A Deep Energy Retrofit, Fall of 2013

This post absorbs still-relevant content of withdrawn post Working With Cotton Insulation Batts.  At March, 2026, recycled denim as expensive wall insulation or sound treatment is a dead market. Thick batts for floor insulation are no longer sold anywhere.

Some cotton batts continue to be sold at Home Depot, labeled Henry UltraTouch. "R13" and "R19" bags under-fill 2x4 and 2x6 walls. Under-fill is from irregular thickness and extreme difficulty of cutting-to-fit. Odd circumstances are not served by powerfully-stamped perforations. Parting to pass wires is impossible. With my reporting, I seek to discourage all others from thinking that cotton batts are somehow "green."

My experience with cotton batts has relevance now, only as an element in my only experience with a deep energy retrofit. 

The deep energy retrofit began right in 2008, with a ground-mounted 3.5 KW PV array. The array safely beautifies and might fully electrify a 2.5 acre farm. It seems that the now-old air-source heat bump needs expensive deep-winter electric resistance heat backup. A good ground source heat pump should now be afforded.

Proper weatherization of the home to support this early-adopter heat pump energy independence is to my credit. I hope that the air-source heat pump has never again needed backup of electric resistance heat, and that no grid power has been needed. In this there as been need of an additional investment in large battery capacity. I hope to learn soon whether all of my work has been durable, and that net-zero has been attained.

Making Right Out of Fright

So much work in home weatherization is with starting conditions that are beyond challenging.

Falling-down insulation in a crawl space MUST BE REPAIRED. Batts out of contact with flooring have no insulation value.

The work is daunting.

My worker-willingness was observed as a child, where my father, a poor farmer in Minnesota who grew up with horse teams, likened me to the much-appreciated and rare

 "willing horse."

I wonder whether worker-employees can have the needed motivation. I think that, rather, unasked initiative is punished.

Fixing everything on the main floor, attic and roof took 274 hours in the Fall of 2011.

Here are the many details of these repairs as PDF captioned job photo albums:

Real Sealing Fixes HPwES Sealing In An Oregon Attic

Insulation and Decking

Wrap Ducts, Completion Survey

Microwave Fan Wiring and Duct

LED Light Installations

Roof Penetration For Kitchen Fan

Roof Penetration for Bath Fan

Remove Debris

Here are photos of main floor, attic and roof starting conditions:

Large attic floor pits exposed first floor walls to attic temperatures. Here, drywall was grouted about a chimney well.

Attic photos, work in progress:

Here are achieved conditions in the first floor and attic

174 hours of epairs in the crawl space were done in Fall of 2013

The found condition in the crawl space was blocked by a no trespassing indication.

Very heavy cotton batts hung down 6" from floor contact against twine restraint, there was zero insulation value, Brave workers persisted against common sense, and an invoice was paid. The lasting contribution was only a thorougn new Visqueen ground cover.

The found all-steel HVAC supply ducts were left horribly leaking, draped with R19 fiberglass batts. The duct insulation batts added much to a feeling of confinement. The same with R11 batts generally wrapped around copper water pipes.

 See that steel duct fittings, wyes, take-offs and reducers hang from 9" rips of 1/2" plywood. after perfecting co,p;ete coverage with insulating jackets, hoist to jam an eye bolt through a 1" hole in the plywood, and pin with a 5/8" dia length of resilient farm property tree branch,

Intimacy of flex ducts with the overhead plywood rips will discourage mouse habitat, I hope.

Someday, say in 25 years, the flex ducts will need replacement. It will be easy to unpin the steel fittings to bridge in new lengths of duct,

Consider the energy savings in employing flex ducts vs. solid pipes with wrong assumption of zero leakage either way:

Reduce Energy Waste in Thermal Cycles Of Heat Ducts

An important feature of the new ducting is much-reduced thermal mass exposed to each heating cycle. The mass of steel ducts responding is reduced from 200 pounds, to fifty pounds. 

Compute an energy savings for the reduction of duct thermal mass:

Assumptions:

Crawl space at 200-day heating season average of 50°F.

Average air temperature within ducts in a cycle is 90°F.

Wrapped ducts heat inertia is such that temperatures are followed 50%.

The furnace cycles every half hour, 48 times per day.

The total mass of ducts is reduced from 200 pounds, to fifty pounds.

The temperature change of the total mass of heat ducts is 10°F in each cycle.

The heat consumed in each furnace cycle is:

m * Cp * delta T

where m = 200 pounds mass, or reduced to fifty.

Specific Heat Capacity, Cp = 0.12 BTU/pound/°F

delta T = 10°F

BTU wasted in each cycle: 200 * 0.12 * 10 = 240 BTU

Times 48 cycles

Times 200 days

Result: 2,304,000 BTU per year

Times 0.00001 to convert to therms, is 23.0 therms.

Times $2 per therm is $46 per year.

At fifty pounds, this is reduced to $11.50.

Energy savings now with flexible ducts are $34.50 per year, not counting benefits of real duct insulation. I think wrap of unfaced fiberglass batts loses most of its value with air circulation.

I think this is a fair estimate. Under $50 per year is saved. Even with insulation wrap, the steel components without an interior liner may cycle by more than 10°F. However, the number of days and cycles per day, may be high assumptions for this home.

Please just recognize that lower duct thermal mass is a good deal, comparable to benefits of stopped leakage and of keeping fittings and flexible ducts aligned for least flow resistance. The prospect of much easier replacement of flex ducts terminating at fixed-in-place steel elements makes coping with flex duct failure more bearable. Be happy to have ducts renewed, clean.  every 25 years 

Study Details of the Ducts Hanging From Steel Elements

6" Chamfered Straight Take-Off From 12" 

HVAC Diagnostic Static Pressure Measurements

Link this post to advocacy through ASHRAE, The American Society of Heating, Refrigerating and Air-Conditioning Engineers. At ASHRAE I belong to one Technical Committee, TC 5.2, Duct Design, where I complain against prevailing cheap and poor design of residential HVAC air ducts. As a contractor I tear apart and reimagine some really-awful ductwork hacked in place by "professional" installers seemingly in compliance with the default rule book delegated to ACCA, Air Conditioning Contractors of America. I want superseding, better instruction, to be found at ASHRAE. Absent this improvement I will continue to practice better work as example to others. On this path I must be able to prove the better value of my alternatives, in a growing body of before and after HVAC measurements by simple manometer measurements of system pressures. Collect relevant blog posts by applying Label: HVAC Duct Design

Google the Subject of my proof quest:

HVAC Diagnostic Static Pressure Measurements

Google AI says:

Static pressure measurements in HVAC systems are crucial for diagnostics and ensuring optimal performance. They help identify potential issues like airflow restrictions, undersized ductwork, or dirty filters, which can impact efficiency and comfort. Technicians use manometers, which are pressure-measuring devices, to measure static pressure at various points in the system, such as before and after the air handler, the coil, and the filter. 

This is incorrect only in the mention of detecting a dirty filter.  Any sensible person will replace a filter on-schedule or upon dirtiness observation. The statement is deficient in not mentioning detection of possible air-side fouling of the secondary heat exchanger in a high-efficiency gas furnace.

I have known that a major purpose of the measurements is to evaluate crummy HVAC ducts, and then to celebrate improvements made. Unfortunately I have been slow to accumulate measurement sets in a way that is sensible, recording measurements both before and after improvements made. 

I will quickly learn in a small sampling of customer homes that meaningful measurements identifying good or problematic HVAC duct design are very easy and unambiguous. They are for the purposes I imagine, just confirming my common sense about duct design. Measurements are not for a silly purpose of "measuring" blower flow rate. Rather, they are just design and construction quality documentation. Want least flow resistances rather independent of construction cost.

Find an appropriate graphic by search: 

graphic: a typical HVAC air duct arrangement and gage static pressure measurements

Credit this to energystar.gov

Label the gage pressure sample points as:

P1: Return, before filter box.

P2: Blower suction, after filter box.

P3: Heat Exchanger, wall at exit, if there is separate cooling heat exchange. Else this is at the  Supply Plenum,

P4: (Or P3) Supply Plenum

By themselves, P1 and P3 (or P4) are measurements of builder variables at hand: efficient design of the Return Ducts and Supply Ducts. P1 and P3 (or P4) should be very small numbers., demanded for everyone. Small numbers will correspond to quiet operation, best observed and judged directly.

P2 should be a very small nunber, except with fouling of the air filter, not a builder issue. Demand biggest/ best filtration for everyone.

The blower and heat exchanger investment is a builder and buyer choice not measurable by manometer. Higher efficiency in a gas furnace with the secondary heat exchanger fouled by cheap or none filtration, becomes a bad choice.

Customer Rae, Goodman Furnace, Upflow, With Air Conditioning:

Good duct design in a home with already-existing supply registers is simply the use of the largest-size steel pipes that are available in branching to final register-size pipes.  Keep air guided and moving at similar velocity everywhere.  Allocate registers to ducts such that register air flow velocities are equal.

See that overhead space and the furnace profile limit the size of four equal-sized starter collars on the supply plenum. A best-possible air filter is conveniently serviced at the furnace backside. 

Here is my measurements setup for customer Rae with greatly-improved HVAC ducting enabled by demolition of a three-stories chimney. Learn: why should one demolish a condemned, inactive chimney? The chimney chase closed below-attic is now occupied by round ducts mainly the 12" return path from the upper floor here seen atop a collector box, instrumented with tap P1. Here label four static pressure values that are exemplary for most HVAC with separate exchangers for heating and cooling.

Good furnace mechanics are taught that for every well-running furnace, gage static pressure downstream of the furnace heat exchanger shall be less than 0.5 inch H2O. Foolish mechanics conflate this pressure, related to gas combustion conditions vs. atmosphere, to value P3 - P2, sometimes called Total External Static Pressure, a phrase unrelated to science in Bernoulli Principle. Wikipedia lists Static Pressure. but does not allow the unscientific phrase.

Here is the Rae table of gage pressure measurements and some computed differential pressures:

Wonder whether P1 and P4 as duct resistance measurement, are as small as they can be. They are reasonably small numbers, judging by other measurements reported here. Be concerned about Blower/HX Delta high at 0.9 inch H20. Expect readings less than 0.5 in H2O. With larger numbers, suspect fouling of the secondary heat exchanger, caused by operating for many years without a good air filter.  See that for this home with unusual multiple return air paths, interior door positions matter. Don't wish for such variables, but see that variation is small, not to worry about door positions.

Gage static pressure values P1 and P4 by themselves have important meaning about duct flow resistance since they are uncoupled.  P1 is a  measure of return-air  flow resistance from house interior space up to the air filter. P4 is a measure of flow resistance to house interior space, from the final heat exchanger egress. Supply ducts have a bit more flow resistance than return ducts. Good design has the supply and return resistances equal and smallest-possible.

Customer Meyer, Goodman Furnace, Upflow, With Air Conditioning:

Here is a table of static pressures for customer Meyer, reimagined ducts well-documented in this blog post:

See that P3 - P2 is 0.8 inch H2O, exceeding an expected value near 0.5 inch H2O. Wonder if this is due to fouling of the secondary heat exchanger, where the furnace was operated for years without an air filter.

Looking at P1 and P4 as indicators of ducts return resistance and supply resistance respectively, see that supply resistance is double that of return resistance, and both are acceptably small.

My Home Goodman Downflow. Without Air Conditioning:

Here is a table of gage pressures and differentials for my home, 

Looking at P1 and P3 as indicators respectively, of return ducts and supply ducts flow resistance, see that returns are over-sized. The supply ducts are complex and long, with above-average but acceptable flow resistance.

Homes with heat pump HVAC will also have just three pressures sampled. Hope that in retrofit of a heat pump, ducts are reimagined and replaced.

Leet Carrier Upflow, 7/18/2025

An ideal upflow arrangement, with intelligent, biggest-possible flexible ducts. 

This is not the work of the original furnace installer.

10/5/2014 

The homeowner complained of loud furnace noise. I found, and fixed, this. The repair included reimagining all supply and return ducts in the crawl space. Such service is possible because I can rely on the design and build of an excellent sheetmetal shop. Nonsense must be avoided somehow.

Here is the job photo album:

Furnace Plena and Flex Duct Innovation, Multnomah Village

At 11/25/2025 upon finding a nice comment to this post, edit this job photo album, brightening all photos. I wish iPhone photos by default were not so dark. Know that much of the value I share is in a reader taking the time  to learn details of my consistent inventions.

Here mention the security achieved in flexible duct suspension from overhead floor framing using hangers to wyes and other strong steel duct elements.

 

And another invention, insulative coating of the interior of steel duct elements to reduce swung thermal mass.

10/25/2014

See wye trees that replaced an absurd Medusa D-box. Avoid turbulence everywhere, with steady-velocity guided flow.

See P1, understandably tiny return flow resistance from a large floor register directly-above. . See P4, a modest 0.21 supply ducts flow resistance. See suspiciously small air filter delta; perhaps there is something wrong with P2 location vs a turning vane. Someday wish to relocate P2 closer to the air filter. This furnace runs so very quiet. Think it is at lower blower RPM. P3 leaving the furnace is watched to be near 0.5 in H2O, and is much less.

Martinez Heat Pump Furnace Pressures and Error Discovery

Surprise! 

Here is all-electric HVAC, a heat pump furnace with beautiful cabinet simplicity.

And, surprise again to find that lofty heat pump installers can be stupid. A hack-in air filter slot does not support air filtration.

See the absurd return air path at the P2 probe half the height of the air filter slot. All return-path cabinetry cut into the furnace LHS wall must be replaced, including a horizontal proper air filter 20x25x5 and a big turning box with much taller side cutout of the furnace wall..

A 12" x 20" x 1" filter propped upright, is freely bypassed.

See that furnace and heat exchanger delta-P, P3 -P2, is quite large,  perhaps associated with heat exchanger fouling.  Perhaps a heat pump heat echanger is very unlike that of a 

 

HVAC ducts are very efficient:

P1 is the resistance in return ducts, -0.20 in H2O. 

P3 is the resistance in supply ducts, -0.15 in H2O. 

Here is my plan for improving Martinez return air ducts:

(Wait to do this in September 2025, when A/C won't be missed for a few days.)

Return air flow resistance will be reduced by a factor of four, a decrease of about 0.5 in H2O of static pressure measurement. The furnace blower will run much quieter and at less cost. The new  filter will at last be protective of the heat exchanger. Inquire whether the heat exchanger should and could be cleaned.

Reject cost/ benefit of a rounded, lined turning plenum with turning vanes (Leet). No other ducts are lined. This follows the example of Rae. The found return path area is doubled, and there is a proper filter box. The side opening into the furnace is doubled in size. Inquire with Rheem as to recommended furnace side cutout markings and size.

The turning box and filter box will be 3/4” above the floor. vs. found down box 66” tall and 2 3/4” above the floor.Combined height of the down box and filter box will be 68”.

Customer Menezes, Bryant High Efficiency Gas Furnace, Downflow. With Air Conditioning:

Here is a first instance of observation in both cooling and heating modes. Shifts in location of pressure drop are surprising, rather small, and real. More energy is needed to move cold air into supply ducts in cooling mode. Like the Leet excellent Carrier gas furnace, operation is surprisingly quiet, probably with less energy applied at the blower and longer cycle run times. Get what you pay for.

On-site, we talked about a hostile crawl space. There is some evidence in the static pressure measurements that supply ducts were installed with care. Still, I wish to inspect the crawl space someday.

Cautions Against Mis-Wiring of GFCI Outlets

 Wonder why GFCI outlets by Eaton and Leviton are packaged and labeled differently. Find this familiar and invariant  adhesive tape over the Load terminals of a Leviton GFCI. 

These instructions inside the Eaton outlet packaging are unhelpful to understanding the intended wiring arrangement. 

This is the Leviton GFCI powering a lamp, with correct wiring and normal function. The fairly-visible square green LED is redundant indication of the normal function.

"American" brands Leviton and Eaton share a deficiency that gravity must be turned to advantage to engage wires in floating grips. Better "Chinese" volunteered design has no-fail wire insertion through guide holes.

Eaton outlets have no instructions at all. 

Commonsense in the following rules argues against need of instructions: 

All GFCI outlets have these functional features:

• The outlet will not function at all if miswiring has the (single) hot lead attached to LOAD side instead of LINE. The reset button then does nothing. There are no LED indications.
• Upon trip, the outlet and all outlets daisy-chained from LOAD terminals are dead.
• Outlets daisy-chained from LEAD terminals are immune to GFCI trip, not protected. Perhaps half of circumstances are with want of this.  A GFCI outlet may be in a long daisy chain, not visible or even imagined, at a scene of frustration. It is bad practice if a device failure is caused by another device. Let every needed GFCI outlet be independent. 
• Another half of GFCI wiring circumstances has one GFCI protecting very-nearby other non-GFCI outlets, out of foolish economy, passing from GFCI LEAD terminals.

Yet, not-thoughtful installations where daisy chains branch loads to include lights, fans, dimmers, and more, may be dysfunctional, perhaps completely denying GFCI service..

For functionality, of all brands, I prefer Eaton.

Eaton is unique in having a bright red LED illumination at an edge of the RESET button. A small red LED is said to immuminate with device failure. For all other brands the small LED in green denotes proper service. The Eaton illuminations are more thoughtful.

Trip-Tested Eaton

Corner, tiny green indications of correct function are kind of silly. They are easily obscured by a cord plug.

Yellow tape over LOAD terminals does make sense. Daisy-chained outlets will by preference be strung from GFCI LEAD terminals. A string continuing via LOAD terminals does not miss protection by the GFCI outlet, the implication of the warning tape.

While all outlets now sold in USA are made in China, we will have no loyalty to USA brands Eaton and Leviton, then open-minded to all other brands imported from China. We expect to pay no premium for a "USA" brand not made in USA. At May, 2025, these other well-made brands are found at Amazon: Amazon Basics, ELEGRP, XIMAOESE and GreenCycle, cutting by half, historic prices of about twenty dollars per outlet. Tarriffs have no USA manufacturing to protect against global manufacturing cost difference arbitrage, a corrupt practice hurtful to workers. I have bought one each of all other brands, at Amazon, and find inadequate instruction against "miswiring" for all.   XIMAOESE and GreenCycle are exceptional in including the cautionary yellow tape of Leviton, now fully guarding against additional outlet connections daisy-chained through the LOAD terminals. At GreenCycle, I lose patience with all Chinese direct-sale offerings at Amazon. Be angry that market-flooding by China acts to destroy USA device brands, where we lose control of features. Be more angry that for the past fifty years, tax policy encouraged USA brands of all kinds, to end USA production.

The GreenCycle Ground terminal requires bending around the screw, a now-ancient and hated feature of outlets. Want straight insertion and space for bonding more than one ground wire. With GreenCycle, the green LED is on even after test-tripping. Want only red LEDs, indicating a tripped condition.

All GFCI outlets observed, when viewed as smiley faces, have ground terminals at the bottom. Since it matters very much to know which set of power terminals is Line for Hot connection, and backsides are hard to read, there must be consistency that Line is the upper set, as with current China production for both Eaton and Leviton. Instead there is absolute inconsistency. All knock-off brands made in China for direct sale as with Amazon, have Line terminals as the lower set. This is absurd and unacceptable. Buyers will presume that found wires undisturbed will accept a replacement outlet.

Repair of a Leaking Attic High Vent, Employing A Roof Penetration Adapter

This post of an action in 2016 is important in a Year-2025 campaign for broad employment of "roof penetration adapters" in residential roofing. Click the hyperlink to find a reverse chronology of all of my writing on this subject, in this blog.

Clarification is needed, whether penretration adapters are needed at attic high vent caps. They do have virtue. Please read on.

Here is a leaking condition that had recurred about once a year when high winds and driving wind would create a siphon under an attic high vent:

March 13, 2016

See that this attic vent was blindly cut in by roofers, without awareness of roof joist locations, as a new requirement on a home built in 1970. It would not have happened with a shingle tear-off.

A new problem was introduced in a re-roof, with cap misalignment from the shingle and sheathing cuts.

There are solutions against these misalignments. Immediately defy the siphon by stuffing a shingle bridge under the roof cap.

Better with a new venting opportunity on an already-shingled roof, let the vent locations be piloted from within the attic. Let all holes be circular, cut from the roof about the pilot hole. Insert a penetration adapter then to divert water flowing down-roof. Thereafter in a tear-off, center the adapter on the roof cut.

An imperative of easy and safe freedom in-attic to place high vent pilot holes would have been served by my self-funded campaign of Mandatory Attic Access Walkways. Know that vents clog over time from the inside. Circular high vents might have more promise of long unobstructed service, than roof-peak slots. A high vent cap should be easily removable for cleaning, from the roof,  without pulling any shingle nails.

Get the idea of the new-roof high vent installation process from rework of a found 4"  roof cap, now to serve a strong kitchen exhaust fan with 7" ducts. My roof adapter has bore 7 1/2" ID, 1 1/2" tall, formed of spun aluminum. The available roof cap is Famco 8". The 2x8 board has deck screws that when worked into composition shingles about the pilot screw will enable a well-defined sheathing cut by Sawzall. This is the same-size adapter plate that was tried by a very competent roofer under my instruction, to fix my leaking roof high vent.

Construction Metals RV38 Vent Cap Over Added 7.5” Roof Penetration Adapter

The penetration adapter is one shingle row deeper than the roof cap. A lower shingle bridge stops possibility of siphon action, even without the penetration adapter. The penetration adapter is redundant except as a can’t fail locator of the sheathing and shingle cuts. The adapter is locked in posution where it is attached to a duct. For a non-ducted roof cap, set it permanently upon the sheathing cut using upper-edge screws or nails. With a firm-set adapter rim, cut replacement composition shingles by stabbing-around with an awl; then breaking away the divots.

A further virtue of the adapter is in allowing a powered fan to be set in place of a static vent, without disturbing roof shingles, tiles or metal sheeting. Want all powered fans to be freely removable for service or replacement, without disturbing roofing.

The leaking vent was repaired including employment of a 7 1/2” spun-aluminum duct penetration adapter. Opportunity of better alignment with the roof cut was missed, and the use of shingle bridges is unknown. There has been no return of leaking. If respected, this will be an example to a roofer someday of ensuring alignment with the sheathing cut. Wish that better alignment of the adapter will be chosen. Wish the cut of plywood ang skip sheathing had not been excessive. Sharp corners are stress risers for lumber splitting.

We find that very many bath fans and kitchen exhaust fans are badly, even dangerously ducted. In any existing home the rather frequent retrofit of a new fan is not subject to building codes/ permitting review. Awful access to the attic contributes to the problem. I see this as a unique engineer/ inventor/ contractor/ technical writer. From age 60 twenty years ago, I have been a general contractor focused mainly on attics and crawl spaces. I know that dirty, awful work, delegated, is without learning and practice with ever-better materials, methods and procedures. I demand that practice of myself and am unaware of another like-minded attics person anywhere.

A roof penetration adapter is very useful when a roof cap is served by a through-roof duct. We shall want that all roof caps and roof-top powered fans are removable for service or replacement, without need af attic entry.

A ducted roof cap with a stem invites others to bond the duct to the cap; then removable only with attic entry. My preference is that ducts shall be galvanized steel "warm air" pipes. The crimped end is upward toward the roof, engaging often as an elbow, to full, tight penetration of the adapter plafe throat. Slight maneuvering of the adapter in a reroof shall not decouple the duct somehow strapped-up.

There shall not be compromise aiming of a duct pipe or flimsy plastic thing toward an existing attic high vent cap. There shall be no capping over framing or roof sheathing at a static vent cap, with a common "duct starter" plate, the usual practice wrongly taught to weatherization contractors.

 At 2025 there is a wrongful trend to offer duct-served roof caps only with long stems that reach into the attic. There is at least one cap manufacturer with thinking somewhat like mine.

 InOvate/ BetterVenting 

Product RoofNeck

Employed with 4" stemless caps for clothes dryer ducting. This is a rolled-in plate with a pipe length necessarily reaching both directions from the plate. I do not want a rolled-in pipe engagement.

Where the adapter has a stem penetrating roof sheathing, it is securely indexed to the roof cut, but there is likelihood the pipe below will be locked to the stem by screws and tape. The RoofNeck is likely to be a nuisance in a re-roof, requiring roofer access to the attic. Roofers do not want this.

For an adapter without sheathing penetration, as I  offer of spun aluminum, some other means must index the plate to the duct. A crimped pipe end protruding through sheathing, works. I do not prefer an adapter with a long stem. It is so very much harder to interleave with shingles, requiring more nail pulling in an insertion through already-placed shingles.

Want one-piece adapter plates with a simple neck at least 1 1/4" tall, as formable in stamping of aluminum. If a short neck protruding downward is needed as a location index, let it be a separate galvanized steel pipe length, crimped end, inserted from the attic. The stamped upper protrusion has needs of leak-free permanence. The downward-extending