Date: June 13, 2000

TO:Patrick Coleman, Alabama Power

Single Phase Harmonics Task Force, IEEE PES/T&D Committee

FM: Ed Yandek, GE Lighting

Member, IEEE PES

Chairman, NEMA Lamp Section Technical Committee

Deputy U.S. Technical Advisor, IEC CISPR/F TAG

Member, U.S. Technical Advisory Group, IEC SC77A

Member, EMC Coordinating Committee, USNC IEC

Member, WG1, TF5A, TF5B, IEC 77A

Subject:Comments—Draft Guide for Harmonic Limits for Single Phase Equipment

Dear Patrick:

Per our earlier email conversations, I promised to provide comments regarding the above document by the end of June.

Please take these comments as constructive and pass them along to other TF members for their consideration.

I fundamentally disagree with the approach taken by the guide – which recommends a single arbitrarily restrictive limit for all products. My statement does not mean that I believe some action by product manufacturers to establish limits for harmonic emissions may not be warranted. Only that I disagree with the broad coverage of this guide and the inflexibility of a single restrictive limit. In my own industry - lighting - we have already taken significant steps to offer products with appropriate levels of harmonic emissions (line current THDs) that match the intended usage and application environment. I could not support this guide for the lighting industry as it is currently written.

Appliances and other products are rapidly increasing in numbers and types. While our concern is with the few that are cited, many others are on the drawing boards that would benefit from having some limits.

If there were some incentive for manufacturers to market low power factor devices, you would see some of the same concern. Those with the largest wattages would be of major concern. Collections of many lower wattage units would also be of concern.

A “one-size fits all approach” maximum THD of 15% as proposed by the draft guide is too simplistic and does not serve the needs of either utilities, product manufacturers, or end users/consumers. It is absolutely essential to develop such guidelines with the intimate participation of those directly impacted - in this case, product manufacturers. Without the participation of such stakeholders, the guide will not be credible nor will it represent the proper best solution.

I will divide my comments into two categories - those that are general, and those that are specific to the document.

General Comments

1. This guide seeks to recommend harmonic limits for single phase equipment—i.e., product equipment. To my knowledge, there has been very little, if any, active participation in the development of this product guide by product industry and manufacturers. Although I welcome this opportunity to comment, impacted stakeholders should be an integral part of such an effort. At this point I would have to characterize the guide as a document that seeks to impose indirect requirements on those that have not participated in its development. This situation opens the guide (and the TF) to a most fundamental criticism of process.

2. There is no evidence presented that justifies a single 15% current THD limit across all product families, power levels, and usage environments. A single 15% THD limit allows no flexibility for manufacturers or end users and virtually mandates the use of active power factor technology for electronic equipment and even many linear (passive component) systems. Such a draconian measure is unwarranted and will not be accepted by many manufacturers.

Concerned with more than just “holes in the dike”. The ones that have shown high potential for problems are certainly of concern, but we’re also concerned with limits for new technologies.

We have only had participation by the lighting and the computer industry. We welcome additional participation by other industries.

If you set 15% THD as a site limit, that’s a suitable target. Having different harmonic limits for different technologies may be the way to go. However, a simpler approach is preferable.

Specific Comments by Document Topic

Introduction

Many in the product sector agree that at some point, guidelines or standards for harmonic emissions of individual single phase equipment will be advisable to prevent objectionable system voltage degradation. However, it is premature to conclude that such measures need be taken now, and that such restrictive limits (15% THD) should be applied across the complete wide range of single phase products.

It is not clear that IEEE 519 has not done an adequate job in minimizing the negative potential impact from single phase harmonic loads where 519 is applied. A very large portion of the connected load in commercial buildings is single phase. When IEEE 519 is properly applied it automatically sets total THD limits at the PCC that by definition include the aggregate effects of single phase loads within the structure in addition to any interactive effects with 3 phase loads [1]. While it is true that IEEE 519 is not applicable to residences or other small users, there is no evidence to state at this point that distributed residential loads are the primary source of the harmonics “problem”. Comments I have heard from various utility representatives reinforce my belief that today’s harmonic issues stem primarily from large single point industrial sources, where IEEE 519 can be readily applied. The next largest source of concentrated loads would be large commercial installations, where, again, IEEE 519 can be readily applied.

Engineering studies have not heretofore made a compelling case that all single phase loads must (or should) have harmonic limits. Many engineering studies or papers that are cited to justify this point of view in fact are theoretical, are very limited in the cases analyzed, do not address aggregate disparate loads, and have not been validated with field data. In short, they tend to be very worst case analyses with worst case input assumptions. The definitive study that can demonstrate exactly what the appropriate practical aggregate summation impact is for disparate harmonic loads in a realistic application environment (building and LV/MV distribution system) has not yet been presented—and this includes the studies cited in the draft guide.

There is a need to get a limit out for existing & future products. The actual limit needs to be jointly explored. When do you start – after there’s a problem?

519 is a three-phase standard applied to industrial sites. It only addresses single-phase in the sense of it being a part of the total load.

Because of the sheer numbers and the forecast, distributed residential nonlinear loads constitute a major threat.

We have ample studies. EDF shows strong correlation of system harmonics with tv usage. Japan has a similar study.

Reasonable engineering studies have been done to date. Product design has changed to avoid the unacceptable harmonic levels anticipated by these studies.

Although it is tempting to surmise that any potential future issue is best solved now by reducing harmonic product emissions to arbitrary low levels across the broad product line, this in fact represents a worst case penalty that is not justified. Although the correct solution takes more time and effort, the only ultimately fair solution needs to evaluate impact from various product classes. A fair evaluation of impact needs to include not only the absolute level of emissions, but also the penetration of such products in the market, usage patterns, and the interaction of such loads in aggregate and with other disparate loads. A more realistic profile of the aggregate future load can then be physically tested and measured on real distribution equipment. Only then can we collectively begin to assess whether such projected future load profiles will cause intolerable degradation to the voltage integrity of the distribution system.

Additionally, consideration must also be given to the minimum level of performance (susceptibility to harmonics) that various distribution systems should be expected to provide. Any ultimate decision to greatly reduce harmonic impact in the systems should represent a balance – some products should have reduced harmonic emissions, and some systems (utilities) should implement needed infrastructure improvements to improve harmonic susceptibility.

It is quite clear that impending deregulation concerns have resulted in a slowdown of utility infrastructure improvements - see virtually any issue of the PES Journal for such discussions. As electrical demand continues to rise, the issues of capacity and harmonics collide in a way that makes it even more attractive for utilities to seek a means to “linearize” the aggregate collective load profile as a means to temporarily “add capacity” and defer costs by shifting the cost burden to manufacturers and end users through new product requirements.

I take strong exception to the statement that proposed THD limits will minimize impact to existing equipment design/products. There is a subset of products for which the statement may in fact be true, but it is not true for equipment in general across the board. It is certainly not true for many types of lighting equipment that have already been designed with moderately low levels of harmonic emissions and that are very cost sensitive. Such equipment has not been implicated as a major source of widespread harmonic problems throughout the distribution network or within premises. Reducing the harmonic levels of such equipment further would add unjustified costs (both engineering and end user costs), reduce performance, and negatively impact other product aspects such as size, form factor, reliability, and aesthetics.

I agree, but this effort only tends to delay our work. Large power devices and collections of high-coincident-use devices are critical first steps to go after.

A more appropriate approach would be to narrow the discussion to specific product families and technologies of immediate concern and to concentrate the discussion accordingly.

Guidelines could then be developed jointly that would be embraced by all parties. In such a process the product industry would be a prime determiner of what THD levels could be economically viable over what time period. After a suitable period of time it would then be possible to ascertain (with ongoing system monitoring of THD levels) whether the next tier of products would need to be redesigned or not.

Overview

I agree in general with many of the general concerns expressed in the overview. However, the key to knowing what action to take in setting broad limits is exactly what is admitted to be missing - namely, the impact of aggregate loads in combination on the network as a whole or within a premises or local network.

In addition, it is critical that monitoring of a distributed (system wide) THD profile be an integral and ongoing part of any discussion that would seek to recommend significant or widespread reduction in single phase harmonic emissions for individual products. Such monitoring and data sharing is essential to determine the effectiveness of interim (step approach) limit development for so-called high impact products.

Example - Suppose the HVAC industry would agree to some economically viable harmonic reduction measures for variable speed heat pumps and AC equipment targeted at residences. If the HVAC industry is willing to underwrite the cost of such design and product changes, it is essential that utilities be willing to monitor the effect as such products grow in penetration since a decision on whether even more restrictive requirements would ever be needed must be based on data.

There is certainly reason to suggest that aggregate disparate loads cancel more than might be predicted by worst case simple models. Is this why aggregate THD values have not, in general, risen greatly despite the already significant increase in non-linear loads? Keep in mind I have not said that no action is needed. The trick is to find a level of action that all sides can support and endorse - take that action - monitor and evaluate - and then collectively decide if additional action is warranted. The “step wise approach” has been used to control harmonics in Japan and has proven very effective [2]. There is even reason to believe that harmonic emission reductions implemented on some products in Europe and the US may already be having a positive impact.

Only through cooperative (utility and product industry) efforts can we seek to truly “calibrate” what steps may be prudent - for which products and application environments. And, most importantly, only through cooperative efforts can we monitor the effectiveness of any change and determine if it has been sufficient, or if additional requirements are appropriate.

For more discussion on how best to balance the needs of manufacturers, users, and utilities on this issue, see the United States National Committee of the IEC’s Powerline Harmonics Position Paper [3].

Scope

120, 240, 277 voltages all imply certain application environments. Although the scope of the guide may be appropriately up to 600V, it would not be appropriate to automatically assume the same THD limits should apply respective of “product input voltage”. Commercial and industrial sites have additional harmonic mitigation options that can be applied at the premises. This needs to be taken into account when deciding whether certain commercial and industrial equipment should have limits - or what those limits should be.

The guide should only apply to certain types of high impact equipment. It is not acceptable to set limits for “lighting” equipment, or “HVAC” equipment, or “IT” equipment. Within each product family, there should be the option of:

No limits

Relaxed - Moderate limits

More severe limits (APFC)

This reads like we have a lot of data that indicates that harmonics are not growing significantly and are not a problem. I believe we really don’t know. The EPRI DPQ project gives us some information. The EDF paper indicates otherwise.

I agree we need a cooperative effort. We welcome more participation by the lighting industry and others.

1 A at 240 V is equivalent to 2 A at 120 V – in terms of what gets on the primary. We can handle the voltage issue fairly cleanly.

We could do something like this. I think we need a broader guide and not just singling out spectific targets.

This is like a bad - better - best approach, although no limits would be the worst situation.

Limits should be technically realistic from an implementation point of view. Moderate limits should be what can be achieved realistically and cost effectively via means other than by active power factor correction.

Limits must allow higher THD levels for some products vs. others - even in the same product family. Products below a specified power should not have limits at all. The guide should be applied to those types of loads for which there is general agreement (and supporting evidence) that impact will be intolerably high if penetration is high. Products that can be agreed to have minimal impact (not necessarily zero impact) should be exempted from coverage.

The scope is a key indicator of the underlying strategy for the guide. It would be better to agree on a relatively small number of high impact products for which limits are logical (and to specifically call then out in the standard) than to try to paint every product with the same brush regardless of justification or impact. Such a guide can readily be revised periodically to keep up with technology changes or market innovation since, realistically, new product families do not penetrate the market overnight.

Purpose

The purpose of the guide should be to give guidance to product manufacturers and product standardization committees who may wish to accept the recommended limits for single phase equipment.

Although existing IEEE 519 is site applied, it automatically incorporates premises single phase/three phase interactions (mitigation) since it is applied at the PCC.

Loads Covered In This Standard

The word Standard should be replaced by the word Guide.

This section describes, by inference, loads that are of greatest potential concern. If that is truly the intent, then the products (loads) that are covered should somehow correspond to products (loads) for which there are ultimately limits. Only covered loads should have any limits at all. Limits should not apply to loads that are not specifically covered. The guide is very unclear in this regard.

We may need to do this. This appraoch gets complicated in a hurry. Instead of a guide, you’ll have an encyclopedia.

OK.

With a loophole like this, this is where the variable speed heat pump sneaks in without any limit. We have to be very careful and not ignore limits applied to future technologies.

As written, the guide discusses certain loads in an informative manner and then recommends that all loads have a very low THD level of 15%. Specific covered loads should be called out in the scope section of the guide. Section 4.0 is more appropriately a rationale and justification for why certain loads are covered (have limits). By narrowing the scope the guide will become much less open to criticism of being overly broad, and it would be more realistic to involve the appropriate product manufacturing sectors.