Poring over the Manual-J, it looks like the windows have a U-factor of about U0.8 BTU/hr per square foot per degree-F, which would be typical of single-pane wood-sashed multi-light double-hungs, with NO storm windows. That's also the primary driver of both the cooling and heating loads.

Is that REALLY what you have for windows?

If yes, fixing up the weatherstripping and installing tight Low-E exterior storm windows would reduce the window U-factor down to about U0.30- U0.35 for far less money than replacement windows of equivalent performance, knocking about 20,000 BTU/hr off the heat load. It'll knock some off the cooling load too, but that won't be quite as dramatic.

West facing windows are probably driving the peak cooling load. Most bungalows have ~ 2' overhangs, which would provide sufficient shading to knock down the summertime gains from south facing windows to something reasonable. It's the east & west facing glass that comprises the bulk of the solar gains, but the west facing glass takes those gains late in the day when air temps are much higher than at dawn, and the thermal mass of the house has been soaking up sun all day. Thus from a peak load perspective shading, the west windows has priority over the rest.

The U-factor of the walls in the Manual-J are around U0.05, (roughly the equivalent of 2x6/ cellulose with R5 of continuous insulating sheathing), but I don't really understand the description of your double-wall construction, or whether that is a realistic U-factor to use.

Hey Dana,
Nope - that's not what I have for windows. I did specify double pane windows in the tool - I'll double check to make sure that it is entered correctly. That being said - I know for a fact the windows are a big part of my problem. The new windows I installed (a 6' half round, a 6x5 fixed and two 3x5 casement) are all vinyl frame, low-e argon filled double pane with triple seal weather stripping. The windows I have newly installed have the following energy star info (ER 32, UF 1.56, Zn 2, PG 70, P/N-DP 3360, WP 720, Air=A3) The old windows are double pane, broken seals, shrunken weather stripping and wooden frames - a little better than a sieve and duct tape...

I have almost no direct facing west windows. The majority of my windows are across the front of the house which is primarily south facing. Only 2 windows on the North side as well. Here is a link to google maps so you can see the orientation of the house: https://www.google.ca/maps/@44.9147219,-75.5898415,142m/data=!3m1!1e3

The south facing walls (with all the windows) are 2x6 construction. They currently have R19 fiberglass pink. After I'm done renovating - they will have R21 Roxul. The east, west and north walls are all 12" thick - double wall construction. The unrenovated walls are about R36 - the renovated walls are actually R40 and much tighter. I'll describe them again and see if I can do a better job... Here are the layers of the wall from the inside to the outside (of the renovated wall).
1) 1/2" drywall
2) 2x4 wall with R12 fiberglass pink
3) 1/2" sheathing
4) Moisture barrier (6 mil poly)
5) 4" space filled with R14 Roxul
6) 2x4 wall with R14 Roxul
7) 1/2" sheathing
8) Tyvek
9) 1/2" vented air gap
10) 5/8" tongue and groove pine siding

I'm happy to answer any and all questions. I'd love some tutoring as to what the numbers on my Manual-J actually mean!! Also happy to provide you with the login info for the tool I used for that calculation if you want to look at any of the input information and ask additional questions.

Consider me a charity case!

Sean.

Garbage-in= garbage out. The U-factor of your doubled-up wall with the R14 continuous thermal break in the middle should be about U0.028 or so. So at a 75F temperature difference (68F- -7F= 75F) the steady state losses per square foot of those walls would be about

75F x U0.028= 2.1 BTU/hr per square foot at design condition.

The 2x6/R21 walls would come in at about U0.065, for a per-square foot loss of:

75F x U0.065= ~4.9 BTU/hr per square foot of wall area at design condition.

If you installed 1.5" of continuous rigid rock wool or EPS sheathing it would drop those walls to the U0.05-ish implied for all walls by the Manual-J.

The beater wood sashed double panes with the broken seals are probably performing at about U0.6, which is still a BIG difference from U0.8. That would make the window losses (a still huge) 30,000 BTU/hr rather than the ~40,000 BTU/hr indicated in the Manual-J, more than a 10% difference in the total number.

A new low-E argon filled wood sashed double-pane casement will typically come in around U0.28-U0.32.

Run yourself an I=B=R spreadsheet for the heat load of the rooms you're fixing up to get a handle on the magnitude of what you need to get out of the radiant floor. The basic methodology is outlined in these bits o' bloggery:

http://www.greenbuildingadvisor.com/blogs/dept/musings/how-perform-heat-loss-calculation-part-1

http://www.greenbuildingadvisor.com/blogs/dept/musings/how-perform-heat-loss-calculation-part-2

The accuracy of that sort of calculation isn't very precise, but it will put an upper bound on what you'll need to fully heat the space, which is an important first step. There are better radiant-specific software load & design tools out there for this, but I don't own or use them, and there are also sophisticated energy modeling tools that would be more precise, but the accuracy depends on just how accurate your construction details are. (garbage in= garbage out) A dumb I=B=R calc will usually overestimate the true load by 15- 20% or more, but you may have that much "noise" in your raw inputs.

I'm less familiar with all of the window performance terms used in Canada. In Canada window performance is rated on the net solar gain, and not necessarily a simple U-factor that can be inserted into a dumb load calculation. All of that stuff can be modeled if you need higher precision, but for these purposes it's less relevant. If a dumb calculation gets you within 20% of the right answer, it's usually "good enough", though not be good enough for folks trying to hit an arbitrary performance number for say PassiveHouse certification, etc.

I would expect a typical tightened up 2x6 house that size with some foundation insulation and current code-min windows to end up somewhere around 45-50,000 BTU/hr @ -7F, literally half what was calculated. You have a higher than average ~20% window to floor area ratio driving the load numbers a bit higher, but it should still come in well under 60,000 BTU/hr. Typical new construction runs a ~15% window/floor area, but both 10% and 25% ratios aren't unheard of. The greater the glazing fraction, the more important window type & orientation become for managing the load numbers.