.

The methods, procedures and data reported here are intended to be consistent with industry standards and protocols as documented by ANSI and reported by recognized lighting industry manufacturers and certification test laboratories. In some cases no unique procedure has been specified and individual adaptations are employed. In other cases the product configurations and functions have evolved in a manner to make previously documented test parameters obsolete, e.g., ballast factor for an integrated ballast-lamp configuration. The principal investigators have conferred with individuals responsible for reviewing standard testing documentation and who are currently testing their own products in order to perform tests and report results that will be most useful in assessing advanced products in a very dynamic technology and market environment.

The information presented is documented so as to be a self consistent reference base that others can duplicate for comparisons under comparable conditions. There is no intent to represent the information as a basis for judging the quality of performance from component to component. The test procedures and corresponding observed performance may provide insight for updating standards and identifying additional useful investigations to be performed. The project performed under the CEC authorization is exploratory in nature and should not be taken as either an endorsement or repudiation of any of the components tested.

Consistent with being a non-commercial, academic research project, the above parties hereby expressly disclaim any liability or responsibility for loss or damage resulting in the use of any apparatus, method, or process disclosed in this report; and for the potential,infringement of any patent or violation of any federal, state or municipal law or regulation arising from the use of any information, apparatus, method, or process disclosed in this report.

Additional copies of this report as well as details on the availability of other publications may be obtained from the California Energy Commission or directly from

College of Engineering

California State Polytechnic University, Pomona

3801 W. Temple Ave

Pomona, CA 91768

(909) 869-2600

Volume 1

Abstract ………………………………………………..…….….iii

Acknowledgements ………………………………………….…..iv

Executive Summary …………………………………………........v

Sections

1. Introduction ……………………………..…………….….1
2. Total Harmonic Distortion (THD) Tests ……………….…8
3. Efficacy, Ballast Factor, Color Rendering Index Tests …..12
4. Starting Characteristics/Inrush Current ……………….…14
5. Radiated Emissions Tests…………………………………16
6. Conducted Emissions Tests ………………………………23
7. Bibliography ……………………………………….……..26

Volume 2

 

Appendices

 

1. Total Harmonic Distortion (THD) Test Data
2. Efficacy, Ballast Factor, CRI Test Data
3. Starting Characteristics Test Data
4. EMC Radiated Emissions Data
5. EMC Conducted Emissions Data

Fifty-seven (57) lamps and ballast combinations were tested for operating characteristics. The tests were performed in the lighting and photometrics laboratory at California State Polytechnic University (Cal Poly), Pomona, CA. The tests performed included precompliance RF emissions, conducted RF emissions, lumininous efficacy, starting current and sputtering times, harmonic content of both voltage and current, power, power factor, color rendering index, and color temperature. Equipment used to perform the tests were obtained through grants from the California Energy Commission, the Department of Energy, the National Science Foundation, and local industry. The laboratory is part of the Center for Lighting Education and Applied Research (CLEAR) at Cal Poly Pomona. It has been equipped with instrumentation representative of the most recent industry practices as defined by active industry advising groups to CLEAR. Test were performed

by students and faculty associated with the Illumination Engineering Program at Cal Poly Pomona.

The investigators wish to acknowledge the provision of test equipment received through grants from the California Energy Commission, the Department of Energy, the National Science Foundation (NSF-ILI), and local industry. Sample lamps and electronic ballasts were provided by industry. We wish to acknowledge the support of members of the NEMA sub-committee for ANSI C78, Dimensional & Electrical Characteristics of Fluorescent Lamps and ANSI C82, Fluorescent Lamp Ballasts. We also wish to cite the contributions of Long Trinh of Hewlett Packard for assisting in the testing protocol for the RF measurements and Richard Docter of Raytheon for allowing us to use their EMC Test chamber. We would also like to express our thanks to John Fluke Co. for their donation of software which enabled the capture of transient current during lamp starting.

A special thanks to professors Robert G. Irvine and Armanda Hull for their efforts in performing tests and collecting data. Thanks to Mark Bailey for his assistance in constructing some of the test apparatus.

Investigators are especially thankful to Paul Rhonemus, a student and employee of Raytheon, for his efforts in making the EMC measurements and establishing testing protocol in accordance with FCC Title 18. We are appreciative of other students who worked many hours on this project.

Results of the tests indicate that the electronic ballasts for all components investigated by this project for both compact fluorescent lamps (CFL’s) and F32T8 lamps comply with the requirements of FCC 18 for residential radiated EMC emissions. Radiated emissions from electronic ballasts were observed above 200 MHz even though their fundamental frequency was near 40 KHz.. Some of the ballasts may not comply with the conducted EMC emissions requirements of FCC 18. Other computer equipment connected to the laboratory power grid could have adversely affected the resolution of the conducted RF emission’s tests. Neither the state nor industry have established minimum or suggested standards for a number of the tests that were performed. Current crest factors higher than 4 were observed. The effects of values this high on peripheral equipment such as light switches and occupancy sensors were not investigated. Additional studies should be performed in this area. Delays in excess of 1 second were observed in the starting of some lamps and inrush currents persisted for more than 3 seconds in other cases. The starting delay may be annoying to the typical end user. The persistent inrush current may also affect peripheral equipment as stated earlier. The ballast/lamp combinations tested had a wide range of efficacies and many of the combinations did not meet the minimum requirement of 40 lumens/watt as required by Title 24, Residential Manual, section 2.4 for use in kitchens and bathrooms. Current Total Harmonic Distortion (THD) as high as 175% was observed. Its affect on power quality is a growing concern for power companies. This can adversely affect the neutral current on branch circuits. Large neutral currents can result in electrical fires; therefore, it is suggested that further studies be conducted to determine the risks of electrical fires caused by the extensive use of compact fluorescent lamps in order to conserve energy. None of the compact fluorescent lamp (CFLs) assemblies had their color temperature on either the product or packaging. Information of this type would help consumers make informed decisions concerning the selection and application of compact fluorescent lamps.

This project to characterize the operating parameters of electronic ballasts operating with compact fluorescent lamps is in response to the California Energy Commission’s (CEC) interest in comparing the lighting performance, energy efficiency, and side effects associated with different vendor’s components.

The California State Polytechnic University at Pomona, CA (Cal Poly Pomona) is interested in performing projects of this nature in its Center for Lighting Education and Applied Research (CLEAR) both for the educational opportunity provided students to gain experience on real industry activities and as a partner with the public and private elements of the lighting industry. Prior to undertaking this project for the CEC, CLEAR had just finished furbishing its photometrics laboratory with instruments representative of the latest industry capabilities. This project afforded the University a vehicle to demonstrate and verify the performance of the laboratory instrumentation. Many prominent individuals from a number of the major lighting companies provided consul and lighting components to the principle faculty investigators throughout the conduct of the program. The dissemination of the results and procedures may provide both an enhanced public awareness of the attributes and variations of the products used in this study and provide industry with insight regarding the testing procedure employed.

Developments in miniature electronic ballasts and compact fluorescent lamps have resulted in a number of companies that manufacture an integrated ballast and lamp assemblies. In many cases the specifications of either the ballast, lamp or both is not known. This investigation was an effort to determine the characteristics of a number of lamp/ballast combinations. An uninformed public could misuse some lamp/ballast assemblies if they are not aware of their characteristics. Some of the test data would be of little use to the general public in that it pertains only to compliance with Federal Regulations. The majority of the data would assist the public in making a decision on which lamp assemblies best suit their application. The use of compact fluorescent assemblies in the home is increasing and to date there is little data available on their characteristics. It is important that the lighting professional and general public be able to select the correct type of ballast/lamp assembly both for quality of lighting and energy efficiency. Characteristics of Compact Fluorescent Lamps are not contained in the California Energy Commissions Directory of Certified Fluorescent Lamp Ballasts and Certified Luminaire Manufacturers, Pub. P400-95-026, April, 1995. It is our recommendation that lighting assemblies should eventually be considered like air conditioners and hot water heaters. In that only those tested, characterized and approved by the California Energy Commission should be permitted to be installed and used in the State of California as outlined in Appliance Efficiency Regulations for…, pub. P400-92-029, April, 1995. A relatively new factor of fixture performance is luminaire efficacy rating (LER). This is similar to that used by refrigerator manufactures where the rating is based on the annual cost of operating the appliance. The LER rating is based on 3000 hours of operation per year at a cost of $0.08/KHr. It should be noted that the General Services Administration (GSA) requires that federal buildings have a lighting power density of less than 1.1 W/SF.

Test equipment available in the lighting and photometric laboratory at Cal Poly Pomona University was used to perform the tests to determine the characteristics of lamp/ballast assemblies. A section of the laboratory is shown in Figure 1-1.

The project performed here is a research project predicated upon the availability of latest industry standard components and laboratory capabilities available at the university. Since the photometric laboratory was in the process of being brought into being at the time the project task was being formulated and product line and corresponding testing was in a state of evolution, some of the originally designated tasks were either modified, deleted or added as appropriate. For example, lamp performance at cold temperatures and audio tests were not performed. The ballast factor test was modified to accommodate a sealed ballast/lamp assembly. The program anticipated testing only twenty ballast. However, the testing was extended to fifty-seven ballast/lamp assemblies as a consequence of manufacturers supplying a large number of their most recent products. Figure 1-2 shows some of the ballast/lamps that were tested. Many additional samples are available for testing that is beyond the scope of this contract. Results of the tests are summarized in Table 1-1.

Figure 1-2. Ballasts/Lamps used in the Research Project

Table 1-1. Summary of Test Results for Ballasts/lamp Assemblies

Ballast/Lamp Under Test	Description	Fundamental Frequency	FCC 18 Pass	Comments	FCC ID	Lumens	Watts	Efficacy (lm/w)	LER	CRI	Color Temperature OK	Manufactures Rated ouput of lamp (lumens) Source: Osram Sylvania /G.E. Cat.
													X	y
1	Sylvania Delux Electronic 15 W	41.5KHz	xx		EB329101A	578.11	14.3	40.4	3.4	84.04	2643	900	0.4655	0.4134
2	Lights of America 2342 42W, 0.55 A	43.4 KHz			N/A	2580.3	40.3	64.0	9.7	81.8	2707	N/A	0.4591	0.4103
3	Sylvania 18 W, 0.3 A	32.4 KHz	xx		EB329070A	805.4	18	44.7	4.3	80.2	2641	1250	0.4633	0.4092
4	Sylvania 18 W, 0.3 A	29.1 KHz	xx		EB329070A	931.5	18.4	50.6	4.4	80.1	2659	1250	0.4639	0.4127
5	Fulham 15 W, pf > 0.95	Mag. Noise	xx	Pass, spikes at 42.2 MHz and 80 MHz	N/A	507.1	15.3	33.1	3.7	87	6358	900	0.3143	0.3375
6	Lights of America FC8T9/WW/RS	Magnetic		1 KHz BW Periodic Noise	N/A	907.7	22.6	40.2	5.4	52.5	2998	N/A	0.4335	0.3969
7	Genura 23W	Magnetic		Borderline between 200 and 220 MHz	N/A	300.6	14.5	20.7	3.5	85.1	2949	1550	0.4471	0.4186
8	Philips SLS 11, 11 W	49.1 KHz	xx	spike at 78 MHz	CIWSLSC	377	10.6	35.6	2.5	83.5	2655	600	0.4678	0.419
9	Sylvania 20W	41.9 KHz	xx		EB329102A	609.1	16	38.1	3.8	83	2648	1200	0.4657	0.4145
10	FEIT Electric 18 W	Magnetic			N/A	495.21	15.2	32.6	3.6	84.1	2629	1250	0.4656	0.4118
11	Sylvania 15 W	43.4 KHz	xx	Pass	EB329101A	272.8	14.5	18.8	3.5	83.6	2689	900	0.4612	0.4118
12	Sylvania 11W	35.7 KHz	xx		EB3DULUX-EL001	303.2	11.2	27.1	2.7	83.8	2540	600	0.4756	0.417
13	Sylvania 15 W	37.4 KHz	xx		EB3DULUX-EL001	559	14.5	38.6	3.5	83.5	2579	900	0.4718	0.4159
14	Philips SLS 11, 11 W	49.1 KHz	xx	spike at 80 MHz at limit line	HK7J032329	540.4	10.7	50.5	2.6	82.8	2716	600	0.4608	0.4145
15	Philips SLS 9, 9 W	49.6 KHz	xx	spike at 80 MHz at limit line	CIWSLSC	276.1	8.7	31.7	2.1	83	2762	575	0.4604	0.4199
16	Osram Dulux Electronic, 23W	40.3 KHz	xx		H7JO32329	681.1	19.8	34.4	4.8	83.4	2620	1550	0.4686	0.4159
17	Motorola Gold Edition Electronic Ballast, Rapid Start, M1-CF-32-B-S-120 for 32W & 26 W, with Philips 26W/30/4P lamp	32.3 KHz		Noise raised ~ 40 dB in some places. May not pass	N/A							1800
Ballast/Lamp Under Test	Description	Fundamental Frequency	FCC 18 Pass	Comments	FCC ID	Lumens	Watts	Efficacy (lm/w)	LER	CRI	Color Temperature OK	Manufactures Rated ouput of lamp (lumens) Source: Osram Sylvania /G.E. Cat.
18a	Lights of America,2726 Single Solution 488 (3-way) 8w/26/w/16w, 0.4 a	113.6 KHz	xx	spikes at 40 MHz and 80 MHz. Below limit line 10 dB.	IZM 2633	383.3	9.3	41.2	2.2	83.2	2610	N/A,1800,N/A	0.4664	0.4076
18b	Lights of America,2726 Single Solution 488 (3-way) 8w/26/w/16w, 0.4 a	115 KHz	xx	spikes at 40 MHz and 80 MHz. Below limit line 10 dB.	IZM 2633	969.6	17.6	55.1	4.2	83.2	2612	N/A,1800,N/A	0.4597	0.3991
18c	Lights of America,2726 Single Solution 488 (3-way) 8w/26/w/16w, 0.4 a	115.4 KHz	xx	spikes at 40 MHz and 80 MHz. Below limit line 10 dB.	IZM 2633	1049.8	17.1	61.4	4.1	83.2	2617	N/A,1800,N/A	0.4656	0.4101
19a	Lights of America,2726 Single Solution 488 (3-way) 8w/26w/16W, 0.4 A	56.8 KHz	xx		IZM 2633	441	10.1	43.7	2.4	83.2	2594	N/A,1800,N/A	0.4589	0.3958
19b	Lights of America,2726 Single Solution 488 (3-way) 8w/26w/16W, 0.4 A		xx		IZM 2633	1487.1	22.6	65.8	5.4	83.2	2644	N/A,1800,N/A	0.463	.409+1
19c	Lights of America,2726 Single Solution 488 (3-way) 8w/26w/16W, 0.4 A		xx		IZM 2633	990.8	15.6	63.5	3.7	83.2	2623	N/A,1800,N/A	0.4663	0.4121
20	Lights of America,2332, 448, 32 W, 0.45 A	42.2 KHz	xx		IZM 2100	1387.1	29.6	46.9	7.1	83.4	2589	2200	0.4661	0.4075
21	Lights of America,2332, 448, 32 W, 0.45 A	41.3 KHz	xx		IZM 2100	1731.7	30.2	57.3	7.2	82.5	2669	2200	0.4594	0.4061
22	Lights of America 2127 IC (248), 27W, 0.43A	42.22 KHz	xx	Tentative pass. Possible problems at 42 and 79 MHz.	IZM 2100	1389.2	30.4	45.7	7.3	84.9	2546	N/A	0.4683	0.4056
23	Lights of America 2127 IC (248), 27W, 0.43A	48.36 KHz	xx	Tentative pass. Possible problems at 42 and 79 MHz.	IZM 2100	1210.3	27.7	43.7	6.6	85	2522	N/A	0.4691	0.4039
24	Lights of America FCL30EX-L Lamp, 2630, 30 W, 0.46 ballast	39.6 KHz	xx		IZM 2027	1235.2	31.1	39.7	7.5	85.8	2632	N/A	0.4665	0.4138
25	Intelect Plus, PUV-T13RS-EOL, ballast for 1 lamp CFM 26W/32W/42W, 4 pin	40.5KHz,	xx		N/A							1825,2200,N/A
26	Intelect Plus, PUV-T13RS-EOL, ballast for 1 lamp CFM 26W/32W/42W, 4 pin	40.5KHz,	xx		N/A							1825,2200,N/A
27	Intelect Plus, PUV-T13RS-EOL, ballast for 1or 2 lamp CFQ26W, lamp CFM26W,CFT27W and 28W 2D	40.8KHz	xx		N/A							1825, N/A, N/A
28	ES Super LampGard, ES-2-CFQ-26-120-G-LG, 2-Lamp, 26W, compact 4pin, quad	36.6KHz	xx		N/A							1825
29	Motorola Gold Edition Electronic Ballast, Rapid Start, M1-CF-32-B-S-120 for 32W & 26 W, for 1 lamp 4pin CFL 32W & 26W	31 Khz loaded 73.4 KHz unloaded	xx	Tenative Pass. See also #17	N/A							2200.1825
30	Motorola Gold Edition Electronic Ballast, Rapid Start, M1-CF-32-B-S-120 for 32W & 26 W, for 1 lamp 4pin CFL 32W & 26W	31.4 KHz	xx	Tenative Pass. See also #17	N/A							2200.1825
31	Motorola Gold Edition Electronic Ballast, Rapid Start, M2-CF-26-B-S-120 for 2 lamp CFL 26 W 4pin	70.9KHz	xx	Passed EMC. Burned up lamps. Did not investigate cause.	N/A							1825
32	Motorola Gold Edition Electronic Ballast, Rapid Start, M2-CF-26-B-S-120 for 2 lamp CFL 26 W 4pin	39.8 KHz loaded & 69.9 KHz unloaded	xx	Passed EMC. Burned up lamps. Did not investigate cause.	N/A							1825
Ballast/Lamp Under Test	Description	Fundamental Frequency	FCC 18 Pass	Comments	FCC ID	Lumens	Watts	Efficacy (lm/w)	LER	CRI	Color Temperature OK	Manufactures Rated ouput of lamp (lumens) Source: Osram Sylvania /G.E. Cat.
33	Motorola, M2-IN-T8-D-120, 2 Lamp, F32T8,F25T8,F17T18	39.8 KHz (32W)	xx		N/A							3000
34	Motorola, M2-IN-T8-GP-D-120, 2 Lamp, F32T8,F25T8,F17T19	39.8 KHz	xx		N/A							3000
35	Motorola, M2-RN-T8-ILL-D-120, 2 Lamp, F32T8,F25T8,F17T20	47.8 KHz	xx		N/A							3000
36	Lights of America, 2630, 30W, 0.25 A	40.8 KHz	xx		IZM2027	1850.1	33.6	55.1	8.1	86.4	2753	N/A	0.4585	0.4153
37	Intelect Plus, PUV-22RS-EOL, ballast for 1 or 2 lamp CFQ26W, CFM26W, CFT27W, and 28W 2D	40.8 KHz (26W)	xx		N/A							1825,N/A,N/A
38	Advance Transformer, Mark X, REZ-1T42,for FM42W lamp, 0.42 A	39.4 KHz (40W)	xx		N/A							N/A
39	Advance Transformer, Mark V, RIC-2S32, for (2) F32T8 or (2) F25T8 lamps, 0.54 A	51.3 KHz (32W)	xx		N/A							30,002,125
40	Advance Transformer, Mark V, RIC-2S32, for (2) F32T8 or (2) F25T8 lamps, 0.54 A	52.4 KHz	xx		N/A							30,002,125
41	Advance Transformer, Mark X, REZ-1T42,for FM42W lamp, 0.42 A	41.9 KHz (40W)	xx		N/A							N/A
42	Intelect, PUV-T13RS, for (1) 26W/32W/42W, RS triple tube or (1) 32 W heliax CF Lamp	30.3 KHz (26 W)	xx	Tentative Pass. Possible problems at 50 through 130 MHz.	N/A							2200
43	Intelect, PUV-T13RS, for (1) 26W/32W/42W, RS triple tube or (1) 32 W heliax CF Lamp	30.3 KHz (26 W)	xx	Note: Large harmonics see data sheet	N/A						2583	1825,2200,N/A
44	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	122.4 KHz 58.7 KHz 121.3 KHz	xx		IZM2633	602.8	23.8	25.3	5.7	84.6	2556	N/A,1800,N/A	0.473	0.4149
45	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	124.1 KHz 58.3 KHz 123.4 KHz	xx		IZM2633	697.3	23.6	29.5	5.7	84.4	2590	N/A,1800,N/A	0.4701	0.4143
46	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	119.6 KHz 57.3 KHz 119.2 KHz	xx		IZM2633	607.1	23.3	26.1	5.6	84.7	2580	N/A,1800,N/A	0.4712	0.4149
47	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	121.0 KHz 56.9 KHz 120.3 KHz	xx		IZM2633	612.2	24.1	25.4	5.8	84.8	2583	N/A,1800,N/A	0.4708	0.4146
48	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	118.9 KHz 57.6KHz 118.9Khz	xx		IZM2633	1521.3	24.07	63.2	5.8	84.7	2698	N/A,1800,N/A	0.4579	0.407
49	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	121.3 KHz 58.3KHz 120.6Khz	xx		IZM2633	562.2	24.2	23.2	5.8	84.8	2575	N/A,1800,N/A	0.4728	0.417
50	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	124.1KHz 58.3KHz 123.4KHz	xx		IZM2633	646.1	23.7	27.3	5.7	84.6	2583	N/A,1800,N/A	0.4712	0.4154
51	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	121.7KHz 57.6KHz 121.3KHz	xx		IZM2633	477.9	23.8	20.1	5.7	85.1	2554	N/A,1800,N/A	0.474	0.4162
52	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	118.5KHz 56.9KHz 118.2KHz	xx		IZM2633	657.1	24.1	27.3	5.8	84.6	2589	N/A,1800,N/A	0.4705	0.4149
Ballast/Lamp Under Test	Description	Fundamental Frequency	FCC 18 Pass	Comments	FCC ID	Lumens	Watts	Efficacy (lm/w)	LER	CRI	Color Temperature OK	Manufactures Rated ouput of lamp (lumens) Source: Osram Sylvania /G.E. Cat.
53	Lights of America, 3-way, 2726, Single Solution (508) 120 V, 60 Hz, 8W/26/16W, 0.4 A max	123.1KHz 58.0KHz 122.4KHz	xx		IZM2633	1188.3	24	49.5	5.8	83.2	2666	N/A,1800,N/A	0.4618	0.4098
54	Philips, PL-s 9/27/SYS, 120 VAC, 11 W			No EMC test	N/A	232.2	9	25.8	2.2	81.6	2838	600	0.448	0.4062
55	LightWave, Model 9826,26 W, self-ballasted lamp 48DF	44.3KHz		No EMC test.	N/A	1625.4	26.5	61.3	6.4	80.9	2811	1825
56	LightWave, Model 9826,26 W, self-ballasted lamp 48DF	44.3KHz		No EMC test.	N/A	1260.8	28.6	44.1	6.9	82.1	2752	1825
57	LightWave, Model 9826,26 W, self-ballasted lamp 48DF	44.3 KHz		No EMC test.	N/A	1212.3	28.67	42.3	6.9	82.6	2686	1825

Objective. The purpose of the total harmonic distortion tests was to determine the power line harmonics from the fundamental frequency of 60 Hz to the 50^th harmonic (3 KHz). Both the voltage and current harmonics were measured. Additional characteristics such as current (Arms), voltage (Vrms), power (W), voltamperes reactive (VAR),power factor (PF), voltage crest factor (CF V), current crest factor (CF C were measured. Total Harmonic Distortion is defined in IEEE Standard 519-1992.

Test Configuration. Testing was performed by connecting each ballast/lamp assembly to North Atlantic 3100 Power Analyzer, serial no. Ph 056, see Figure 2-1. Each lamp was allowed to stabilize and then the data was sent to an HP LaserJet 4 printer because the power analyzer is capable of printing data but is not capable of storing the data. Temperature in the laboratory was 21 ^oC during the majority of the tests. Temperature variation during the tests was approximately 2 ^oC. Lamp-base temperature rise as defined by ANSI C78.25 was not part of the study. The supply voltage applied to the device under test was supplied from the laboratory regulated supply system and then through a Statco Power Supply Type 033-2575.

Results. Power factors (PF) ranged from 0.5 to 1.0 and current crest factors (CF A) over 5 were observed as well as current total harmonic distortion greater than 150%. There are no industry standards that limit or regulate this parameters for Compact Fluorescent Lamp assemblies. As such, each manufacturer sets their parameters based on their own criteria. Test data for this test are contained in Volume 2, Appendix A.

When the data in this study is reviewed in light of studies, ^9,10 it seems to the authors that the electronic ballasts typically used in conjunction with sealed ballast/lamp assemblies utilizing CFLs do not have the desired characteristics that are possible if state of the art ballast designs were used.

The fundamental frequency of operation of the electronic ballasts are shown in Table 1-1. It ranges from 29 KHz to 116KHz. The harmonics associated with the fundamental frequency varied over a wide range. Two examples are given in Figure 2-2 and Figure 2-3. Note in Figure 2-2 that the fundamental frequency for lamp no. 29 was 73 KHz when the lamp was not connected and that the fundamental frequency changed to 31 KHz when the lamp was loaded. The frequency components on lamp no. 29 are clearly distinguishable in the figure; however, for lamp no. 28 shown in Figure 2-3 the fundamental frequency is very difficult to identify because of the large number of nearby harmonics. Broadband emissions of this sort being emitted from the ballast can easily interfere with any nearby electronic equipment. This study did not investigate the effects of these broadband emissions.

Figure 2-1. Total Harmonic Distortion Test Setup.

Some utilities have set limits on Total Harmonic Distortion (THD) for lighting systems at less than 30 %. Some of the ballast/lighting assemblies that were tested exceeded this THD limit.. Current Total Harmonic Distortion (THD) as high as 150% was observed on some of the assemblies. . This can adversely affect the neutral current on branch circuits. Large neutral currents can result in electrical fires; therefore, it is suggested that further studies be conducted to determine the risks of electrical fires caused by the extensive use of compact fluorescent lamps in order to conserve energy. It is not only electronic ballasts that cause this problem but also large groups of computers. A total system approach must be used in analyzing the affects on an electrical distribution system not just the affect of the electronic ballasts alone. At present manufactures do not normally disclose this type of inform on their product data sheets. Lighting system designers can be creating a potential electrical fire hazard if this type of information is not taken into consideration during the initial stages of design.

Figure 2.2. Fundamental Frequency of Operation for Lamp No. 29

Figure 2-3. Fundamental Frequency of operation for Lamp No. 28

Objective. The purpose of these tests were to determine the efficacy, ballast factor (BF), color rendering index (CRI) of ballast/lamp assemblies. Only those systems that were comprised of a sealed ballast lamp assembly were tested for these parameters. In addition the following parameters were measured: color temperature and x-y coordinates on the CIE chromaticity diagram.

Test Configuration. Testing was performed by installing each ballast/lamp assembly into the 5-foot integrating sphere, see Figure 3-1. The lighting assemblies were powered from the regulated laboratory supply through a Statco Power Supply Type 033-2575. After each ballast/lamp assembly was installed in the integrating sphere it was allow to warm-up and stabilize. The light properties of each assembly was measured using a Labsphere spectrophotometer model DAS-1100 connected to the sphere by a fiber optic cable. Electrical power being consumed by each ballast/lamp assembly was monitored by a North Atlantic 3100 Power Analyzer serial no. PG-142

. Output data from the spectrophotometer was transmitted to a computer for analysis. Analysis of the data was performed by Labsphere Spectral Lamp Measurement System Ver. 3.2, 1996 software supplied with the spectrophotometer.

Temperature in the laboratory was between 20 and 22 ^oC during the tests. The integrating sphere had interior temperatures approximately 2 ^oC higher than the laboratory temperature. No atmospheric controls such as temperature and humidity are incorporated into the integrating sphere. Temperature rise in the sphere was minimized by leaving the sphere open between tests.

Calibration. The integrating sphere and spectrophotometer were calibrated using lamp "b" supplied with the spectrophotometer. The calibration data is traceable to the National Institute for Standards Testing (NIST).

Results. The results of the tests are shown in Table 2-1. It should be noted that the ballast factor for the sealed CFL ballast/lamp assemblies were obtained empirically. Most of the lamp assemblies were factory assembled making it very difficult to separate the ballast from the lamp without adversely affecting the functional capabilities of the assemblies nor would it be representative of the assembly. Therefore, in order to determine an estimated ballast factor (BF), which is a useful figure of merit, the total light output from the ballast/lamp assembly was compared with the light output from a factory lamp of equivalent wattage as listed in major lamp manufacturer’s catalogues. Test data for this test are contained in Volume 2, Appendix B.

Figure 3-1. Integrating Sphere and Spectrophotometer Test Setup.

Objectives. The purpose of the starting characteristics/inrush tests was to determine the amount of inrush current that occurs and the time required for the ballast/lamp assemblies to reach a steady state condition.

Test Configuration. Testing was performed by connecting each ballast/lamp assembly to a Philips Model PM3394B 4-channel digital oscilloscope, current sense resistor and computer, see Figure 4-1. The current sense resistor was a 1% one ohm resistor which measured 1.0156 ohms using a Fluke model PM6306 Programmable Automatic RCL meter set to 60 Hz. Triggering the oscilloscope was performed by connecting channel 1 to the load side of the power switch then setting the oscilloscope to trigger on channel 1. Current was measured by connecting channels 3 and 4 inverse across the one ohm current sense resistor. The difference between channels 3 and 4 was displayed on the oscilloscope and transmitted to the computer via an RS232 link and software provided by John Fluke Co.

Results. Inrush currents as high as 3 amperes were observed but they were of very short duration. Currents as high as 70 amps have been reported in the literature¹³ for large groups of fluorescent being switched on simultaneously. Because of the very short rise time of the inrush current very high frequency EMC emissions occurred which were in excess of the steady state values allowed by FCC 18. These are discussed in Section 5 Radiated Emissions Tests. The delay times that occurred before each lamp reached steady state current varied from 28 ms to 4 seconds. Test data for this test are contained in Volume 2, Appendix C.

Figure 4-1. Test Setup for Measuring Starting Characteristics/Inrush Current

Objective. The purpose of the radiated emissions tests are to determine if there are excessive RF signals being broadcast from the device under test into free space. The tests described herein were performed to provide an approximation of EMI compliance of RF lighting devices currently on the market in the State of California. Certification that a device complies with FCC regulations can only be provided by an FCC Certified laboratory. RF Lighting devices are required to meet the conducted and radiated emissions requirements of FCC Part 18 for RF lighting devices. Forty-four lamps were tested to a pre-compliance level only. Table 5-1 provides a list of the RF lighting devices tested for EMI pre-compliance and provides a summary of the radiated and conducted emission test results.

Lamp Number	Description	Fundamental Frequency (kHz)	Radiated Emissions 30 MHz to 300 MHz	Conducted Emissions 30 MHz to 300 MHz
1	Sylvania Delux Electronic 15 W	41.5	Pass	Pass
2	Lights of America 2342 42W, 0.55 A	43.4	Pass	Fail
3	Sylvania 18 W, 0.3 A	32.4	Pass	Pass
4	Sylvania 18 W, 0.3 A	29.1	Pass	Pass
5	Fulham 15 W, pf > 0.95	-	Pass	Fail
6	Lights of America FC8T9/WW/RS	-	Pass	Fail
7	Genura 23W	-	Pass	Fail
8	Philips SLS 11, 11 W	49.1	Pass	Not Tested
9	Sylvania 20W	41.9	Pass	Pass
10	FEIT Electric 18 W	-	Pass	Fail
11	Sylvania 15 W	43.4	Pass	Fail
12	Sylvania 11W	35.7	Pass	Pass
13	Sylvania 15 W	37.4	Pass	Pass
14	Philips SLS 11, 11 W	49.1	Pass	Pass
15	Philips SLS 9, 9 W	49.6	Pass	Pass
16	Osram 23 W	40.3	Pass	Pass
17	Motorola Gold Edition Electronic Ballast, Rapid Start, M1-CF-32-B-S-120 for 32W & 26 W, with Philips 26W/30/4P lamp	32.3	Pass	Fail
18	Lights of America,2726 Single Solution 488 (3-way) 8w/26/w/16w, 0.4 a	118.2	Pass	Fail
19	Lights of America,2726 Single Solution 488 (3-way) 8w/26w/16W, 0.4 A	56.8	Pass	Pass
20	Lights of America,2332, 448, 32 W, 0.45 A	42.2	Pass	Pass
21	Lights of America,2332, 448, 32 W, 0.45 A	41.3	Pass	Pass
22	Lights of America 2127 IC (248), 27W, 0.43A	42.2	Pass	Fail
23	Lights of America 2127 IC (248), 27W, 0.43A	48.4	Pass	Pass

Lamp Number	Description	Fundamental Frequency (kHz)	Radiated Emissions 30 MHz to 300 MHz	Conducted Emissions 30 MHz to 300 MHz
24	Lights of America FCL30EX-L Lamp,2630, 30 W, 0.46 A ballast	39.6	Pass	Pass
25	Intelect Plus, PUV-T13RS-EOL, ballast for 1 lamp CFM 26W/32W/42W, 4 pin		Pass	Fail
26	Intelect Plus, PUV-T13RS-EOL, ballast for 1 lamp CFM 26W/32W/42W, 4 pin		Pass	Fail
27	Intelect Plus, PUV-T13RS-EOL, ballast for 1or 2 lamp CFQ26W, lamp CFM26W,CFT27W and 28W 2D		Pass	Fail
28	ES Super LampGard, ES-2-CFQ-26-120-G-LG, 2-Lamp, 26W, compact 4pin, quad		Pass	Fail
29	Motorola Gold Edition Electronic Ballast, Rapid Start, M1-CF-32-B-S-120 for 32W & 26 W, for 1 lamp 4pin CFL 32W & 26W		Pass	Fail
30	Motorola Gold Edition Electronic Ballast, Rapid Start, M1-CF-32-B-S-120 for 32W & 26 W, for 1 lamp 4pin CFL 32W & 26W		Pass	Non Operating
31	Motorola Gold Edition Electronic Ballast, Rapid Start, M2-CF-26-B-S-120 for 2 lamp CFL 26 W 4pin		Pass	Non Operating
32	Motorola Gold Edition Electronic Ballast, Rapid Start, M2-CF-26-B-S-120 for 2 lamp CFL 26 W 4pin		Pass	Non Operating
33	Motorola, M2-IN-T8-D-120, 2 Lamp, F32T8,F25T8,F17T18		Pass	Fail
34	Motorola, M2-IN-T8-GP-D-120, 2 Lamp, F32T8,F25T8,F17T19		Pass	Fail
35	Motorola, M2-RN-T8-ILL-D-120, 2 Lamp, F32T8,F25T8,F17T20		Pass	Pass
36	Lights of America, 2630, 30W, 0.25 A		Pass	Pass
37	Intelect Plus, PUV-22RS-EOL, ballast for 1 or 2 lamp CFQ26W, CFM26W, CFT27W, and 28W 2D		Pass	Fail
38	Advance Transformer, Mark X, REZ-1T42,for FM42W lamp, 0.42 A		Pass	Pass
39	Advance Transformer, Mark V, RIC-2S32, for (2) F32T8 or (2) F25T8 lamps, 0.54 A		Pass	Pass
40	Advance Transformer, Mark V, RIC-2S32, for (2) F32T8 or (2) F25T8 lamps, 0.54 A		Pass	Fail
41	Advance Transformer, Mark X, REZ-1T42,for FM42W lamp, 0.42 A		Pass	Pass
42	Intelect, PUV-T13RS, for (1) 26W/32W/42W, RS triple tube or (1) 32 W heliax CF Lamp		Pass	Fail
43	Intelect, PUV-T13RS, for (1) 26W/32W/42W, RS triple tube or (1) 32 W heliax CF Lamp		Pass	Fail
44a			Pass	Fail
44b			Pass	Pass
44c			Pass	Fail

Test Configuration. The tests were performed in an open-air test site located in Phelan California using the procedure described in Hewlett Packard Application Note 1290-1, Cookbook for EMC Pre-compliance Measurements.. The test configuration is shown in Figure 5-1. The test equipment used is listed in Table 5-1.Correction Factors used for the Cable, Preamplifier, Bi-conical Antenna, and Log periodic Antenna are listed in Tables 5-3. through Table 5-6.respectively. HP provided the calibrated correction factors as part of the HP 84110EM EMC Preproduction Evaluation System.. Photos of the Test Site. Are shown in figures 5-2 through 5-7. The tests were performed in accordance with the Electro Magnetic Interference (EMI) Test Procedure For the performance of Radiated and Conducted Emissions.

Equipment Name	Model
EMC Analyzer (S/N 1305)	HP8590EM
Biconical Antenna	HP 11955A
Preamplifier	HP8447F
Instrumentation computer	IBM Compatible
Report Generator Software	HP85878A

Requirements. The requirements for RF Lighting devices are defined in FCC Part 18.305 (c) for Industrial, Scientific, and Medical (ISM) consumer equipment. Paragraph 2.2.2 of the appendix to FCC Part 18 defines the IF bandwidth as not less than 100kHz for measurements between 30 MHz and 1000 MHz. Table 5-7 defines the RF field Strength limits referenced to a measurement distance of 30 meters.

Frequency Range (MHz)	Field Strength Limit (mV/m)
30 to 88	10
88 to 216	15
216 to 1000	20

Results. The Results of the Radiated Emission tests are provided in Appendix D. The plots of appendix D provide only Vertical polarization results; however, periodic checks in the Horizontal polarization did not indicate the presence of any emissions above the levels shown in Appendix D. Radiated emission tests in the range of 300 MHz to 1 GHz were not performed due to the excessive ambient noise present. Test data for this test are contained in Volume 2, Appendix D.

Conclusion. The devices tested passed the requirements of in FCC Part 18.305 (c) for Industrial, Scientific, and Medical (ISM) consumer equipment. No emissions exceeding the defined requirements were detected. The fundamental frequencies of the devices tested (Table 5-1) were low (29 kHz to 118 kHz) compared to measurement range of 30 MHz to 300 MHz. No out of tolerance conditions in the range of 30 MHz to 300 MHz were observed; therefore, based upon analysis, it is not expected that any out of tolerance conditions would exist as a result of the fundamental frequencies in the region of 300 MHz to 1 GHz. This conclusion is based upon the spectral roll off of a rectangular pulse (worst case) using equation 5-1. Based upon a fundamental frequency of 120 kHz, the magnitude of harmonics at 30 MHz will be less than -50 dBc; harmonics at 300 MHz will be less than -70 dBc. Therefore, if out of tolerance conditions were not observed in the range of 30 MHz to 300 MHz, then no out of tolerance conditions in the range of 300 MHz to 1 GHz would be expected to be observed either.

Figure 5-1. Test Setup for Measuring the Fundamental Operating Frequency

Ojective. The purpose of the conducted emissions tests was to determine if there are excessive RF signals being conducted from the device under test into the power line.

Test Configuration. The tests were performed by connecting the ballast/lamp assembly to a Hewlett Packard line impedance stabilization network (LISN) model HP11967D as shown in Figure 6-1. The test configuration is shown in Figure 6-1. The test equipment used is listed in Table 6-1. Correction Factors used for the Line Impedance Stabilization Network (LISN) and Transient Limiter are listed in Table 6-2 and Table 6-3, respectively. HP provided the calibrated correction factors as part of the HP 84110EM EMC Preproduction Evaluation System. The tests were performed in accordance with the Electro Magnetic Interference (EMI) Test Procedure For the performance of Radiated and Conducted Emissions as provided by Hewlett Packard.

Equipment Name	Model
EMC Analyzer (S/N 1305)	HP8591EM
Transient Limiter	HP11947A
Line Impedance Stabilization Network (LISN)	HP11967D
Instrumentation computer	IBM Compatible
Report Generator Software	HP85878A

Requirements. The Conducted Emission requirements for RF Lighting devices are defined in FCC Part 18.307 (c) for Industrial, Scientific, and Medical (ISM) consumer equipment. The Conducted Emission requirement is that no conducted emissions exceed 250 mV over the frequency range of 0.45 MHz to 30 MHz using a 50mH/50W LISN. As can be seen in the test setup, the LISN used was a 250mH/50W LISN. The larger impedance of the LISN may have adversely effected the measurements resulting in larger peak values. Also, the actual physical layout of test equipment deviated from ANSI C63.4 recommended layout. However, during various test runs, changing the location of the equipment did not appear to have an appreciable effect on the measured result.

Results. Each device has two data sheets, one records the conducted emissions on power line 1 (L1), the second records the conducted emissions on power line 2 (L2). Results for Lamps number 30, 31, and 32 were not obtained because these lamps would not light at the time of the test. Prior to measuring the device conducted emissions, an ambient measurement was performed. The Ambient measurement (equipment powered off) was approximately 10 dB below the test limits, surpassing the recommended ambient margin of 6 dB minimum. As can be seen in the plots, broadband noise marginally exceeds the FCC limit in twenty-one of the forty-four RF Lighting devices tested. An error exists on several data sheets, the Test Name erroneously indicates that these tests are "Ambient" tests. The Test Name should read Conducted Emissions not Conducted Ambient Emissions for all data sheets containing a lamp number in the Title. Twenty-one of the forty-four RF lighting devices tested failed to comply with the requirements of FCC Part 18.307 (c) for industrial, scientific, and medical (ISM) consumer equipment. Three devices could not be tested due to device failure. A number of the ballast/lamp assemblies failed the conducted emission test. There are some deviations from recommended methodology and equipment used which should be recognized in assessing any given ballast/lamp assembly failure. There was an ambient background of 10 dB less than the test limits and the recommended in 6 dB. There may be some impedance contributions because the LISN devise in the laboratory is a 250 mH/50W LISN and the FCC specifies that a 50 mH/50W LISN device be used for the tests. Test data for this test are contained in Volume 2, Appendix E.

Limitations. The effect of conducted emission upon residential appliances that are controlled by or use microprocessor was not investigated. Because a large number of the lamp/ballast combination failed the conducted emissions test, future tests should be undertaken to determine the effects of conducted emissions upon home security systems and electronic based appliances such as microwave ovens, washing machines and dryers. The effect of temperature on the lamp/ballast combination was not undertaken because the environmental chamber is still under construction.

Dwyer, Roy, Evaluation of Harmonic Impacts from Compact Fluorescent Lights on Distribution Systems, IEEE Transactions on Power Systems v. 10 (Nov. ’95) p1772-80.

 

Heald, Tom, Phantom of the Ether, Electronics Now v. 66 (Jan. 95) p53-7.

 

McKinnon, Barry, Electromagnetic Interference in Hearing Aid T-Coil Applications, CAA Conference in Ottawa, Canada, 1994.

 

Roche, W. J., High-Temperature Behavior of Compact Fluorescent Lamps, Journal of the Illumination Engineering Society, v. 22 (Winter ’93), p97-102.

 

Collins, B.L., Performance of Compact Fluorescent Lamps at Different Ambient Temperatures, Journal of Illuminating Engineering Society, v. 23 (Summer ’94) p 72-84.

 

Serres, A.W., On the Photometry of Integrated Compact Fluorescent Lamps, Journal of the Illuminating Engineering Society, v. 24 (Winter ’95) p58-62.

 

Embriz-Santander, E, A, Comprehensive Harmonic Study of Electronic Ballasts and their Effect on a Utility’s 12 kV, 10 MVA Feeder, IEEE Transactions on Power Delivery v. 10 (July ’95) p 1591-9.

 

Chen, W., Documenting the Effects of Luminaire Venting on CFL Downlight Performance, Journal of the Illuminating Engineering Society, v. 26 (Winter ’97) p. 115-24.

 

Shimizu, K., Electronic Ballast Circuit for Fluorescent Lamps that Reduces Circuit Harmonics, Journal of the Illuminating Engineering Society, V. 26 (Summer ’97) ; 26-31.

 

Qian, Jinrong, Analysis, Design, and Experiments of a High-Power Factor Electronic Ballast, IEEE Transactions on Industry Applications v. 34 no.3 (May/June ’98) p616-24.

 

Ji, Yunfen, Starting Performance of High-Frequency Electronic Ballasts for four-foot Fluorescent Lamps, IEEE Transaction on Industry Application v. 33 (Jan/Feb ’97) p 234-8.

 

Katoaka, Shozo, Preventing IR Interference between Infrared Waves Emitted by High-Frequency Fluorescent Lighting Systems and Infrared Remote Controls, IEEE Transactions on Industry Applications v. 33 (Jan/Feb ’97) p239-45.

 

Ed Shen, DMM: Demystifying Maintenance Maladies, EC&M, April 1999, p 52-56.