Harmonic Currents

That my friend, is a real good question. If you ever find one please let us know. I would love to find one.

It would be good ammo I have for a pet peeve that has to do with neutral currents on 120/208 v 3ph 4w load centers.

I know that pc's and electronic ballasts draw 3rd,5th,7th harmonic currents and that the 3rd's add up on the neutral;I also know that pc's have a poor power factor ( .6 to .7) and that ballasts have a pf of .9 due to an added capacitor; I also have learned that the power drawn by these devices thru the harmonic currents is vars; so it would seem natural to add a capacitor to any load that needs vars ( just like in pf capacitor correction). The only thing I'm not sure of is whether capacitors can supply vars at harmonic frequencies or just at the fundamental ( 60hz) ? My unsureness comes from the fact that when applying capacitors to improve pf ( at 60hz) we are told to beware of harmonics on the system, because the caps act like sinks and so we have to put filters in front of the caps. Help!

It is a good topic to discuss. I am working in a company where my line harmonics ITHD is 32 to 38% and power connection is 8000 KW. As I could not solve, I am living (managing) with them by localizing harmonics to particular feeder. Some people are installing electronic tube light ballasts to save power but ultimately loose much on the other side. Your neutral wire will draw high currents and you need to install neutral wire that is bigger than all the three phases put together. I know some good PC SMPS manufacturers manufacturing good quality supplies with built-in PFC and very low standby-power consumption. Power Integrations make ICs (TOP switch & TYNY switch) to make good power supplies; but every manufacturer do not make use of good technology available from Fairchild, Powerint, National semi, Motorola, etc. I have opened hundreds of PC power supplies to only find that only one/ two got filters inside. Though there is a place in every SMPS to mount filtering components, majority do not provide, they pocket the money. Then what is the way out? Drag them to consumer forums? File a court case? Black list them? then- who will do it?, where? how? when? As soon there is gray market, this will continue, to stop gray market, then will have to for branded PCs; though very costly??? there are questions and answers.

"opened hundreds of PC power supplies to only find" Hi KV : Since you have seen many SMPS first hand, and since my idea is to apply a capacitor on each typical branch circuit ( phase-to-neutral pc loads ), would you know how one can calculate the proper rating of this capacitor? Do you simply need the amount of vars required to supply to the circuit with , say 5 pc's ( SMPS) ? Do you have to worry about any PIV ( peak inverse voltage) when shutting off the pc's or when shutting off the whole circuit? I think one would have to shut off the capacitor automatically whenever the pc load is off ? then what happens when the cap is sized for all 5 pc's to be on and then someone shuts off 2 of them ?

Dear friend, I am a bit confused with your question. Instead of trying to answering you I shall try to explain harmonics effects & some possible solution. All PCs, Printers, cell phone chargers, that is portable power supplies based on SMPS design generate harmonics. In my country govt fines good amount of money for not maintaining Power Factor at mains. These harmonic generating supplies besides generating hun-wanted harmonics also lower the Power Factor. As a rule all such small power supplies of PCs, Printers, TVs, etc must 1. Must have in-built active power factor correction circuit 2. A filter that blocks generated harmonics by entering mains. Many Chainese PSUs not having these built-in PFC (power factor correction) and filter circuits to keep their prices lower.

PC power supplies got smoothing electrolytic capacitors at high voltage DC. The value of these capacitors depend upon your input supply variation. A push pull flyback SPMS normally designed to vary the duty cycle from 10 to 50 % and whinthin this range SMPS will have to act at mainum/ maximum in-put voltage and minimum and maximum out put load conditions. If you have too big capacitorrs, they damage bridge diodes or NTC or both.

Hope I have explained to your requirement.

KV : In north america, the manufacturers don't care about pf or harmonics. Let me use a specific situation : I have a 120vac( phase-to-neutral) 6ohz load( SMPStype) that has a rating of 10 amp rms/1200va/840watts/0.7 pf/ I 3rd= 30%/I 5th=20%/I 7th=14% . Therefore, I fund= 9.3 amp and vars = 857 . Can I put a capacitor across input of the load, to correct the pf , which means it will provide vars to the load instead of the upstream source supplying this vars ? If I put a 0.73 kvar capacitor across the load, it will correct the pf to 0.99. But we know that some of the .73 kvar is for displacement pf and some for harmonic pf !? So, is this acceptable and normal for a capacitor to be utilized as a source of harmonic currents ? and is there no danger of the capacitor overheating ?or being forced to supply or absorb any more harmonic currents from other upstream parts of the distribution?

Dear ovide brudo: Yes you can connect capacitor to improve PF, but the chances of capacitor capacitor failure will be high. Try use APP type capacitor. As power factor correcting capacitor offers low impedence to higher frequencies (3rd, 5th, 7 th harmonics) and these harmonic currents will try to flow through the capacitor. Optionally I recommend that you block generated harmonics at the SMPS by adding a 60Hz band pass filter and connect capacitor before it. Next solution is have a filter only at the capacitor so that other frequencies are not entering the capacitor. Good filter design consists of series & parallel combination of chokes and capacitors.

Hello Guest : So, if I understand correctly, 1) traveling from source to load, I connect the capacitor ( from phase to neutral) and then connect the 60hz band pass filter in series ( or in line of the phase conductor) and finally the load( which is phase to neutral )? and this is going to prevent harmonic currents from flowing from the load to the source and into the cap? If this is how the filter protects the capacitor, okay. But I don't understand how the load can continue to operate without those vars at the harmonics?? I thought that because they are needed by the load(SMPS) that instead of having the source supply them, that a capacitor( made for harmonics) could supply these harmonic vars( just as traditional pf correction capacitors supply 60hz vars)?

2) What are APP capacitors?

Hello. I am KV: Normally people prefer Metallised Polypropylene (MPP) capacitors as they are cheap and self healing. All Polypropylene (APP) capacitors are costly but can withstand loads. I can not explain here as it is too lengthy subject. Power factor correction capacitors normally mounted at Mains against 3 Phase. In your case you as you are using single phase and load is very small, you can connect at your load. give me your mail ID I shall forward some good literature. kvsubramanyam@hbl.in

Hi Ovide

Capasitors of the self healing type can be overloaded du to harmonic, in capasitors the loss increase with the freq. And the temperatur sensitive film used will be damaged.

Take a look at: http://www.schneider.su/docs/energysaver/Rectimat.pdf

Maby it answers some of your quistions ?

Kind Regards

Hello Neslo : "self healing type " : I tried the link on two separate days and no luck connecting. In any event, does this mean I should avoid capacitors of the self healing film type? Which type do you recommend for harmonic applications?

Hi Ovide

I just tried the link, and it worked on my pc..? (you need adobe pdf reader)

The link giving some useful information to consider, as I thought could be useful for you.

Regarding self healing type of capasitors, they are ok, but only below the max operating temperature. The operating temperature is rather low app. 70 celcius and it is important that this max is respected, if not lifetime will bee shortened dramatically.

I also find your "profile" quistion interesting, but dont belive you will find answers without measuring from case to case...

Kind Regards

Neslo

This is an excellent question.I am equally interested in Knowing if such profiles are available.Harmonic Profiles for the items you have mentioned should be taken by the manufacturers.I doubt whether any one does the analysis on their products.

Hi. For harmonic problems I went to www.internationalpowersolutions.com . They mostly deal with motors and VFD's but they do individual calculations for each item to determine the proper size of capacitor required. The were able to save me 12% on our overall power requirements, which added up to about $6,000/month. I don't know if they can help you with this application but it is worth a try. Good luck.

Hello : I tried your link to IPS but it seems they disappeared; both their 866 tel and their email do not work ? ( the date on their website was 2001 ).

Hi Ovide. I just called up International Power Solutions and they said that they changed their 1800 number but forgot to put it on their website. It will be done this week. The new number is 1-800-704-7602. Hope that they are able to help you out.

ovide brudo,

I don't know. When I put a Dranetz unit on the supply to a 125 Hp VFD about 9 years ago I found the harmonics % to be high, but do not remember the exact amount or the spectrum. You will have harmonics higher than the 25th.

For ballasts I think the results will be similar and will vary from one brand of electronic ballast to another, depending on internal design. It is possible that the big companies will be willing to supply this information (try Phillips, Advance, etc.).

Anybody out there with an analyzer available and a desire to respond to this interesting question?

--JMM

Hi JMM, I've already tried the ballast manuf and they don't offer any harmonic profiles to the public; all they stipulate is that , for example if one purchases a ballast with 20% THDI , "most" of it is 3rd harmonic ! ( not good enough if I want to make calculations for harmonic mitigation design) In any event, I am prepared to test these items myself and document the harmonic profiles; I have a source for renting a Fluke 43B meter, but the biggest problem is obtaining samples of ballasts( c/w lamps and fixture) from the so-called friendly lighting fixture reps; and as for testing various pc's etc, well it's even harder to get samples .

I do not know your location, in India you get lot of 3rd party service providers are available to conduct harmonics study and give you advice. I have few Fluke39, and 41B meters. Soon going to buy Fluke 434 so that I can do analysis for the 3 phases at a time. ABB web site got very good downloads on the subject. I am using 800 battery chargers-cum-regenerative-dischargers from 10KW to 100KW rating. I am able to manage by:1. Loacalyzing harmonics 2. By modification of chargers, but to only some extent.

There was a good article in the Electronic Design News, in the 1991 April 11 issue, "Power-factor-corrected switching power supplies" by Dan Strassberg and in the 1991 July 4 issue, "Specialised ICs correct power factor in switching supplies" by Dave Pryce. If you cannot access them drop me a mail and I will scan & send you them.

Hi Qqberci , Yes please send me whatever you can ( my email is = obrudo@rogers.com ) . However, how valid would 1991 be for equipment manufactured today ? within 10% , 20% or more?; this is somewhat important for applications purposes. Any newer reports from the european community? I tried asking for harmonic profiles from a copper development association in Belgium, but no answer; I also found a thesis prepared by students in Turkey , measuring harm. profiles on pc's,printers, etc. but they did not have models or dates of equipment and when I wrote to them , no answer.

The theory and the terminology are the same so you will be able to find the up-to-date types of devices. I will scan the article and send you.

Please see my message to all in the top level of the Electrical Forum asking for collaborators for a proposed study of neutral currents on 120/208V 3ph 4w systems.

Some who have posted in this thread may have an interest.

The basic problem is, THD cannot be accurately quantified in a test lab; the measurement standards as they exist now are dependent upon too many variables having to do with the installation, which is an unknown as far as the device manufacturer is concerned. This has been in discussion within IEEE for at least a decade that I know of. The original premise was to change the requirements of IEEE 519 to have manufacturers correct for Total Demand Distortion, in other words, the harmonics the individual devices create as a percentage of the demand it has on the power grid, not Total Harmonic Distortion as it is now because THD is affected by other installed devices, wiring, transformers etc. etc. By controlling TDD, each individual device would be evaluated for it's own specific contribution rather than its effect on the entire system, but by enforcing standards, the aggregate effect of all the TDD of each devices having reduced harmonics will end up mitigating the THD at the facility anyway.

The problem with this being implemented is, it will require each and every manufacturer of SMPS devices to test and publish their TDD values, irrespective of installation issues. This is essentially what you are asking for, but you have already experienced the issue first hand. This is a hugely expensive undertaking for manufacturers, one which they are vehemently opposing. VFD and large UPS manufacturers, because their devices are "big ticket", can afford to do the studies and publish results. But small devices such as ballasts and PC power supplies are so cost sensitive the manufacturers don't want to undertake the project, no matter how little the amortized cost will end up being. It's going to take government interference, and most governments are busy making silly laws.

Hi JRaef : Thanks for your candid remarks; I was afraid of the reality out there, but I'm still looking for ways to mitigate these SMPS harmonic currents right at the load.

Could you take a look at my comment #18 and let me know your thoughts ? I'm trying to confirm with someone that a capacitor is basically a "current source" and so ,while we apply them normally to correct pf ( or basically supply vars at fundamental freq or 60hz) why can't we apply them to supply vars at 3rd,5th,7th harmonic ( which is really what the SMPS is asking for -- harmonic currents -- and we know that these are vars and not watts ) ??

As I had promised I scanned the appropriate EDN articles. The file size is a bit more than 29 MB so I have put it to my server. The link is http://qqberci.samunet.hu/Pfc.pdf

still dont understand why all the fuss. A UPS unit with <5% THD can serve for your pc's while you use electronic ballasts with higher pf.