Measuring Duct Leakage
using the
Generalized
Subtraction Correction Algorithm


Myron Katz,   Norman Witriol  &  Jinson Erinjeri

 

Feb 26, 2006                                 Copyright Myron Katz

 


 

Acknowledgements

 

This talk is an outgrowth of a research project by the Trenchless Technology Center, Louisiana Tech University, Ruston.  One objective was to investigate and characterize residential duct leakage.

 

    We gratefully acknowledge the Louisiana Department of Natural Resources (under a US DOE grant) for its sponsorship of this research, DNR Interagency agreement No. 2030-04-03  and the valuable support of the DNR Energy Section staff, in particular Paula Ridgeway, Wade Byrd, Harvey Landry, Buddy Justice, Howard Hershberg, and Tangular Williams.


 

Introduction

 

  • Generalized Subtraction Correction Algorithm (GSCA) is an improvement and extension of Modified Subtraction algorithm presented in Minneapolis Blower Door Operation Manual. 

Introduction

 

•        Accurate use of GSCA depends upon automated data collection using TectiteTM to get two, very accurate whole-house leakiness tests

 

•        The presentation assumes audience familiarity with procedures available at www.EnergyConservatory.com


Introduction

 

•        GSCA used for duct leakage testing in: Testing HVAC Duct Leakage in Existing Residential Buildings in North Louisiana, August 2003.

 

•        GSCA’s mathematical model was published in ASTM’s, Journal of Testing and Evaluation, Nov., 2004.

 

 

 


 

GSCA can be performed in 3 modes depending upon available equipment:

 

Fully Automated:  computer controlled APT

                      with 4 or more pressure channels

Semi-Automated: computer controlled

                      DG700 or APT with at least 2

                      pressure channels

Fully Manual: digital manometer – either a

                      DG3 or DG700 used manually

 

This talk will explain Semi-Automated via DG700.


 

Beyond Modified Subtraction

 

Generalized Subtraction Correction Algorithm (GSCA) only requires three extra manual data collection steps; these are shown in blue.

 

To greatly enhance GSCA’s accuracy, we recommend adding six more steps; these are highlighted in red.

 


 

GSCA Procedure in a Nutshell

 

Do pressure-pan tests to choose a sample-supply-register closest to average reading.

 

1.     Depressurize home to pressure P.  Record:

    House-Leakiness Flow as Q.

    Pressure in the Attic with respect to inside, PA.

    Pressure in the Ducts, at the sample-supply-register, with respect to the attic, PD.

 

2. Repeat Step 1 at 8 pressures, P, from 15 to 50 Pa.

 

3. Tape registers, repeat 1 & 2 to record , A & D.

 

 


 

Why another procedure?

 

•        More Energy Savings

•        More accuracy

•        Less time

•        Less expensive equipment

•        More useful diagnostic results

•        Automatic paper trail

•        Better fit to Existing and New construction

•        Reveals data collection errors in REAL TIME

•        Measures infiltration between home and attic

•        Improves quality and quality assurance


 

Why another procedure?

 

More Energy Savings!

Very high precision duct leakage testing improves HVAC installation

   If duct leakage tests are accurate to 5 CFM25, homeowners can demand a challenging but attainable level of duct leakage (such as less than 5% duct-leakage-to-outside as in the Florida Energy Code) in their contract with HVAC installer.   Compared to 29% duct leakage published in our study, this could result in more than 20% higher effective SEER.


 

Why another procedure?

 

Whole-House Leakiness more accurate if:

 

•        Reports as Effective Leakage Area

•        Avoids  “Can’t-Reach-Fifty-Factor”

•        Uses industry-standard multi-pressure tests

•        Calculates flow exponent for that home

•        Uses baseline pressure measurements

•        Uses automatically collected data

•        Uses hundreds of data collections

•        Uses air temperature measurements

•        Estimates its error and data consistency

 

ALL OF ABOVE AUTOMATED BY TECTITE 


 

Why another procedure?

 

Duct Leakage more accurate if

 

•        Derives from two, very accurate, whole-house leakiness tests

•        Utilizes two attic pressure readings

•        Uses duct-leakage flow-exponents

•        Utilizes pressure-pan tests

•        Estimates its own error and data consistency

 

GSCA boasts all five of these.


 

Why another procedure?

 

More accuracy –

 

ANCIENT HISTORY

 

   The Modified Subtraction algorithm has “an average uncertainty of 27% for total duct leakage and 53% for either supply or return leakage…”  

    Mark Modera, Field Comparison of Alternative Techiques for Measuring Air Distribution System Leakage, ASTM STP 1255, 1995.


 

Why another procedure?

 

More accuracy

  

A 2002 Ecotope study by Francisco, Palmiter & Davis demonstrated very careful data collection using Modified Subtraction produced very similar results to tests with Duct Blaster assisted by blower door.

 

The following is a chart from their study.



 

Why another procedure?

 

More accuracy

   Estimate Duct leakage using actual pressure between ducts and the volume that contains them.  For our clients, this is usually the attic.  However, attic pressure is not even measured in the Modified Subtraction algorithm!

 

The following chart displays attic pressure within homes tested by Myron Katz during 2000-2003.



 

Calculated Duct Leakage
using
Outside vs. Attic Pressure

 

Attic Pressure

with respect to House (Pa)

Modified Subtraction vs. GSCA

Mean % Error

47 - 50

3

45 - 47

9

43 - 45

15

40 - 43

22

30 - 40

28

 


 

Why another procedure?

 

Less time required for testing

•        Avoids errors in data collection

•        Fewer setups for house and duct leakage

•        Automatically zero meters

•        Tape masks falling off leave warnings

•        Most data collection & storage are automatic

Although these are extra

•        Pressure-pan tests should be done anyway

•        Attic pressure tests reveal infiltration pathways


 

Why another procedure?

 

Less expensive equipment

 

•        Blower door APT with 4 or more pressure-channels (most automated)

             Est. Cost: $2500.

 

•        DG700 (collects 99.8% of same data – less automated)  

             Est. Cost: $750

 

•        Duct Blaster (if GSCA not used)

             Est. Cost: $1000


 

Why another procedure?

 

More useful diagnostic results:

 

•        Energy savings (as explained above)

•        Pressure Pan Tests guide duct-repair work

•        Attic pressure tests inform weatherization work

•        Baseline data identify stack effects

•        First steps to Zone Pressure Diagnostics

•        Data collection for future testing refinements


 

Why another procedure?

 

Quality & Quality Assurance

 

•        Announces data collection errors in REAL TIME

•        Reviewing Graph Screen can uncover data collection errors.

•        Tectite creates an automatic paper trail which improves quality assurance

 


 

Clearest Benefits of GSCA

 

•        GSCA will recognize wide range of data collection errors -- all easily reparable in real time.

•        GSCA output is exceptionally consistent with properly performed Duct Blaster tests and outperforms Duct Blaster in some situations.

•        GSCA requires less equipment

•        GSCA able to consistently detect 2% changes in Duct Leakage

•        GSCA significantly enhances accuracy of whole-house leakiness testing


Limitations of GSCA

 

Just like modified subtraction:

•        GSCA can only be used on a home with a single HVAC system.

•        Parts of duct system not in  conditioned volume must be in single pressure volume ― i.e., in a single attic or in a single crawlspace

 


 

Advantages of GSCA vs Duct
 Blaster assisted by Blower Door

 

•        Less equipment needed

•        Faster setup

•        Unaffected by clogged evaporator coil

•        Same effort leads to more accurate  

whole-house leakiness measurements

•        Enhanced knowledge of house infiltration

•        Almost fully automated & paper trail


GSCA Procedure in a Nutshell

 

Do pressure-pan tests to choose a sample-supply-register closest to average reading.

 

2.     Depressurize home to pressure P.  Record:

    House-Leakiness Flow as Q.

    Pressure in the Attic with respect to inside, PA.

    Pressure in the Ducts, at the sample-supply-register, with respect to the attic, PD.

 

2. Repeat Step 1 at 8 pressures, P, from 15 to 50 Pa.

 

3. Tape registers, repeat 1 & 2 to record , A & D.


 

Thank You J

 

Questions?

 

Step-by-Step Next…


 

STEP-BY-STEP Procedure

 

1. Set up computer and record HVAC register names

2. Place a pressure probe in the attic

3. Set up blower-door hardware

4. Set up Tectite software & measure temperature

5. Measure home’s baseline pressure and cruise

6. Find sample-supply-register via pressure-pan tests

7. Setup data collection at sample-supply-register

8. Measure pressures in attic & at sample-supply-register

9. Depressurize at a series of pressures and save data

10. Copy whole-house leakiness data to calculator

11. Check data’s consistency, accuracy, etc

12. Tape registers and repeat steps 8 – 11

13. Calculate Duct Leakage and read its accuracy

14. Enter House & Duct Leakiness into RemRate

 


 

1. Set up computer & record register names

 

•        Set up Laptop near intended blower-door installation.

•        Copy Tectite files and associated Duct Leakage Calculator file into a new folder named to identify this client.

•        With homeowner’s help, identify & name every supply and return HVAC register.

•        Open “Duct Leakage Calculator.xls” & record HVAC register names in column D in the spreadsheet on rows 13 – 33.

 


 

1. Set up computer & record register names

 

Set up Laptop near intended blower-door installation.


 


1. Set up computer & record register names

 

Copy Tectite files and Duct Leakage Calculator into a new folder named to identify this client.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 


1. Set up computer & record register names

 

With the homeowner’s help, identify and name every supply and return HVAC register.

                  

1. Set up computer & record register names

 

Open “Duct Leakage Calculator.xls” & record HVAC register names in column D in the spreadsheet on rows 13 – 33.

 

 


2. Place a pressure probe in the attic

 

Into Attic, place a Spool  or           Tube.

 

     

 


 

3. Set up blower-door hardware

 

Use normal procedure

    a. Connect pressure probes and sensors to controllers, fan and outside.

     i. Connect GREEN tube, i.e., the outside

        pressure-sensing tube to

          REFERENCE of 1st pressure channel

     ii. Connect RED tube, i.e., the fan-speed

         pressure-sensor of fan to

         INPUT of 2nd pressure channel

b. With a 9-pin communications cable, connect computer to DG700.


 

3. Set up blower-door hardware

 

Connect GREEN tube to outside pressure-sensing tube.            

Connect RED tube to fan-speed pressure-sensor of fan.

 


3. Set up blower-door hardware

 

Connect computer to DG700.

 


4. Set up Tectite software

 

  1. Start Tectite by clicking on “Cruise.bld” icon.
  2. Skip Customer Info & Building Info screens.
  3. In Climate Info Screen  

                   Measure and Input current Temperatures.

  1. Skip Comments screen.
  2. In Test Settings screen use default settings.

                   (Auto, CGSB pressures, no active zones.)

  1. In Test Graph screen accept Auto Mode.

 

    Turn on DG700.


 

4. Set up Tectite software

 

a.       Double click “Cruise.bld” file to start Tectite.

b.  Skip Customer Info & Building Info screens

c.  Climate Information Screen:

                                             Measure and input current temperatures.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 


4. Set up Tectite software

 

e. Skip Comments screen

f. Test Settings screen

 Use default (Auto, CGSB, no active zones.)

 

 


 

4. Set up Tectite software

 

In Test Graph screen accept Auto Mode.

                  Turn on DG700.

 

 

 

 

5. Measure baseline pressure & cruise

 

a.       With the flow-cover on the blower-door fan and    the fan-speed controller set to minimally on, select     the Cruise option.  Perform the Baseline test.

b.   Tell the software to continue and use an OPEN   fan, the software will automatically direct the blower   fan to depressurize the home to 50 Pa.  Note to what  pressure the home can be depressurized;  if less     than 50 Pa, record this value in the Maximum Cruise Depressurization field, H5 in the Duct Leakage Calculator.  

c.    Continue the cruise during the following pressure-pan steps.


 

5. Measure baseline pressure & cruise

 

With the flow-cover on the blower-door fan and the fan-speed controller set to minimally on, select the Cruise option.  Do Baseline pressure test.

 

 

 


 

5. Measure baseline pressure & cruise

 

Cruise.   Tell the software to continue and use an OPEN fan.

 

 


 

5. Measure baseline pressure & cruise

 

If the test environment does not reach 50 Pa, in the Excel file record the value reached in the Maximum Cruise Depressurization field, H5. 

 

5. Measure baseline pressure & cruise

 

d.  Continue the cruise during the following pressure-pan steps.

 

 


 

6. Find sample-supply-register via pressure-pan tests

 

a.  Enter pressure-pan readings into column F and rows 13 – 33 of the spreadsheet.  (Clear unused values: do not leave a “zero” in any field in this column.)

 

b.  Read the Average Register Pressure on row 35.

 

c.  Find a register that has the closest value to the value found on row 35. (This register, hereinafter referred to as the “SAMPLE-SUPPLY-REGISTER,” exhibits a pressure most typical of the duct system.) 


 

6. Find sample-supply-register via
pressure-pan tests

 


 

6. Find sample-supply-register via pressure-pan tests

 

Enter pressure-pan readings into column F and rows 13 – 33.

 


6. Find sample-supply-register via pressure-pan tests

 

a. Enter pressure-pan readings into column F and rows 13 – 33.

b. Read Average Register Pressure,

               on row 35.

 

See next slide.

 



 

6. Find sample-supply-register via pressure-pan tests

 

a. Enter pressure-pan readings into column F and rows 13 – 33.

b. Read Average Register Pressure,

               on row 35.

c. Determine sample-probe-register.

   Choose the register with pressure-pan reading closest to average register pressure.

 

See next slide.


 

7. Setup data collection at sample-supply-register

 

b. No further data collection at this supply register is required; the rest of the testing is standard.

 

The procedure for this datum is identical to the standard,

         Modified Subtraction

duct-testing procedures – the procedure you have used until now.


 

8. Measure and Record pressures in
Attic & at Sample-Supply-Register

 

•        While in cruise mode, measure the pressure in the attic and record the data in J63

•        While in cruse mode, measure the pressure at the sample-duct-register and record the data in K63.  (If this is not the taped test, you should use the value you already measured with a pressure pan.)

 


 

 

8. Measure and Record Pressures in
Attic & at Sample-Supply-Register

 

Make a manual attic pressure reading relative to the house at the highest pressure.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

 


8. Measure and Record pressures in
Attic & at Sample-Supply-Register

 

c. Take SUPPLY-PROBE-REGISTER

readings at the highest pressure.

 

 


 

9. Perform an automated House-Leakiness Test at a series of pressures and save data

 

a. Save and rename the Tectite building file to “AsFound.bld”. 

b. Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

d. Save the Tectite file again.


 

9. Depressurize at a series of pressures and save data

 

Save & rename the Tectite file to “AsFound.bld”

 


 

9. Depressurize at a series of pressures and save data

 

b.      Select Start Test in the Test Graph Screen.                      

Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 


 

9. Depressurize at a series of pressures and save data

 

b. Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 

 


 

9. Depressurize at a series of pressures and save data

 

b. Fully or Semi-Automated Mode: Select Start Test

    in the Test Graph Screen.                                                                       

Tectite will depressurize the house automatically,

   and perform preliminary and final baseline tests.

 

 


 

9. Depressurize at a series of pressures and save data

 

b. Fully or Semi-Automated Mode: Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 

 

9. Depressurize at a series of pressures and save data

 

b. Fully or Semi-Automated Mode: Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 

 

 

 

9. Depressurize at a series of pressures and save data

 

b. Fully or Semi-Automated Mode: Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 


9. Depressurize at a series of pressures and save data

 

b. Fully or Semi-Automated Mode: Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 

9. Depressurize at a series of pressures and save data

 

b. Fully or Semi-Automated Mode: Select Start Test in the Test Graph Screen.  Tectite will depressurize the house automatically, and perform preliminary and final baseline tests.

 


10. Copy whole-house leakiness data to calculator

 

a. Change to the Test Results screen of Tectite.

 b. Update the data of the spreadsheet file in row 63                                               

         (first among Untaped Cases):

 From Building Leakage Curve data

    Input Flow Coefficient into Column E

    Input (flow) Exponent into Column F

 From Airflow at 50 Pascals input CFM into Column B

  To monitor and maximize accuracy

    Input the % error of the same datum into Column C

    Input Correlation Coefficient into Column G

 Record measured House Leakiness via

  From Leakage Areas data input ELA into Column I

 


 

10. Copy whole-house leakiness data to calculator

 

a. Change to the Test Results screen of Tectite

 


 

10. Copy whole-house leakiness data to calculator

 

a.       Change to the Test Results screen of Tectite.

 

b. Update the spreadsheet file into row 63

 

   i. From Building Leakage Curve data input Flow Coefficient into Column E

 

  ii. From Building Leakage Curve data input (flow) Exponent into Column F

 

 iii. Input the attic pressure reading, (from step 8) into Column J.

 

 iv. Input Average Supply Register Pressure (from F35) into Column K

 



 

10. Copy data to calculator

 

From Airflow at 50 Pa

                Input CFM into Column B

 

To monitor and maximize accuracy,

  Input the % error into Column C

    From Building Leakage Curve

  Input Correlation Coefficient into Col G

 

To record measured House Leakiness,

    From Leakage Areas data

  Input ELA into Column I




 

11. Check data’s accuracy, etc.

 

•        When errors are found they are corrected and testing is done over.  In this simplified description of these checks are skipped. 


 

12. Tape registers and repeat 8 - 11

 

Record as in Step 8 and 10.  Except:

  1. Save and rename the Tectite File: “AsTaped.bld” before the testing begins.
  2. Use the same test pressures as before.
  3. Take Attic and Sample-Supply-Register readings at the highest pressure.
  4. Put data on row 73, (1st Taped case).
  5. Save the calculator and Tectite file.

 

12. Tape registers and repeat 8 - 11

 

 

Record as in Step 9.  Except:

Save & rename Tectite file to “AsTaped.bld”

 


12. Tape registers and repeat 8 - 11

 

Use the same test pressures as before.

 

 

12. Tape registers and repeat 8 - 11

 

  1. Put data on row 73, (1st Taped case).
  2. Save the calculator and Tectite file.

 

 

13. Calculate Duct Leakage

 

(At this point you have all of the data needed to calculate duct leakage to outside.)

 

  • Copy any row from the untaped data, e.g. row 63, from Column A to Column K into cell A90.

 

  • Copy any row from the taped data, e.g. row 73, from Column A to Column K into cell A91.

 

  • Place a “1” in Cell M83 if attic & supply pressures are collected manually. If not put “0”.

 

  • Read Duct leakage to Outside and an overestimate of its error on line 103.

13. Calculate Duct Leakage

 

  • Copy the cells from any row from the untaped data, e.g. row 63, from Column A to Column K into cell A90.

 

  • Copy the cells from any row from the taped data, e.g. row 73, from Column A to Column K into cell A91.


 

13. Calculate Duct Leakage

 

3.   Place a “1” in Cell M83 if attic & supply pressures are collected manually.  If not, put “0”.

 


 

13. Calculate Duct Leakage

 

Read Duct leakage to Outside and an overestimate of its error on line 103

 


14. Enter Data into RemRate

 

    1. In the Infiltration/Ventilation screen, enter the value of Effective Leakage Area (ELA) found in cell I-90 into both Heating Season Infiltration Value and Cooling Season Infiltration Value and set the data type to Eff. Leakage Area.

 

    1. In the Ducts screen, enter the value in cell H-103 into Total Duct Leakage and set the data type to CFM @ 25 Pascals.

 

14. Enter Data into RemRate

 

. In the Infiltration/Ventilation screen,

enter the value of effective leakage area (ELA),

cell I-90, into both Heating Season Infiltration Value and                                  

Cooling Season Infiltration Value and                                                     

set the data type to Eff. Leakage Area.

 

 

 

 

 


 

14. Enter Data into RemRate

 

b. In the Ducts screen, enter the value in cell H-103 into Total Duct Leakage and set type to CFM @ 25 Pa.

 

 


GSCA Procedure in a Nutshell

 

 

Do pressure-pan tests to choose a sample-supply-register closest to average reading.

 

  1. Depressurize home to pressure P.  Record:

     House-Leakiness Flow as Q.

Pressure in the Attic with respect to inside, PA.

Pressure in the Ducts, at the sample-supply-register, with respect to the attic, PD.

 

2. Repeat Step 1 at 8 pressures, P, from 15 to 50 Pa.

 

3. Tape registers, repeat 1 & 2  to record , A & D.

 

 


Thank You J

 

Questions?