wwyd about gsfl?

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

WWYD about GSFL?

PRESENTED TO THE REGIONAL TECHNICAL FORUMCarrie Cobb, BPA

Rob Carmichael and Laura Tabor, Navigant ConsultingJuly 16, 2013

A update on BPA’s non-residential lighting market research

*

*Yoda Eckman


2


It’s about Baselines

3

WHERE WE ARE WHERE WE ARE GOING

Pre-existing

condition

Current

Practice-“what’s in the ceiling”: pre-condition to post-condition -Reflects the choices a typical

consumer makes

ONLY GSFL!!!

“As a general rule, the RTF will use a baseline that is characterized by current market practice or the minimum requirements of applicable codes or standards, whichever is more efficient.”


4


THE CHALLENGE… THE METHODS…

THE CONSIDERATIONS…

5


Research ChallengeIssue:

What is the baseline for GSFL (general service fluorescent lamps—AKA linear fluorescent) in the non-residential sectors?

Why it’s important: To ensure that, in light of the DOE standard, we continue to pay for only above-standard energy savings, but still capture cost-effective savings opportunities.

Why it wasn’t straightforward:This ain’t your grandpa’s incandescent wattagestandard…let’s look at the complications.

6


7

There is no maximum allowable wattage per lamp.

The ballast matters (a lot).

Nearly endless GSFL fixture permutations.

How will manufacturers/distributors/contractors respond to new efficacy standard?

Got data?

Uses Cases

Different users will have different perspectives and needs for expressing the baseline.

Commercial Lighting Challenges

Regulatory and

Technical

DOE standard is an efficacy requirement (lumens/Watt) for a single lamp. A brighter lamp at the same (or greater) power is fair game.

Lamps live in fixtures, paired with ballasts, which affect system performance. The standard applies only to lamps.

Market Content

Programmatically, the building code metric, LPD, is rarely helpful in the context of incentivizing fixture retrofits.

Number and mix of lamp installations is a much different question than number and mix of fixtures installations.


8

Scope of Perspective

Lighting Analog

Utility Baseline Typically Expressed in:

Code or Standard Typically Expressed in:

Programmatic Application:

Super-system Building or Space Watts/Sqft (LPD) Watts/Sqft (LPD) New Construction

Programs

System Fixture Watts Lumens/Fixture Watts* Lighting Retrofit Programs

Sub-system Lamp Watts Lumens/Lamp Watts Upstream Programs

Sub-system Ballast Watts (Output Watts / Input Watts) Upstream Programs

Perspectives on Commercial Lighting Baseline


Data Collection

9

Interviewed 20 distributors; in-depth, 90-min interviews. • All NW states represented; some rural, some

urban, some small, some large.• Biggest differentiator among distributors is not

size/location, but market niche. Received complete 2010–2012 GSFL

and HID sales data from 11 distributors. Expect at least 2 more.

Captured an estimated 45-50% of the market distributor non-residential market.


Further Data Collection and Analysis

10

Interviewed 3 manufacturers Reviewed national sales data Reviewed channel-specific (retail, wholesale, etc.)

national data Market insights further supported by Navigant’s

research in support of previous GSFL and ballast rulemakings as well as ongoing GSFL DOE rulemaking


FINDINGS

11


12

T12 Lamp Sales ( in thousands )


13


Not only T12 sales plummeting…

Many distributors plan to stop stocking T-12s next year

Remaining sales of T-12s are 100% maintenance market

Big-box stores (non-distributor channels) are not selling significant volume of T-12s to C&I market

Retailers/manufacturers not interested in maintaining SKUs14


15


16


17

80% T-8


18

The market is DOMINATED by…


No matter how you slice it…

Four Foot T8/T12 Applications

Market SegmentAverage

Wattage Per Lamp*

Watts Above/Below

AverageNew Construction 28.7 -1.0Maintenance 30.9 +1.3Retrofit 28.2 -1.5All Applications 29.7 -

19

*8ft lamps scaled to approximate number of 4ft replacements


What about the existing T-12 systems in buildings?

T-12s are out there• ~25% in 2009 per CBSA• Data from current CBSA available within the

year• Saturation expected to

be lower due to big program push

• Likely in low operating hour profiles

20


When a 40W T12 Burns Out…

21

≤10% 20% 30% 40% 50% 60% 70% 80%0

1

2

3

4

5

6

7

Percent of Customers Purchasing New T12s Resp

onde

nts

(n =

12)

75% would upgrade to T8 (majority), T5, or LED.

On average, distributors estimates that only 25% of customers would install new compliant or stockpiled T12s.

Furthermore, recall lamp burnout represents the most likely scenario of a T12 being chosen.


Defining the market

The standard is working and covers all GSFL, not just one type• Baseline for GSFL; not just T-12s• Market Average = retrofit + maintenance +

new construction

22


BPA’s new baseline

23

Lighting Systems Categories

Nominal Lamp

Wattages in MML

Existing Average Lumens

Application's Typical Lamp

Wattage

Market Average System

Input Per Lamp

# of Lamps in Existing System

Market Average Fixture

Adjustment (AKA "old ballast

factor")

New Baseline System Watts

Lamps Included in

Average

All 4' T-12 34

All 4' T-12 40

All Slimline 8' T12 60

All Slimline 8' T12 75

All HO 8' T12 95

All HO 8' T12 110

All VHO 8' T12 185

All VHO 8' T12 215

4ft T84ft T12

4ft T84ft T128ft slim

5729 95 191.8 0.97 91.8 8FTHO4ftT5HO

4625 60

8FTHO4ftT5HO134.10.72134.18370 185 1

New Current Practice (Market Average) BaselineTypical Existing Condition / Obsolete Equipment

52.90.8852.9 1

1984 34 129.4 0.86 29.4


THINKING AHEAD:THE NEW MATH

24


ANSI Testing Standard

ANSI is in process of changing standard New DOE ballast efficiency metric in 2014 Topic for consideration in lighting protocol

when subcommittee meets again Not explicitly being incorporated into the

baseline BPA will implement

25


Background on Lamp Power DrawDrivers of GSFL Fixture Power Consumption

Lamp Power Number of Lamps Ballast Efficiency Ballast Factor

26

These are actually two separate, independent*

factors. The regional calculators put them into

one bucket called “ballast factor.”

*For all intents and purposes of this project. In reality, one can affect the other.

Nominal (“rated”) lamp power is NOT the same as

actual power draw from the lamp. And the reason

it’s different is NOT ballast factor.

The actual reason it is different is an artifact of the way manufacturers are required to test and report

performance to DOE AND the fact that most lamps now operate on electronic ballasts.


Background on Lamp Power Draw (Cont.)

Definitions

Nominal Lamp Power(32W for the standard T8 lamp)

This is simply what manufacturers market their lamps at. It is a category. More like a “Venti” than “20oz.”

ANSI Lamp Arc Power at HF(29.0W for the same T8 lamp)

Actual lamp draw when operated at High Frequency (more appropriate because most lamps will run on electronic ballasts now in the commercial sector).

27

Okay, so which one is ‘correct’?

Both are ‘correct’ in the context in which they are reported, but for most linear fluorescents lamps today, the actual lamp draw is

the ANSI Lamp Arc Power tested at High Frequency because most commercial lamps will run on electronic ballasts.

𝑺𝒚𝒔𝒕𝒆𝒎𝑰𝒏𝒑𝒖𝒕 𝑷𝒐𝒘𝒆𝒓 (𝑾 )=(¿𝑳𝒂𝒎𝒑𝒔×𝒃𝒂𝒍𝒍𝒂𝒔𝒕 𝒇𝒂𝒄𝒕𝒐𝒓 × 𝑨𝑵𝑺𝑰 𝑳𝒂𝒎𝒑 𝑨𝒓𝒄 𝑷𝒐𝒘𝒆𝒓 )𝑩𝒂𝒍𝒍𝒂𝒔𝒕 𝑳𝒖𝒎𝒊𝒏𝒐𝒖𝒔 𝑬𝒇𝒇𝒊𝒄𝒊𝒆𝒏𝒄𝒚


Real Magnitude of this “Artifact”

28

Ballast Type

Lamp Nominal Wattage

ANSI Lamp Wattage on

Magnetic Ballast

Magnetic Delta from

Nominal

ANSI Lamp Wattage

Electronic Ballast

Electronic Delta from

Nominal T12 Electronic 34 32.0 -5.9% 29.8 -12.3%T12 Electronic 40 39.0 -2.5% 36.3 -9.2%T8 Electronic 32 30.8 -3.8% 29.0 -9.4%T8 Electronic 25 NA NA 24.2 -3.2%T8 Electronic 28 NA NA 26.0 -7.1%T8 Electronic 30 NA NA 27.5 -8.3%T8 Magnetic 32 30.8 -3.8% 29.0 -9.4%T12 Slimline Electronic 60 60.5 0.8% 57.4 -4.3%T12 Slimline Electronic 75 75.0 0.0% 71.2 -5.1%T8 Slimline Electronic 59 60.1 1.9% 57.0 -3.4%T12 High Output Electronic 110 106.0 -3.6% 100.0 -9.1%T12 High Output Electronic 95 90.0 -5.3% 84.9 -10.7%T8 High Output Electronic 86 NA NA 84.0 -2.3%TSSO Electronic 28 NA NA 27.8 -0.7%T5HO Electronic 54 NA NA 53.8 -0.4%


Back Up Slides

29


30

Weight by market distribution of number of

lamps per fixture

Weight by market distribution of ballast types

for each lamp type

Professional Judgment

DOE Rulemaking Data on Lamps/

Fixture

BASELINE WATTS: New Construction

MaintenanceRetrofit

Weight by market distribution of lamp

types for each baseline category

Initial & Mean Rated Lumens

ANSI Lamp Wattage

LAMP CHARACTERISTICS

100+ Linear Fluorescent Permutations50+ HID Permutations

Ballast Type

Lamp Type

Number of Lamps per Fixture

AVERAGE FIXTURE WATTS

50 Linear Fluorescent Permutations50+ HID Permutations

Ballast Type

Lamp Type

AVERAGE SYSTEM WATTS

~20 Linear Fluorescent Types~25 HID Types

Lamp Type

AVERAGE LAMP WATTS

Ballast Factor

Ballast Efficiency

BALLAST CHARACTERISTICS


Distributor Ballast Shipment

Data

GOAL: Roll detailed sub-type data up to a single representative value for each purchase event that can be used as an input in the baseline calculation. As shown, the resulting baseline watts value reflects the role of ballasts on overall lamp-ballast system wattage.

Distributor Lamp Shipment Data


Market Average Wattage Calculation


31

Obsolete

Obsolete Equipment. Equipment is considered obsolete if either of the following apply:• Components of the Efficient Case system are longer

available in the market due to the equipment no longer being manufactured; or

• Regulatory requirements, such locally applicable codes and standards for equipment or equipment standards promulgated by the U.S. Department of Energy prohibit sales of the equipment necessary for the Efficient Case system to operate.


Characteristics of Results

32

• The baseline wattages reflect the market-weighted system input wattage per lamp for the

fluorescent lamps shown to the right. (Nominal overall average is 31W/lamp).

• The lamp baselines account for the average ballast and system characteristics.

• The results show the dominance of the 32 W T8 lamp: there isn’t a big difference no matter

how you slice it.

• These values can be easily rolled into any system configuration.


text

Market Measure Applications

New Construction,

Addition

New Building

New Space (Addition)

Renovation or Remodel

Scheduled Regular Maintenance

“Onesie-twosie Maintenance”

(Equipment Failure)

Old system no longer satisfactory

Cannot purchase old system components

anymore

Energy efficiency-motivated upgrade

Purchase Triggers Programmatic Measure Applications

New Construction

Maintenance

Retrofit

100%

50%

100%

50%

RTF Baseline Condition

Current Practice

Current Practice

Current Practice

OR

Pre-Condition

10%

50%

90%

100%

Baseline Perspective (Building or Fixture/Lamp) & Metric

Perspective = Building

Metric = LPD

Perspective = Fixture/Lamp

Metric = Watts

Perspective = FixtureMetric = Fixture Watts

Basis for the Baseline

Maximum code LPA, or actual current practice as

determined by code compliance work

Average Unit Shipped in Maintenance Market

Average Unit Shipped in Retrofit Market

OR Average Pre-Condition

Major Renovation

Group Relamp and Reballast

Group Relamp Only

50%

Lamp Replace on Failure

Ballast Replace on

Failure

System Upgrade

90%

10%

Perspective = Fixture/Lamp

Metric = Watts

Average Unit Shipped in New Construction Market

Perspective = Building/Space

Metric = LPD

Average LPD of retrofitted spaces (Current Practice)

ORAverage Pre-condition LPA


Caption Contest

34

“This is your pre-condition baseline and

here’s your current practice.”

“So say the T12 represents the 6th

Plan

potential and T8 is non-

programmatic…”

wwyd about gsfl?

Documents

t i o nfindings11b o

t i o nnot

t i o nits

t i o nin

t i o n2 b o n n e v

t i o n67 b o n n e

t i o n8 b o n n e v

considerations5b o n