I need to get one thing out of the way before we go any further. Insulation is not a single decision. It’s not “pick a product, fill the cavities, move on.” I wish it were that simple. It would make my job a lot easier and this article a lot shorter.
I just walked a retrofit where I used three different insulation strategies in the same house. Closed cell foam in the basement. Dense-packed wood fiber in the walls. Blown-in wood fiber in the ceiling. Three products, three install methods, three sets of reasoning. If that sounds like more work than it needs to be, I’d push back on that. The lazy version is using one approach everywhere and hoping it holds up. The right version means understanding what each zone of the house is actually up against, then matching the tool to the problem.
Here’s the thesis, and I’ll say it now so you’re not waiting for it: the win in a retrofit is continuity. No gaps. No thin spots. No weak links where the comfort you paid for leaks right back out. Everything I’m about to walk you through, basement to roofline, is in service of that one idea, even though the materials change completely from floor to ceiling.
The basement: moisture makes the call, not me
Basements don’t play by upstairs rules. The second you go below grade, you’re dealing with soil moisture, and soil moisture only moves one direction. From more to less. More is the dirt outside your foundation. Less is the inside of your house. So it’s pushing toward you constantly, as vapor, whether you’ve thought about it or not.
Ignore that and you get condensation on cold concrete. Ignore the condensation and you get liquid water. Ignore the liquid water long enough and you get mold. That musty basement smell everyone’s smelled at some point? That’s not “old house character.” That’s a moisture failure with a smell attached to it.
So down here, I frame the wall two inches off the existing concrete or block, and I fill that gap with closed cell spray foam. One move, two jobs. I get continuous insulation, and at two inches thick, that foam is also acting as a vapor barrier, which keeps the concrete surface warm enough that condensation stops being a threat.
I don’t foam everything down here. That would be overkill and it would cost more than it needs to. Where I’ve got weird cavities I can’t staple into, like around an old cast iron pipe I’m not about to fight, I switch to TimberBatt. I can cut and mold that stuff into tight, irregular spaces without reaching for foam I don’t actually need there.
The floor gets its own logic, too, because you almost never know if there’s a vapor barrier under an existing slab, and cutting concrete to find out is expensive and disruptive for no good reason. So I just assume there isn’t one. Clean the slab, apply a moisture mitigation coating (the same kind used in commercial flooring), and build a sleeper floor on top with standard two-by material. It doesn’t need to be pressure treated, because the coating underneath is already doing that job. I still tape the underside of the sleepers with zip tape anyway, forming a capillary break, because I’m not building for the next five years. I’m building for the next fifty to a hundred, and I want that floor protected even if something changes down the road that I can’t predict today. TimberBatt fills the cavities before the subfloor goes down.
The rim joist: the spot everybody forgets exists
Above grade, I want the same R-value across the whole plane. Wall, rim, doesn’t matter, same performance. The rim joist is the classic place that doesn’t happen, because it’s a mess of framing joints and leak paths that were never easy to reach even before the house got old.
For air sealing there, I use AeroBarrier. It’s aerosolized, non-toxic, and it gets misted into a pressurized house where it follows the actual airflow straight to every leak until the leak seals itself. I used to chase these gaps with caulk, tape, and canned foam, and it took forever and it still wasn’t always right. This is faster and it’s better. I’m not going to pretend otherwise just because it’s newer.
Once that’s sealed, I match the insulation there to whatever the wall above is running. There’s no engineering reason the rim should perform worse than the wall it’s bolted to. Depending on the cavity, that’s TimberBatt cut to friction-fit, or TimberFill dense-packed if I want full coverage with zero voids.
The walls: where cross-furring earns its keep
This is the zone where I brought TimberHP in, because this is where their products do the heaviest lifting.
Standard framing is two-by-four studs, sixteen inches on center, wood sheathing behind it. On this house, I added two-by-threes across that framing every sixteen inches. The only place old framing touches new framing now is a strip an inch and a half by an inch and a half. That’s it. Ben’s math on this project puts the thermal bridging reduction at somewhere around 96 percent compared to studs with nothing interrupting them, which is a big number for a fairly simple move. I want that confirmed with TimberHP before it shows up in anyone’s marketing copy, because I’d rather be right than fast on a number like that. But the method holds regardless of the exact percentage.
With that lattice built, TimberFill goes in behind netting. Saddle-staple the netting so it holds tension and doesn’t bulge, then dense-pack the cavity full. We pull core samples on this job to check install density, and that’s not a box I check for the photo. Dense-pack only works if the density is actually there. Skip that step and you’ve got expensive fluff, not insulation.
I’ll be straight about why TimberHP specifically. Their material starts as forest product residuals. Wood chips left over from mills cutting pine boards or lumber, sourced from FSC and SFI certified small diameter pulp trees. That market used to flow to paper mills. As paper’s declined, wood fiber insulation has picked up that leftover supply chain. That’s not a sustainability tagline someone wrote for a brochure. That’s an actual materials story, and it’s part of why I use them.
The performance backs it up. TimberBatt runs around 3.8 per inch, and because of how densely it’s packed, it’s self-supporting. That matters, because cellulose has a real history of settling over time, which is where insulation earned a “junk” reputation it didn’t always deserve. Fiberglass batts settle too if they’re not stapled right. Wood fiber, installed at the density it’s designed for, doesn’t have that problem.
I’ll also name the tradeoff, because I’d rather tell you now than have you find out the hard way. Wood fiber takes more attention to install correctly than tossing in a fiberglass batt and calling it done. You need the right cut, the right density, the right approach around penetrations. Once you know what you’re doing it’s not slower. But there’s a learning curve the first time, and nobody sells you that part.
One more thing worth knowing: wood fiber acts as a moisture buffer. It absorbs a real amount of humidity and helps even it back out, which means steadier interior humidity instead of swings. Borate salts get added in manufacturing for fire retardancy and to resist mold. If “borate” makes you nervous because it sounds like a chemical, it’s the same mineral family used in dilute form in saline eye drops. Everything’s toxic at the wrong concentration. This isn’t some exotic additive. It’s a well-understood choice that’s been used in building products for a long time, and I’m not going to pretend I have some secret concern about it that I’m not telling you.
The ceiling and roofline: where the small stuff turns into the big losses
This is where retrofit reality gets you. A house from the seventies or eighties almost never has a raised heel at the roof edge. That means the insulation depth gets squeezed right where the ceiling meets the roof, right at the plate line. You might be supposed to have twelve to sixteen inches of insulation there and actually have four to six. That pinch point is also where the most thermal bridges converge, so it’s carrying more of the heat loss than its size would suggest.
On this project we thickened that detail on purpose. Built a trade ceiling that bought us more head height in the room and more insulation depth at the edge, at the same time. That’s a real tradeoff, giving up some ceiling height to fix a performance problem, and it’s a conversation you have straight with the homeowner. You don’t bury it in the plans and let them find out later.
The air barrier gets handled differently up here, too. The drywall on the ceiling is pulling double duty as the air barrier, so we ran AeroBarrier with that drywall already in place while the walls below were still open cavities with no insulation in them yet. That keeps the air barrier continuous around the whole envelope, even though the sequencing looks backwards if you’re used to new construction.
For the ceiling cavity itself, especially the larger, irregular spaces that don’t match a standard batt or a clean joist bay, blown-in TimberFill is the right call. You’re not fighting the framing. You just fill the space completely, and you’ve got one continuous blanket of insulation that isn’t settling, isn’t shifting, isn’t going anywhere.
Why any of this actually matters
None of this is about chasing a bigger number on a spec sheet. I don’t care about bragging rights on an R-value. I care about continuity. Basements need moisture handled first and insulation second. Walls need the thermal bridging cut and the density verified, not assumed. Ceilings need the one spot old construction never accounted for protected on purpose.
Get all three right and the house stops having moods. Rooms stop running five degrees apart from each other. Corners stop feeling like a different season than the rest of the room. That’s what the homeowner actually feels, even though they’ll never see a single one of these details once the drywall’s back up.
Retrofits are never a clean slate. You’re working with what’s already there, adapting to what you find when you open it up, and building every detail robust enough to handle the fact that you didn’t get to design this house from day one. That’s the job. I like the job.