Heat Loss Calc. Help

[molo]

Hello,

I'm trying to size a natural gas forced air furnace and I know that the wrong size furnace can result in discomfort, harm to the appliance, and excessive fuel usage. I've spent some time online reading about heat loss calcs, and it seems that one of the most accurate ways to size a new system is by looking at the (edit: input/output of the existing furnace, the volume of gas used, and making a calculation based on that). There is also a natural gas H2O tank who's usage I would have to separate from the furnace usage (not sure how, but the water usage is metered and this may also help). Here's some basic info about the space, and thanks for any help:

1. Rochester/Buffalo NY area

2. 700 sq. ft second floor with 4" of fiberglass in walls and ceiling and no insulation in the floor (above a heated space).

3. 8 double pane vinyl windows (3' x 2.5')

4. Current heat source is a 40-50 year old natural gas furnace in the basement

5. The existing furnace has very large duct work coming from it.

6. There is also a natural gas hot water tank (may make it difficult to determine past fuel usage of heater only)

[jimbo]

Looking at the existing heater is NOT the answer to finding the right size. On the other hand, if the home is comfortable, that can be a clue.
The "official" method is Manual J, published by the Air Conditioning Contractors of America ( ACCA). http://www.energysavers.gov/your_hom.../mytopic=12340

Most contractors these days have that on a laptop, but of course you can do it "the old fashioned way". But if you don't do a thorough load and heatloss calculation, you are just 'winging' it.

[Dana]

Go ahead and run the fuel use calc, assume 75% efficiency on the oversized beast and just IGNORE the contribution of the HW heater. About 8-9 times out of 10 it comes in lower than a Manual-J anyway. Unless you have 8 people living there who all take daily showers the mid-winter HW factor will be less than 10% of the gas bill in most houses in NY, and less than 20% of the annual bill. Typical Manual-J comes in at 25% over measured reality.

Manual-J on a room-by-room basis is useful for designing ducts or zoning systems by for determining the whole house heat load using an existing heating plant to measure the load is usually better for hitting the whole-house number with any precision. At the very least it puts an upper bound on it. Unless you're measuring the leakiness of the house with a calibrated blower door test you can't really hit it dead-nuts-on with a Manual-J, and it's almost always hitting high (sometimes 2x high). The heat load simply can't be more than what's implied by the source fuel use.

[ballvalve]

Dont forget to add in for the kid that chased the dog outside and left the door open for a half hour.

And the wife that makes 5 trips to the car for groceries and never closes the house door.

And the European wife that must sleep with the bedroom window open, also forgetting to close the vents and leaves the bedroom door open.

[Dana]

Those will all factor in to the "as-used" heat load measurement, but in most cases would still be closer than a Manual-J or similar.

A kid leaving the door open for a half hour every day in winter doesn't seem like a very likely scenario, not nearly as likely as a kid taking a shower for a half hour ever day. :-)

Using deep overnight setbacks on the T-stats can sometimes introduce a low double-digit error too, maybe even enough to offset the error introduced by the ignored hot water fraction.

Bottom line, if you size by the measurement you won't be cold, and you're less likely to oversize it than relying on necessarily-imperfect heat loss calculations.