GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Musings of an Energy Nerd

Installing Fiberglass Right

It’s hard to do a perfect job

Proper installation takes time. Few fiberglass contractors take the time to install batts carefully.
Image Credit: Fine Homebuilding

Of all of the commonly used types of insulation — including cellulose, rigid foam, and spray polyurethane foam — fiberglass batts perform the worst. As typically installed, fiberglass batts do little to reduce airflow through a wall or ceiling assembly; rarely fill the entire cavity in which they are installed; and sometimes permit the development of convective loops that degrade insulation performance.

Knowing this, why would any builder choose to install fiberglass batts? The answer is simple: because fiberglass batts cost less than any other type of insulation.

Before we totally dismiss all fiberglass batt installations, however, it’s important to note that there is a big difference between the typical fiberglass batt installation and a best-practice installation. If a conscientious builder installs fiberglass batts carefully, it’s possible — although not easy — to get the best of both worlds: adequate thermal performance at a relatively low price.

Lots of sloppy installations

Study after study has shown that most fiberglass batt jobs are sloppy. In 2002, the California Energy Commission contracted with researchers Marc Hoeschele, Rick Chitwood, and Bill Pennington to conduct a study of new California homes. In its March 2003 issue, Energy Design Update reported, “The performance problems uncovered by the study were particularly disappointing in light of the fact that all 30 houses [studied by the researchers] were enrolled in programs promoting building-envelope improvements and duct tightness.”

The article went on to note that “the California Energy Commission’s ‘envelope protocols,’ which include standards for air sealing and insulation installation, were widely ignored. … Not a single builder managed to implement any of the following standards:

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. homedesign | | #1

    The Air Barrier is the Key Component
    A perfectly tailored fiberglass installation ... installed by the A-team....will not make up for a less than perfect air barrier.
    A fiberglass job will require much more attention to the air barrier than cellulose or foam.
    So .. you will need the A-team for the air barrier and the Batt installation.

  2. greenophilic | | #2

    Blame Energy Star and lackluster contractors
    Good post as always. One must wonder why the prescriptive performance programs out there have not simply banned the use of fiberglass batts. The constant refrain that I always hear is that all insulation products work well when installed right. That is true, but if one product has statistically higher likelihood of incorrect installation, then it should simply be relegated to those builders who don't care to participate in performance-based programs. Energy Star and others simply need to remove fiberglass from their acceptable roster of insulation materials. If they don't, they are not providing the performance benefit they are selling, as it is clear that no standard can make an installer do fiberglass correctly. Nor is it easy for the verifier to assess quality of install, which is often hidden behind kraft paper.

    One other thought, Harley indicated that all spray foam homes were better than cellulose homes, which in turn were better than fiber glass homes. Do you think this actually has to do with the material itself, or is the insulation choice simply indicative of a contractor's overall dedication to quality controls? Probably it's a mix...

  3. GBA Editor
    Martin Holladay | | #3

    Good question, Brennan
    You asked "Do you think this actually has to do with the material itself, or is the insulation choice simply indicative of a contractor's overall dedication to quality controls?"

    I pondered the same question when I wrote the April 2005 article for EDU. Here's what I wrote:

    “Although the data clearly show a consistent correlation between insulation type and airtightness, the reasons for the correlation are unknown. It is possible that builders who choose cellulose or spray foam insulation — both of which have a reputation for resisting air flow — may be more meticulous in performing air sealing tasks than builders who choose fiberglass batts. Perhaps builders’ attention to detail in certain areas unrelated to insulation performance — for example, careful sealing between wall bottom plates and subfloors — partially explains the measured differences in airtightness. If this theory is true, the extra dedication to air sealing may be motivated by the builders’ desire to justify the added cost of spray foam or cellulose over fiberglass batts. However, the principle of Occam’s Razor favors a simpler explanation: that the measured differences in airtightness are due to differences in the material characteristics of the different insulations.”

  4. homedesign | | #4

    Minimum Standard for Air Tightness
    If code required a blower door test and had a minimum standard for air tightness...
    then the half-arse builders would have no choice.
    They would either upgrade their insulation material or they would be forced to employ some other air barrier strategy that meets the standard.
    The result would likely be a lot less use of fiberglass.

  5. Expert Member
    Michael Maines | | #5

    Compressed batts
    One of the rules for good installation you mention, avoiding compression of batts, gets undo attention. For years I've heard repeated that you can't compress the fiberglass or you ruin the R-value. The truth is that you reduce the potential R-value, in other words waste material, but you still end up with higher R-value than you would by using less material.

    I think this one myth has caused more gaps to be left in the insulation system, resulting in reduced performance, than any other one of the many things that can be done wrong with fiberglass installation.

  6. GBA Editor
    Martin Holladay | | #6

    You're right
    Michael, you make an excellent point. Compressing a fiberglass batt reduces the R-value of the batt but raises the R-value per inch. In other words, batt manufacturers have not optimized the product for R-value per inch; they have optimized the product for a low manufacturing cost.

    Owens Corning has published a chart showing the advantages of fiberglass compression. According to the chart, a 2x6 cavity can be filled with a so-called R-19 batt, resulting in a cavity R-value of R-18. The same cavity could also be insulated with a 9 1/2-inch R-30 batt compressed into the 5 1/2 inch cavity. That would result in a cavity R-value of R-21. Better R-value through squishing.

  7. Expert Member
    Michael Maines | | #7

    Thanks for clarifying my description. Just after I posted that, I flipped open the new JLC to see your Q+A on this same topic, which made me second-guess what I said. So I pulled the same Owens-Corning chart to double check. Glad we're in agreement! And curious that they don't seem to publish that chart anymore....

  8. GBA Editor
    Martin Holladay | | #8

    More on fiberglass compression
    Let's consider a wall cavity with a horizontal electrical wire that bisects the studs. A lazy subcontractor decides to insulate the cavity by tucking the insulation behind the wire (instead of delaminating the batt to go on both sides of the wire).

    The lazy installer has violated the "no compression" rule. But as Michael points out, there really isn't any problem resulting from the compression of fiberglass batts, so the "no compression" rule is a misnomer. The problem with the lazy installer's action isn't the compression — it's the uninsulated air gap on the interior side of the wire.

  9. user-626934 | | #9

    Energy Star 2011
    Great article as always, Martin. For Energy Star homes beginning in 2011, it looks like fiberglass batts might be (for all practical purposes) removed from the equation. Part of the proposed revamping of Energy Star for New Homes (which starts January 1, 2011, see for more info) is to REQUIRE Grade I insulation installations. This would, in my opinion, effectively eliminate fiberglass batts entirely. In my 18 months of working as a RESNET Home Energy Rater, I've never seen a fiberglass batt installation that came anywhere close to Grade II, let alone achieve Grade I!

  10. GBA Editor
    Martin Holladay | | #10

    That's interesting
    John, I didn't realize that Energy Star builders were quite that bad. Very interesting information! We all imagine that Energy Star builders are at least a notch above the run-of-the-mill builder -- and even Energy Star builders can't seem to manage even a Grade II fiberglass insulation job. More evidence, if any is needed, of a gross contradiction: the code requires installing fiberglass batts according to the manufacturers' instructions, and these instructions require installation techniques which, for all practical purposes, are never achieved.

  11. Coyo | | #11

    Martin - great article!
    Good point also about the compression vs created airspace and your clarification that fiberglass needs to be installed by a six sided air barrier and fit perfect to work as advertised. I agree with you about the drastic effects of gaps and voids in performance-we measured over 40% reduction in R-Value of a R-13 fiberglass bat installed in a 2x6 wall cavity-which leaves a sizable air gap to the sheetrock. My experience with fiberglass batts is that they only work perfect under one condition: laboratory testing conditions at 70 degrees in the hotbox with ideal moisture contents and no airflow. As soon as you introduce some airflow and moisture to fiberglass it's performance deteriorates rapidly. Around 99% of all moisture that works its way into the building envelope is due to air flow and we currently have no regulations or set goals about air tightness, which about guarantees the two things to happen that fiberglass insulation can't handle well.

    And I agree with John that I have hardly seen any Grade 1 installs...however I doubt that the new Energy Star reg's will get fiberglass batts removed from the equation in 2011 - the fiberglass industry is just to entrenched and powerful in its current marketplace. They have by far the biggest market share and are most widely used. I believe that to be the only reason why we also don't see more better batting alternatives like Mineral or Rockwool on our market. As we are creatures of habit most people also like to use what they always used - and are simply unaware of the problems. That's way we need folks like Martin and his good writings to keep the rest of us straight. Thanks, and keep up the good work.TC

  12. Darin Zurliene | | #12

    Insulation ?
    Fiberglass isn't an insulation it is simple filter and a poor one at that. It only can resist heat and cold from 30F to 90F or to a Delta T no greater than 50F.
    It is interesting that there are continuing questions raised on what is or isn't insulation and what material performs with actual insulating values.
    It started with the Federel Trade Commission in its faulty testing reuirements of materials used for insulating homes, 75F is what the material is tested at, which doesn't doesn't meet real world applications, that is when the HVAC is idle. The FTC also ruled just recently (Tuesday, May 31, 2005: 16 CFR Part 460) that even though there is sufficient proof that fiberglass is ineffective as insulation, the FTC ruled it would be confusing to the consumer, if manufacturer labeled the product to which would inform them to the effective parameters of resisting heat and cold. The FTC opinion also stated that it would not have significant effects on the national economy, or on the cost of home insulation products, or on covered parties or consumers.
    The confusion is greater now than it has ever been, with no help from the FTC who is clearly afraid of loosing the tax dollars from the fiberglass industry, because they consume a large amount of gas to manufacture and fiberglass is still used in most homes.
    When working with the home owner, a HVAC contractor, an energy provider, or a builder. It is simply put. If fiberglass works so well, why are you calling on other products to give you a quote on insulation. They have little or no faith in fiberglass to provide them with efficiency or comfort.
    When insulation is being discussed, fiberglass shouldn't even be allowed at the table.

  13. GBA Editor
    Martin Holladay | | #13

    Thanks for your comments
    Thanks, Darin. For those who don't know him, Darin Zurliene is the owner of Spray Insulation, Inc., an installer of cellulose insulation with headquarters in Bartelso, Illinois.

    While many technical objections to the approved methods for testing insulation have been raised, the current testing methods at least require competing manufacturers to adhere to similar standards and test methods. For more information on how insulation products are tested, see "Understanding R-value."

    Contrary to Mr. Zurliene's assertions, fiberglass is an insulation product.

  14. Ben Ainslie | | #14

    First off, Martin, it made my day to see you taking a stance against fiberglass insulation. Your stance against the ASTM testing c1363 which rated a wall system with P2000 a 10.2 per inch led me to believe you were some a fiberglass-guy. For that I apologize. I thought that because the standard astm c571 was created by the fiberglass industry 40 years ago. What you and Mr. Zurliene may not know is that when the ICC did the testing for P2000 they also tested an identical wall system with R-19 fiberglass and it tested at R-11. They then added moisture to both wall assemblies and the fiberglass tested at a R-9. This is the fiberglass plus the wall system. Oak Ridge labs, which both of you are familier with I'm sure, stated that a wall with perfectly installed fiberglass in a perfect climate will never perform higher than R-13.
    The FTC is slowly coming around though. Every year new insulation product are being switched over to the C1363. As of now, by law P2000 has to display both the old ASTM C571 which rates it a 4.1 and the new "wall system" test which rates it 10.2, on the product. Fiberglass still holds 90% of the insulating industry. Money is power in this country so don't keep your hopes up that they will have to switch to the new testing anytime soon. Like you said Martin, people by it because it is cheap. The biggest hurdle in green buiding is getting the consumer to believe the price hike on high performance products like hard pack cellulose, XPS, EPS, and Icynene is a worthy investment.

  15. Mark Williams | | #15

    Compression is a Misnomer
    Martin, I disagree about your statement saying that compression, e.g. around a wire is a misnomer problem. You point out that the air gap is a problem, but due to compression, you've changed (reduced) the R-value in the vicinity of the compression. Both the compression and the air gap contribute to the problem.

    Overstuffing, but filling the cavity IS OK. You 'waste' R-value in a sense, but you got more R per inch and as long as the cavity is filled, we have a higher insulation level.

  16. GBA Editor
    Martin Holladay | | #16

    Hmm ... not sure I get you
    You wrote, "due to compression, you've changed (reduced) the R-value in the vicinity of the compression. Both the compression and the air gap contribute to the problem." Either I misunderstand you, or I disagree.

    If you take a batt designed to fill a 3 1/2 inch deep cavity and you compress that batt by sliding it behind a wire bisecting the space, then the insulated volume behind the wire will be insulated to a higher (not lower) R-value than the same volume would have had if the batt had been split.

    Unfortunately, the area in front of the wire will probably remain uninsulated. The air gap is the problem.
    However, if, after the drywall is installed, the insulation contractor comes back and carefully drills a hole at the center of the air gap, and fills the air gap with blown cellulose or blown fiberglass, the wall will perform better than it would have with a split batt. The compression was never a problem.

  17. John | | #17

    and the derating would be....
    I am curious how much a class 2 installation should have its effective R value derated.

    If for example a R-19 batt installed to class 1 standards achieves R-19, what does a class 2 installation achieve? Would you derate it to say R-16?



  18. GBA Editor
    Martin Holladay | | #18

    More information
    I'm not sure of the answer to your question, but the answer can be inferred from an example given in Bruce Harley's article: "Consider a typical example: a 2 x 6 wall with R-19 insulation rated at 6 1/2inches,that gets a rating of Grade II at the rough inspection. The HERS rater would assign the assembly an R-17 wall, with appropriate on-center stud spacing and interior and exterior sheathings, and would choose Grade II from a set of radio buttons. The software would do the rest."

    Presumably the de-rating from R-19 to R-17 refers just to the R-value of the installed insulation, and doesn't include the de-rating attributable to the wall framing.

  19. user-956864 | | #19

    How to deal w batt faced wings and drywall glue?
    Please discuss the wings on batt facing that I want the contractor to staple to the studs for full compression. Then glue (which the builder feels is the best, most permanent way to attach drywall correctly to the studs) won't work. I am looking for full compression of the cavity and can't get it when the insulation guy tucks it in instead of using the face wing as intended. Maybe we get rid of the wings and go with new friction fit so the drywaller can glue, I get full fill on all 6 sides. Is gluing the superior way to attach the sheetrock? I am a rater frustrated w the insulation guy for not knowing the standard and figuring out how to do it correctly and still leave a stud for the drywaller to attach to.

  20. GBA Editor
    Martin Holladay | | #20

    Response to Rick Jenkins
    Manufacturers of kraft-faced fiberglass batts recommend that the stapling flange should cover the edge of the stud facing the room, and should be stapled to the edge of the stud (the side measuring 1 1/2 inch). See the illustration below, from "How to Install Fiberglass Insulation," a guide from Guardian Fiberglass, a batt manufacturer.

    If you want the edges of the studs to be left exposed so that you can glue drywall to the studs, then you should choose unfaced fiberglass batts. Because these batts don't include a vapor retarder, your local building inspector may insist that you paint the drywall with vapor-retarder paint.

  21. user-948834 | | #21

    All this talk about fiberglass is about batts. Any batt, no matter what the material, is not a good choice. Fiberglass done right is called BIBS. BIBS is approved according to FTC 460. BIBS is a tested in the field insulation system. Cellulose, or any other insulation, is not. It is foolish to say fiberglass is no good, based on a poor installation. I have blower door tested homes with "dense packed" cellulose installations and found settling in as little as 3 months. Proper installation is key with any system. Dense packed fiberglass in a retrofit application, blown in at 2.25# cubic foot can no way settle and has the same infiltration as 3.5# cellulose and is R-4.23 per inch. Cellulose is great in an open blow, but I would no way put it in a wall.

  22. GBA Editor
    Martin Holladay | | #22

    Response to Tim Johnson
    I agree than blown-in fiberglass is far preferable to fiberglass batts.

    I disagree, however, with your disparagement of cellulose insulation. Certainly cellulose is a "tested in-the-field solution" -- whatever that means. The R-values of insulated cellulose walls have been extensively tested, and cellulose installers have decades of field experience.

    Your statement that dense-packed cellulose installed to 3.5 pounds per cubic foot can settle is simply untrue. I agree with your statement that "Proper installation is key with any system" -- a statement which is just as true for blown-in fiberglass as it is for dense-packed cellulose.

    Properly installed, dense-packed cellulose will not settle. Of course, any insulation can be installed improperly, so it is always important to choose a skilled installer -- whether you specify fiberglass or cellulose.

  23. feldmann2775 | | #23

    Recommended vapor barrier for wood TNG walls
    Great article, very helpful. I am currently building a home and am strongly leaning towards mineral wool insulation in the outside walls (2x6). Martin, I don't know if you like that any more than fiberglass, but I prefer not to use spray foam. There will be no drywall in the house (well, garage only) - all the interior wall and ceiling coverings will be 3/4 inch TNG basswood. While technically a hardwood, basswood is very light weight. I have no idea what the air or vapor permeability will be of the finished product. Would you still recommend a "smart" vapor retarder in this case? I am on the Zone 6/7 border. The sheathing will be 1/2 inch of plywood. No insulation exterior to the sheathing. Exterior finish is stucco, and I plan to follow GBA's advice of using a drainage mat (the 10cm version if I can find it). Any advice appreciated.

  24. GBA Editor
    Martin Holladay | | #24

    Response to Matt Feldmann
    With any air-permeable ("fluffy") insulation, including fiberglass batts or mineral wool, it's essential to have an air barrier on both sides of the wall. So the main problem with your plan is that tongue-and-groove boards are leaky. (The vapor permeance isn't a huge deal, but technically in your climate zone, the building code requires an interior vapor retarder, so you have to think about that issue as well.)

    The easiest way to ensure that you have both an interior air barrier and an interior vapor retarder is to cover your walls with drywall and tape the seams, and then paint the drywall with vapor-retarder paint. Once this work is complete, you can install any type of finish material -- including tongue-and-groove basswood if that's what you want -- on the interior side of the drywall.

    For more information on air barriers, see Questions and Answers About Air Barriers.

  25. feldmann2775 | | #25

    Response to Martin
    Thanks Martin. So you don't think a careful application of MemBrain would suffice in this case? I am a little concerned about tearing the plastic during install of the tongue-and-groove, but considered putting drywall shims or furring strips over the studs (after MemBrain installation) to keep the boards from rubbing against the plastic during their installation. This would be less expensive, but if you think drywall is the way to go, would 1/4 inch suffice? Or here is another thought - what about 1/4 inch of plywood, with taping or caulking the joints?

  26. GBA Editor
    Martin Holladay | | #26

    Response to Matt Feldmann
    MemBrain is quite fragile. Moreover, I don't know if anyone has studied whether the many fastener penetrations that are required to install tongue-and-groove boards would reduce the effectiveness of the MemBrain as an air barrier. I suspect they would. (Builders have plenty of experience putting fasteners through drywall -- an approach that doesn't seem to hurt the ability of the drywall to work as an air barrier.)

    Drywall is an inexpensive and time-tested air barrier; it's the best material in this case, in my opinion. I would use 1/2-inch drywall if I were you. With thinner drywall (or thin plywood), the fastener penetrations might cause air leaks.

  27. shorthand_charlie | | #27

    Thank you for all of the helpful posts and articles on this site! I just became a GBA member so I could really dive in. What a wealth of info as I have been designing and now building a small home for my family. I'm in it deep now.... :)

    The home is 1100 sqft, zone 5 in Colorado, 2 levels with a 3:12 monopitched roof built with 11 7/8 I joists, vaulted cathedral ceilings. I am using meticulously taped OSB as my primary air barrier. My current dilemma is to make final decisions on insulation.

    I will be doing continuous rigid foam on the exterior of the building--2" on walls (R13) and 5" on roof (3.5 " polyiso and 1.5" of EPS ~R28 total) I am currently trying to decide on the best cavity insulation for the project. I am planning on using cellulose in the walls but am unsure about the best cavity insulation for the unvented cathedral ceilings. I am unsure about the ease and quality control of blowing cellulose into the roof cavities. I have been intrigued by mineral wool batts but it seems that the 16.25" bats that would fully fill the cavities in the I joists are only available to an R15 or R23 (at least in the US). The R30 batts are only available in the 15.25" batts for stud cavities. Combining layers (say an R15 plus and R23 batt) gets cost prohibitive.

    I started thinking to fiberglass batts. An unfaced R38, 16" wide by 12" deep is readily available, would completely fill the cavity, and is quite a bit more affordable per R value. But I know that fiberglass batts have been shown to be less effective due to difficulty in proper installation. I have done a lot of fiberglass installation in the past and understand the difficulties and really planned on staying away from the stuff.

    However, I can't imagine a more simple installation scenario: long, regular runs, consistent width, with minimal to no wiring/plumbing interruptions. The I joist flange would even ensure no sagging with minimal compression. It is starting to sound like a winner to me. And that is surprising me because I, like most of the posters/readers on this site have been turned away from the pink stuff.

    Am I crazy? Are there dilemmas or risks that I am missing? The bump up in R value could possibly require more exterior foam but at R28 as planned, I'm still above the code required threshold for exterior R (42.5% of total R). I'm not up to the roof framing yet but the air barrier sure should be consistent and tight from walls to roof as I have planned. I've been sealing the hell out of everything. I'm thinking air movement and infiltration should be pretty minimal in the roof cavities. I'd love to hear any thoughts. I'd love to be talked out of this idea. But right now, with my brain wrapped around so many different aspects of the building project, this idea is making sense to me. In half an afternoon, I could easily have all 600 sq feet of roof all insulated. Thoughts? And a big preemptive thank you!

    (btw, I've never done blown in cellulose. From my research, I think I'm prepped for a good wall installation, but the cathedral roof cavity install of cellulose still seems iffy to me. Maybe I'm missing something there too...)

    1. GBA Editor
      Martin Holladay | | #28

      If you are building a roof that combines exterior rigid foam and fluffy insulation in Zone 5, then about 41% of the total R-value of the roof needs to come from the rigid foam. (For more on this topic, see "Combining Exterior Rigid Foam With Fluffy Insulation.")

      If you install R-28 of rigid foam on the exterior side of your roof sheathing, then the maximum R-value of the fluffy insulation is R-40. So insulating your rafter bays with R-38 fiberglass is safe.

      1. shorthand_charlie | | #29

        Thank you Martin for your prompt response. And for all of the other helpful and insightful responses on this page!

        I think I am clear on the proper ratios of cavity to exterior insulation for a roof assembly. I do appreciate the confirmation, however.

        I guess the gist of my concerns expressed earlier have more to do with whether or not I can readily escape some of the pitfalls of fiberglass insulation installation as mentioned in the above article. My specific application (an entire roof made up of consistent, straight, cavities which are minimally interrupted with any plumbing or electrical intrusions (few to none) seems to me to be an example of an easy scenario in which to do fiberglass right--that is, Grade 1. Am I missing any possible obstacles or pitfalls that could make my thermal barrier subpar? In other words, is this a good application for fiberglass?

        And thank you for your copious information on the other aspects of green building (ratios, relevant code, design principles, etc) that have helped me to understand these concepts.

        1. GBA Editor
          Martin Holladay | | #30

          As the article on this page makes clear, fiberglass insulation performs well, meeting the R-values shown on the label, as long as it is installed according to the manufacturer's installation instructions. Few installers do a good job -- but if you are installing it, nothing is preventing you from doing everything right.

          As with all insulated building assemblies, decent insulation performance depends on (1) following the installation instructions provided by the insulation manufacturer, and (2) attention to airtightness.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |