Posted By jonr on 08 Nov 2013 11:56 AM
You will have created the equivalent of putting the furnace and water heater in an unheated attic. I suggest some google searches on that. The pipes will probably freeze if there is no flow through them.

I realize that it's not ideal but it's been in an 'unheated attic' in the basement like this for 60 years but it's been without insulation between that space and the rest of the house. I figure at least insulating this already-unheated space from the rest of the house would be a step in the right direction, right? How would it be any different than what it's been for 60 years down there? Or am I missing something?

I figure at least insulating this already-unheated space from the rest of the house would be a step in the right direction, right?

The heat loss from the house and ducts to the room have probably saved you from a pipe freezing disaster. See the above advice for "insulate the exterior", "insulate the exterior", and "do nothing".

Posted By cbass94 on 07 Nov 2013 09:56 PM
Thanks again for the replies everyone. Given the discussion here and also taking into consideration costs, here's what my current plan is:

Leave the external air source (after reading up on the IMC 2012, it seems like that is best) and then insulate the ceiling of the mechanical room with Roxul R30 (rock wool batts) and insulate the two interior walls with Roxul R23 batts - 2x6 walls for that room. Because I'm leaving the external vent open to the outside, there is no need in my mind to insulate the exterior two walls of that room (remember it's a basement room and is 75% below grade).

Once all of that insulation is in place and that room is effectively blocked off from the living spaces in the house, I'll look more into doing a cold air trap if I feel I need more than just the insulation.

SO, now one more question... since there is no insulation on the interior walls of that room currently (the walls that connect to the rest of the basement living space), there is obviously no vapor barrier either. But in reading about the insulation I obviously see a lot of talk about vapor barriers. The problem is that everything that I've read says the vapor barrier should be on the living space side of the insulation/wall. Since the living space side of these interior walls are already finished there is no reasonable way to put up a vapor barrier on that side. So... do I need to worry about a vapor barrier in this case or just forget about it since I'm using the mold/mildew-resistant rock wool and I live in Utah where the air is usually pretty dry??

Thanks for the help!

PS, lastly I will also insulate all the duct-work in that room since I'll be leaving the external combustion air source.

That's the wrong way to go, since you then put the duct, air-handler, flue pipe and furnace losses outside the thermal envelope, and with the air handler and ducts now outside the pressure boundary of the house, air leakage on the duct systems then drive infiltration losses while running. It adds up to a significant hit in overall system efficiency.  When you start adding thermal partitions inside the house, it makes the whole air-sealing thing 1000% more complicated too. The best approach is insulate & air sealt at the foundation,  put a louvered door or a grill in the interior partition wall to take the combustion air from the interior, and letting any heat loss from the system accrue to the conditioned space instead of the great outdoors.

To insulate a basement foundation wall in SLC (US climate zone 5) the cheapest/simplest way to do it while avoiding moisture problems is a combination of rigid foam against the foundation trapped in place by a studwall insulated with unfaced batts.  The ratio of the foam-R to fiber R is important to keep average mid-winter temp at the foam/fiber interface on the above-grade portion of the wall above the average midwinter interior-air dew point.  The IRC has prescriptive numbers of R5 foam min for a 2x4 wall, and R7.5 foam min for a 2x6 wall.  Unless you're planning to super-insulate the place, an inch of polyiso (R6-ish) with R15 Roxul in a 2x4 studwall works, and delivers a "whole wall R" (all thermal bridging accounted for) of about R17. You can go thicker on the foam if you like, but no thinner.  Because polyiso is slightly hygroscopic (and wood is too), it's best if you put an inch or two of EPS between the cut bottom edge of the iso and the bottom plate of the studwall as a capillary & thermal break.  Tape the seams of the iso with FSK tape, and seal the gap between the top edge of the iso and the foundation with 1-part can-foam.

NEVER put poly vapor barriers on any sub-grade wall, even in UT- it's the most mis-applied building product ever invented.  Assmblies are much more resilient when they're designed to be able to dry, adjusting vapor retardency of layers to levels where moisture from vapor permeance isn't problematic. All sorts of things are vapor retarders (including standard latex paint), but AIR TIGHTNESS is by far the bigger factor in limiting moisture migration. (A square inch of air leak in a studwall is worth a whole house worth of latex painted wallboard, in terms of the total amount of moisture migration.)  Below grade is a special case, since you have moisture drives from both ground water and air to manage. Vapor barriers halt water vapor diffusion, but that's a recipe for creating unintended moisture traps.

It sounds like you live near the East Millcreek area. My #2 son's home located in that area is of very similar construction to what you have described.

My 2.5 cents worth.

I would have to agree with Dana on most of the issues he has identified but would like to stress that unless you have significant ground water or roof drainage near the foundation the vapor pressure from the foundation and floor here in most of SLC is actually quite minimal. After all we are the second driest state in the union. I did energy management consulting with Utah Power and Light (now Rocky Mountain Power) before taking up a college teaching position a number of years ago. Many of the issues encountered in more humid and high water table areas of North America while not inconsequential here they are certainly less of an issue. When you must drill several hundred feet to get any water for a well you begin to understand how dry we really are. Nearby irrigation ditches, landscaping sprinklers and improperly sloped landscaping are actually the greatest source of water for vapor intrusion. So simply making sure that you are not irrigating close to your foundation and that the ground is sloped away from the home can mitigate this issue on older homes. HVAC design parameters for Salt Lake City have been 95 degrees summer and 10 degrees winter. With summer time RH typically running at around 15% and dew points of 45 degrees and winter 30% and DP of -10 degrees being the norms having vapor intrusion is nowhere the issue I saw when living in Ottawa Ontario for example.

As Dana points out cold outside air intrusion onto warm walls and into the conditioned space should be your greatest source of concern. Sealing up your mechanical room, as I suggested in one of my last posts, from the rest of the house would make your HVAC and WH systems essentially, as someone else pointed out, the equivalent to installing in them in an unconditioned the attic BUT sometimes the attic or a sealed insulted basement room is the only reasonable economical alternative. The problem is twofold - first it is just about impossible to retrofit existing construction with the necessary seals required to make it air tight from the rest of the house and second everything in the mechanical room would need to be insulated - ducts, pipes, WH, furnace and you would still have a less than optimal solution.

You might want to check with Questar regarding their current outside combustion air requirements. I'm sure you know that they do free energy audits and might suggest you have your your furnace derated which would reduce the outside combustion air requirements. Most of the furnaces in our area , even new replacements such as yours, are as much as 300% oversized. In recent years Questar has reduced the square inch of ventilation requirement from the old AGA standards.

One thing I would add to Dana's comments on rigid foam board is if your floor joist are running perpendicular to your outside wall in your mechanical room then cut tight fitting rectangular pieces of 2- 3" ESP to place against your rim joist and set them in a bead of fresh canned foam or foam seal around them after placing them. If your floor joist are running parallel to the outside wall of course you can do the same thing with ESP but in much longer sections. You would then butt your rock-wool or fiberglass floor/joist space insulation up to that.

Thanks again for the replies everyone.

When I decided to insulate the room from the rest of the house, that decision was mostly as a result of my understanding of the fuel gas code (which is certainly nowhere near a perfect understanding :unsure:). I figured that if I closed off my external air source, the volume of indoor combustion air I'd have available would not be sufficient even if I vented the furnace room to the adjoining room via a louvered door or wall vents.

After Jonr and Dana1's replies suggesting that I shouldn't do it any other way, I decided I better get exact with my numbers though to make sure I'm doing it right. So, after looking up my furnace and water heater's exact BTU rating, it turns out the water heater is 38,000BTU and the furnace is 96,000BTU for a total of 134,000BTU. In looking at the IFGC 2012, it looks like 50 cubic feet of indoor combustion air is needed per 1,000BTU of capacity. So that works out to be 6,700 ft^3 in my case (134 kBTU x 50 ft^3). The current furnace room is approximately 11'x12' with 7 foot ceilings and the adjoining room (laundry) that I would have to vent to (via wall vents or a louvered door) is also approximately 11'x12' with 7 foot ceilings. This gives a total of 1,848 ft^3. Because this is nowhere near the 6,700 ft^3 required by code, I am still reticent to do this method.

One other option I have is to also replace the laundry door to the next room (family room) with a louvered door. The family room is open to the rest of the house via the stairs so that would meet the 6,700 ft^3 air volume requirement. But I really don't think I want to do that because then the noise from the washer and dryer would infiltrate to the family room much more.

So, given all of that additional info, am I missing something that would still allow me to close off that external air source and insulate the foundation walls of my furnace room without running into code violations/hazards? In my mind, that is the way I'd like to go - close off the external air source - but I only want to do it if it will be the proper way to do it. Maybe it's KI7OM's suggestion of seeing if Questar has modified the requirements so I don't need as much combustion air... I know direct vent appliances would allow this but that's not in the budget for the next few years having just bought the house.

One thing people seem to forget is that this air you are burning is going to come from somewhere. Using combustion air from adjacent space requires infiltration from undedicated openings.

By the way as you read the code, make sure you are clear on levels of construction as folks who endeavor to make their homes "unusually tight" may still have trouble with air from adjacent spaces. You would also want to include other gas appliances that share that air in your calcs (i.e. clotes dryer, stove/oven, gas logs etc.).

BTW The code used to permit combustion air from a dedicated (undampered) register (like a toe kick) on the furnace plenum providing there was a make up air (which could be dampered).

The fact that the location of the infiltration air needed for the combustion air is random and uncontrolled is not a reason to put a big uncontrolled hole in the thermal & pressure envelope (at the bottom of the stack-effect stack, no less) and thermally isolate the heating equipment from the house. A dampered air intake to the room interlocked to the ignition sequencing of the furnace could manage that, but it's kludgey at best.

Direct-vent/sealed combustion appliances are really the right solution in the longer term. Then you can make your house PassiveHouse tight if you like, and never have to worry about backdrafting or combustion air issues.

Even with effectively zero insulation in your CMU walls a 96K furnace is likely oversized 2x for your actual heat load at +11F (the 99% outdoor design temp for SLC: http://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/Outdoor_Design_Conditions_508.pdf ). When it comes time to replace your existing burners you may opt to go with hydronic air handler with a coil right-sized to your heat load at 120F average water temp, and heat it with a direct-vented condensing hot water heater doing double-duty. The smallest Polaris or the 100KBTU/hr Vertex hot water heater would likely cover your loads. Using the existing furnace to measure the heat load of the house where-is/as-is can be done, if you have a mid-winter gas bill with exact meter reading dates & usage. Any insulating or tightening you do will change that heat load. Insulating between the mechanical room and the rest of the house would likely raise the heat load, but insulating the foundation walls would lower it.

Before then, sealing every seam & joint on the ducts with duct mastic improves system performance (and in most 50 year old homes with hard-piped ducts reduces the noise when running.) Seal the seams on the air handler with FSK tape (aluminum duct tape). Caulking at between the sub-floor (or wallboard) and the register-boots helps too. With ducts installed a semi- or fully-conditioned basement there's no point in insulating the return ducts, and only a small return on insulating the supply ducts (but some.) Air sealing the ducts & air handler guarantees the flow is going to and coming from the intended locations, and reduces the amount of air-handler driven infiltration. In field auditing in CA by both utility industry researchers and the Lawrence Berkeley National Labs over the past couple of decades the typical or average amount of duct leakage of existing homes was over 20%, and homes where the leakage breaks 40% are not rare. While that 40% duct leakage is not a 40% loss to the outdoors when the ducts are all inside the conditioned space, it still is a net loss.

If you are going to insulate and air seal the room and ducts from the house for noise reasons, then I would:

1) put a small, thermostatically controlled electric heater in there just to make sure that it never gets to freezing.
2) also insulate the exterior walls (it will still help retain some heat, even with an exterior vent)
3) implement the bucket trap (external air supply) suggested above

Thanks again everyone. I'm still debating this with myself and I really do appreciate all the input. In the long run, of course, I want to do direct vent appliances, and with that in mind it does indeed make sense to do the combustion air vents/louvers to the rest of the house even if it means more noise in the family room for right now because I could get rid of those vents again after I got the direct vent appliances some years down the road. And thanks to all of your comments, I realize that simply insulating that room from the rest of the house is a lot more work than I'd initially thought anyway, so that's been extremely helpful. As of now, you've all convinced me to lean towards the louvered doors to the rest of the house even if it means more noise in the family room through those louvers. Given that approach, I'm going to mull it over for another month or so and really think through that aspect of it.

Very informative discussion everyone, thank you.

The saving of a few btus is not worth your life. As discuss earlier in this thread there are ways to almost eliminate air coming into the furnace room that is not passed directly up the furnace flue, with very little impact on the interior of the building. Dana, I'm sure you know this has been Canadian code for several decades. Canadian codes also specify that there must be a door between any furnace room and any bedroom.
As Joe points out about unusually tight, if this op is worried about a small piece of uninsulated wall, it is likely that he will also try to make the house as air tight as possible.

It might be fine for this OP to use internal air, (I would not do it) but the next person reading this might get himself killed doing the same thing. The colder the climate you live in the more important it is to have free flowing combustion air if your appliances are not sealed. Of course the colder it is the more the appliances run, but also the less the doors and windows are open and they are open for less time. Also the period of no opening at all (night time) is longer allowing for greater time of CO build up. Incidentally Calgary evacuated a large condo yesterday due to high levels of CO.

I can't imagine 40% air loss from ducts unless the pipes are broken. I can't even get my head around 20%. If the unit is running at normal cfm and the end of each run is terminated with an open register, I just can't see sufficient back pressure to force that amount of the air through the small leaks of a well fitted ducting system.

How would it help to seal between the duct boot and the floor unless there is a well sealed door between the two floors or if the registers are kept closed?

Johnr - I grew up in Ogden, Utah in the early 50's working for my GC father. Many of the homes in Salt Lake City built in the '50s were originally coal fired and later retrofitted to NG with outside combustion air supplied at the time of the retrofit. If they were originally fitted with NG appliances in the 50's then they likely did not have the outside air supplied to the appliances and have defacto been grandfathered in without. Most replacement contractors in SLC did not retrofit coal fired to NG conversion jobs with outside air. Back in the 80's I did a lot of work with Rocky Mountain Mechanical and Canyon Heating and Air Conditioning in the Salt Lake valley. If you check with Questar (AGA as well) you will find that they DO require outside combustion air to the NG fired appliances (furnaces and WHs). You may NOT depend on infiltration from the rest of the house envelop as makeup-air. If you were ever to have Questar come into your home for an inspection they likely would shut off your gas until the issue is resolved. Access to the mechanical room cannot be from any bedroom sleeping area and a louvered door bedroom access would be an even bigger issue. This is not just a comfort issue but one of life safety. Though Questar is not quite as insistent on having the hi/low inlets as they once were and are a bit more relaxed on the amount of square inches of opening to the outside you must have, blocking off entirely the outside combustion air inlet to one already fitted would also be a Salt Lake City Building Code violation issue in which if they were to do an inspection they would likely call for Questar to shut off your gas.

Having lived in eastern Ontario, Canada, for five years I became much more acutely aware of the building envelop and air infiltration/ventilation issues. Those same issues pointed out by Dana and FBBP should be taken seriously.

The mechanical room itself, not the duct boot, would need to be sealed entirely from the rest of the house for this strategy to really be effective. This would mean not just an air tight access door to the mechanical room but all mechanical room penetrations to the rest of the house including joist and build-down spaces, electrical, plumbing, and supply and return air ducts. To be effective all would need to be tightly blocked and sealed from the rest of the basement and living space.

Sealing the duct boot to the wall/floor ceiling it is penetrating is important to ensure ALL heat is going to/from the room intended, and not being pushed into/drawn from the framing cavity (aka "thermal bypass duct".)

The Lawrence Berkeley folks didn't just make it all up- air under pressure, even 10 pascals moves all sorts of unintended places if you let it leak. The magnitude of the leakage was a surprise, even to many of those taking the measurements. Lots of small leaks along the pressurized or depressurized paths add up. (They were also able to prove that closing registers in an attempt to economize by not heating a room increases rather than decreases the heat loss out of the house due to the higher infiltration losses, also a surprise to most.

It's damned near impossible to air seal the mechanical room from the rest of the house, even if you wanted to, especially in a ship-lap plank subfloors typical of 1952 construction. (When you think you've done it, run a blower door test and post the cfm/50 numbers.) It's far easier to air-seal at the foundation, and swap out the existing equipment right-sized for the load and direct-vented.

Residential HVAC and Distribution Research Implementation
CIEE/PG&E Final ReportThe DeltaQ test shows that the average leakage for these houses is typical of those seen in previous surveys (Cummings et al. (1990), Downey and Proctor (1994a), Jump et al. (1996a) and Modera and Wilcox (1995)) with 99 cfm (47 l/s or 10% of air handler flow) for supply and 107 cfm (51 l/s or (12% of air handler flow)) for returns. Page 10
found at http://epb.lbl.gov/publications/pdf/lbnl-47214.pdf

10 and 12% would make a whole lot more sense then 40%

CBass, You are 5 pages deep in deciding what to do about something that costs you a very modest amount of money.
Spend that money (you were going to use on temp fixes) on permanent home improvements (i.e. air sealing elsewhere) or save it for a furnace.

Posted By FBBP on 11 Nov 2013 09:11 PM
Residential HVAC and Distribution Research Implementation
CIEE/PG&E Final ReportThe DeltaQ test shows that the average leakage for these houses is typical of those seen in previous surveys (Cummings et al. (1990), Downey and Proctor (1994a), Jump et al. (1996a) and Modera and Wilcox (1995)) with 99 cfm (47 l/s or 10% of air handler flow) for supply and 107 cfm (51 l/s or (12% of air handler flow)) for returns. Page 10
found at http://epb.lbl.gov/publications/pdf/lbnl-47214.pdf

10 and 12% would make a whole lot more sense then 40%

I did not claim 40%as the typ, but only not very rare outliers.  The 10-12% typ leakage as the full sample average is lower than my recollection of ~20%, but OK that was for a survey of 110 houses in just one study. (That is neither the first/last/only set of data.)  In your cited document the paragraph at the beginning of p.11 also reads:


An outlier to be sure, but a movie you would want to guarantee to not be starring in.