandlunsford.com/motorrep/mrservices/fieldservices/fieldservices.html ) and I don't know of one "matter of fact" motor failure or shortened life span due to proven Harmonic Distortion.

I'm asking for actual instances of failures due to Harmonics like the transformer example in my first post. Even in my example, the only proof I have this was the reason for the failure, is the replacement with the K4 transformer solved the problem.

Roger

[ April 28, 2003, 09:09 PM: Message edited by: roger ]

 

[ron]

Re: Harmonic damage and repairs

We performed a harmonic study on an Police Headquarters type building, where the distribution system was approx. 30 years old. Luckily most of the circuits were not near their capacity and although measured high current THD were not a problem (but need to be periodically monitored) due to the net heat buildup not being high. Those circuits that were near their capacity, also maasured high THD and needed remediation. Some had filters added to the transformers and some had the transformers replaced with K rated trans and associated circuits upgraded with double neutrals.

 

[roger]

Re: Harmonic damage and repairs

Ron, how were you monitoring the individual circuits? The audits I have been associated with were simply mutiple ampmeters recording the suspect neutrals at peak loaded times.

The transformers are easier for obvious reasons.

Roger

 

[ron]

Re: Harmonic damage and repairs

Fluke has some lovely meters that measure voltage and/or current THD. It helps confirm whether it is a harmonic problem or a conneciton problem (for heating of course).
Another more basic method, is to use a clampon averaging meter and a true RMS meter. The further the readings, the more harmonics that can be expected.

[ April 28, 2003, 10:00 PM: Message edited by: ron ]

 

[roger]

Re: Harmonic damage and repairs

Ron, thats the problem, we have a pretty good size aray of fluke meters including company and personel owned. But even then, with say 10 meters including power quality types there are always more circuits than can be watched in one session.

Roger

 

[dereckbc]

Re: Harmonic damage and repairs

Roger, I think overall the harmonic?s problems are over exaggerated by the lack of understanding of what causes them, and their effects on certain types of equipment. With careful electrical distribution planning and design most of the problems can be minimized or eliminated. It is true the end equipment causes most of the problems by poor filter design, but it is a fact of life even with good design. I work mostly in the telecom and data center world, and I have witnessed a few problems caused by harmonics.

One is overheating of transformers from high neutral harmonic current. In all cases I have seen was the transformers were not ?K? rated. Solution was straight forward of replacing the transformer with a ?K? rated model. This did not cause any equipment problems. It was observed and corrected to prevent fire danger.

Another problem similar was to high of current on the neutral feeder conductor from harmonics. Again the problem was the design of the electrical system not taking into the account the circuits were going to feed non linear loads with switch mode power supplies. This did cause some equipment problems from high neutral-to-ground voltage, which I will try to explain.

With some electrical distribution systems like the ones used in data centers, all the loads are feeding switch mode power supplies. With any switch mode power supply you are converting the AC line voltage directly to dc, and then converting the dc to high frequency AC, and back again to low voltage dc. In the process high frequency harmonics are developed in the neutral conductor and coupled into the EGC by capacitive and direct coupling. This produces common mode noise in the EGC if the voltage between the neutral and ground is relatively high, say above two volts peak-to-peak. The problem this creates is I/O ports that use ground as a signal reference are interconnected. The common mode noise signal can be sensed by the data bit detectors and interpreted as errors.

There are several remedies such as using optical modems and balanced signal transmission between I/O devices. But the fact is not every one uses these methods. Many are stuck on using RS-232 ports. So since the electrical designer cannot control the communication methods and equipment designs, there are steps that can be taken to minimize the voltage drop between neutral and ground to account for high harmonic content. So until equipment manufactures improve their designs and limit the harmonics and use optical signal transmission methods, chumps like you and I can keep making a living. Hope that day never comes!

 

[roger]

Re: Harmonic damage and repairs

Hello Dereck,

I think overall the harmonic’s problems are over exaggerated

this is my opinion also.

I could see you having more experience with harmonic problems than most.

by the lack of understanding of what causes them, and their effects on certain types of equipment

I would think in general we all have a reasonable understanding of the causes and the effects on equipment.

With careful electrical distribution planning and design most of the problems can be minimized or eliminated. It is true the end equipment causes most of the problems by poor filter design, but it is a fact of life even with good design.

agreed.

As an engineer, what is your opinion of zig zag transformers as a solution. I have had no experience with them to date besides literature.

Anyways, I still haven't heard from many with true experiences of known harmonics caused failures.

Roger

[ April 30, 2003, 05:51 PM: Message edited by: roger ]

 

[dereckbc]

Re: Harmonic damage and repairs

Roger, I did not mean to offend anybody by my statement ?by the lack of understanding of what causes them, and their effects on certain types of equipment?. I apologize to anyone who might have been offended. In my line of work I hear a lot from building engineers who panic when high harmonic currents are observed. When I press them to document any problem associated with it, 95% of the time I never hear back from them. For the 5% I do get documented problems it is usually overheating in the transformer, or too high of current in a neutral conductor, and sometimes signal ground problems.

As far as the zigzag transformer used to reduce harmonics is somewhat misleading. "Trap or divert" is more accurate in my opinion. I have installed a couple on the output of isolation transformers just to try them out. What I noticed is downstream they have little or no effect on reducing harmonics. They divert the harmonic current and dissipate the current as heat instead of passing upstream. You can achieve the same thing by using ?K? rated transformers initially.

Dereck

 

[iwire]

Re: Harmonic damage and repairs

Roger, I spend much of my time in office buildings with panel after panel of furniture feeds, most of the furniture feeds we do are two - four wire Multiwire circuits one set for GP outlets one set for IG outlets.

The only problems I have witnessed have had nothing to do with harmonics, bad connectors in the furniture or bad splices by installers.

I think it is a case of "worst case thinking", meaning all circuits will be loaded to the max and all loads will present harmonics.

I am sure the EE wants to give the customer a quality job, this results in electrical systems that are bullet proof.

What I see happening is distribution so oversized and branch circuits so lightly loaded that you could not cook neutrals if you tried, it helped that for a long time the dot coms had deep pockets and liked to spend.

An example, A data room we did for a software company has 6 PDUs fed from a distribution panel, each has a 225 amp breaker at the feed, 300kcmil copper with double neutrals, the longest run is 125'.

I have taken readings on these feeds with the whole room filled with equipment the highest load was about 40 amps.

 

[roger]

Re: Harmonic damage and repairs

Dereck, no appology neceassary. After looking at my post I see where it may appear that I was being defensive, but I didn't mean it that way.

Bob, this has been my experience with office furniture too.

Your description of the data room service is definately overkill unless there is some major growth expected, and I emphasize MAJOR GROWTH.

Roger

 

[charlie tuna]

Re: Harmonic damage and repairs

my work area is large high rise office buildings for the past thirty years, last twenty-two years running my own business and servicing these buildings we originally wired during tenent build outs. we speciallize in infrared scanning, power quality and quantity metering, and in general "fixing other people's screw up's"!
from our viewpoint, harmonics are on the decrease and harmonic problems peaked out about ten years ago. these were problems where neutral conductors burnt up inside transformers and panels from harmonic current, not loose connections. we still see harmonic current with our data loggers, but at lower levels than before. unbalanced loading is a bigger problem right now than effects of harmonics. a portion of the harmonics have been dealt with by the use of dedicated circuits and oversized neutral conductors.
the biggest problem with modular furniture wiring is laser printers!! its common practice to have one one each work station and the modular connectors cannot handle the current draw and loose their contact pressure ability. if you are blessed with installations with dedicated circuits (individual neutrals) and properly circuited for the connected load your problems will be minimumal. some of these manufacturer's products should be outlawed! the clips that hold the connectors together pop loose too easy and allow the connection to operate while only half connected. we have had some come loose and allow the neutral to disconnect totally creating 208 volts across 120 volt devices--this was on modualar units with common neutrals. we have actually had fires in this stuff. another thing is the comments that vfd's cause damaging harmonics in office buildings - this we have never had a problem with!!!!!!!!! my $.02!!