Heating Degree Days NOT useful for sizing
 

Heating Degree Days are NOT useful for sizing a heating system at Design Temperature??? (which is how it's done)

I took the total January HDD for my location, say it was 930 / 31 days = 30*
60* indoor - 30* = 30*

I can't size a heating system for a Design Temp. of 30* I must use 0*


HDD are useful for estimating about how much heat might be needed for a typical January.

Weather Data Depot: free downloads of heating & cooling degree days

"
What is a degree day?

A degree day is a measure of relative heating and cooling energy required by buildings. It's calculated as the difference between the average daily temperature and the balance point temperature (60 degrees). When the average daily temperature is above the balance point, the result is cooling degree days; when below, the result is heating degree days.

Example 1: Average daily temperature = 80. Balance point = 60. Cooling degree days = 20 CDD. (80-60=20)

Example 2: Average daily temperature = 45. Balance point = 60. Heating degree days = 15 HDD. (60-45=15)

Example 3: Average daily temperature = 60. Balance point = 60. No degree days.

You may ask, "Why not use average temperature instead of degree days?" The problem with average temperature is that highs and lows cancel each other out. A warm day (80 average temp) combined with a cold day (40 average temp) average 60. So do two mild days of 59 and 61. But in the first case there are 20 CDD and 20 HDD while in the second there is 1 CDD and 1 HDD. Using degree days, you can see that the relative amount of energy required for the first set of days is much greater than for the second set of days. But if all you looked at was the average temperature, you would conclude that both sets of days were about the same."

BBP

Good points on not simply using total seasonal heating (or cooling) degree days for design load - alone.

I use both and expected wind and thermal mass. I look at the heat (or cool) required to offset an environmental load as BTUs per square foot per degree day. Even a super insulated home, with outstanding minimal wind infiltration, will require extra heat when it is bitterly cold AND windy.

Thermal mass also allows a home to absorb or buffer temperature extremes. You can quantify this on a very cold, cloudy and still day by turning off the heating system and looking at the fall in home temperature vs time. A semi-log plot (temps vs time) shows a straight line and this gives a very good estimate of thermal mass.

Years ago, I did and experiment where I added a lot of thermal mass to a home I was living in. I had access to dozens of 42 gallon plastic barrels. I filled them with water and compared the regression lines described above. VERY different. As a consultant to DOD, I suggested that potable water tanks (IBC containers) be kept inside living/conditioned spaces in cold regions. That is working well today and minimizes the size of the heating units needed to keep personnel comfortable.

The vast majority of cold mornings are met with a typical 20 F rise in daytime temp. With a significant thermal mass, the temp does not fall to the outside, but lags. Then the daytime heats things up again. Or to be proper, the daytime lessens the load.

Total degree days are very useful as you can look at yearly variance and plan for a prolonged cold spell that is outside the ability of the home to buffer (thermal mass).

For example, here in central Oklahoma we have about 3200 winter heating degree days. But the variance in this can be large - some 20%. January is our worst month for variance in terms of heating degree days and can be 50%! . In that month, we can also have terrific winds and cloudy days when we see little if any daytime increase in outside temperature.

This tells me that I need to plan for at least the minimum design temp - or have available a supplemental source of heat. In many home, there is a 15 kW electric resistance heater (electric oven) that can be briefly used. Correct, not a code approved heating source, but one that can give me some extra BTUs on a cold morning. I usually cook some biscuits which works well by heating my stomach AND the house.

That is where analyses of serial yearly (or monthly) heating degree days come in - not as a static value, but to estimate extra loads not described as an average.

Most HVAC installers have a simple response - just oversize the heating unit! It is not that simple as extra fuel must be provided to provide peak load (yet another waste). Highly mobile military units hate heavy stuff. But they crave warmth.

So much for my musings at 6 am . . .


Steve

Bill, good point! The most extreme day you are likely to see is the one you have to size for. The further that day falls from the average for it's month, the more shortfall you have if only using degree days to size.

Buffering helps, but how much? Steve, about how much water did you use, and about how much buffering does it give you?

Thermal buffering is the sum of many thermal masses. The smallest thermal mass is the air mass. The most substantial are heavy dense things like stone counter tops and such. The resultant thermal mass sum is the aggregate of all the masses acting independently as the environmental temperature changes.

Turns out that just one or two IBC totes of potable water make a huge difference in a 100 meter square space. The problem DoD has (Afghanistan) is that nights are bitterly cold as the air is so dry.

The way temporary field structures are built allows totes to be placed first and then the structure dropped (or built) into place.

My initial experiment was with about 200, one gallon water jugs. The next experiment was with steel drums (42 gal). My wife was frankly not very pleased with 30 steel drums in the house and was VERY glad when the experiment was over.

Those notes are about 30+ years old, but I did some calculations and was amazed at the potential buffering. Turns out the reality was very close. The specific DoD stuff is, believe or not, classified.

Michigan winters have a typical low in the low teens (F). When Kathy was on call, I would put house temp to 70F, run the blowers to fully saturate all the thermal masses. This took about 12 hours. Then I would turn off the heat and record the temperature with an old style recorder (circular device with pen writing thermometer on graph paper). The house would drop rapidly in an exponential fashion to the mid 40's the next am. Then turn on heat so wife had a warm home to come back to!

Didn't take rocket science to do a semi log plot.

Then brought in barrels and filled with water. One problem was that ground water in Michigan is COLD so it took a while to heat up all this water to 70 F(a few days). This time the house did not drop much below the mid 50s.

The house was crappy in terms of air infiltration so I had to choose my nights carefully.

The DoD has huge (hanger size) environmental chambers (Natick, MA) where these concepts were done in a more elegant way.

When we built a later home, here in Oklahoma, I put in a lot of thermal mass. That allowed a smaller geothermal heat pump to work very efficiently as I was really just heating (or cooling) against the average outside temp and not the full extremes. Used lots of concrete . . . and built a large basement with insulation on the outside walls. Could keep that huge house cool with a 12 kBTU GT heat pump in summer in Oklahoma.

Back in the 1980's, I recall some people putting water jugs in walls, sealing them up. At first this is tempting - until you realize that the polyethylene is time unstable and tends to start cracking after a few years . . . .

There is another thread where a solar greenhouse with aquaponics uses water filled steel barrels to buffer air temperature changes. Same concept.

Bottom line is that thermal mass, constrained with the envelope, can be a huge help to minimize energy bills.


Steve

I agree, manual j can be off quite a bit. Our design temp here is 96F however in the past ten years we have long stints of 100f+ days. One year we had over 60 days straight of triple digits. HVACs sized too close to manual J couldn't keep up. Lots of unhappy customers. Heating obviously you need to plan for those extremes as well, just because you average 40f for a low doesn't mean you won't have a freak 20f week. Perhaps a good argument for multistage equipment that could ramp up when you need it.

I can't wrap my head around this post. ..at all.

I live in a house built in 1985. 2100 square feet. Before starting my insulation and air sealing retrofit efforts with about R30 tops cellulose in the attic, R13 in the walls with a 3/4" wrap of non-foil(plastic instead) polyiso wrap(R18, but not that much once considering thermal bridges with 2x4s 16" on center. No slab insulation. Not a pile of windows but enough on the northwest summer side to heat the place in the summer but not provide much benefit in the winter, low quality windows, most with failed seals that condense between the panes. 1500cfm50, roughly 5 ach50.

This house manual Js at about 30,000BTUhr at -11f design load. I haven't bothered calculating the manual J for air conitioning but I'd bet the 2 ton air conditioner was sized under manual J. The furnace has a 57,000BTUhr output.

Prior to doing any retrofit work, this standard code minimum 1985 construction had an actual measured heat load based on run time of 25kBTUhr during the middle of the night while the temperature was a steady -20f or lower for an 8 hour period. In the summer(after air sealing) under design conditions of 88f with the 2 ton air conditioner it more than keeps up. With 72f inside(I don't normally keep it this low but for measurements sake I did) and 95f outside I've extrapolated that the actual cooling load for 75f inside and 88f outside to be around 15kBTUhr. Since I'm plenty comfortable with the temperature at 75f in my house, with the smallest standard split system in my house I could have the system keep my house at 75f while there is full sun and 91f inside with a 1.5 ton system.

With that being said, if I kept the windows the same size and orientation, based on how the surface area of a space gets larger linearly at the perimeter in relation to the heat load, I could grow my house to 4624 sq keep and keep the same equipment that I have now. ..and if you don't buy the linear concept and multiple by square feet, I'd be at 3150 square feet before I actually run the furnace constantly at -20f temperatures and the air conditioner would then be properly sized for a load that I'm very comfortable living with.

After I got done with a basic level of only air sealing and resolving thermal bypasses into the attic for a cost of two $20 sheets of XPS and about 15 cans $60 of Great Stuff. I've brought a 25k BTUhr heat load at -20f down to 19kBTUhr and 17kBTUhr at a -11f outside, 70f inside design load.

I seriously think that if someone can't maintain heat with a 40k BTUhr furnace, that person either has a very large house, they have too much glass(or inefficient glass), and/or there is a need for additional insulation and air sealing.

Air conditioning is a little bit different, it's more dependent on window size, orientation, whether the glass rejects heat, air sealing, cooking loads(are you really running the oven for hours in the middle of a blazing sun day, and internal humidity loads). Even with there being different configurations, I think that a 1.5 ton and 2 ton system could manage in a well designed house where the cooling loads were in mind during its construction in my climate. You'd go up a little bit with temperature but solar and moisture loads on the hottest days are a larger factor. Windows, air leaks, and humidity loads are the largest contributors.

..with that being said, I'm thinking that if the central air even breaks down it's getting replaced with a 40k 90+% condensing furnace and a 1.5ton air conditioner. ..although the reality is that I'll probably toss in a super efficient inverter 12k mini-split heat pump upstairs instead to handle most cooling days and only use the 8.5SEER system I already have on days where I feel it needs a boost. I live in a higher cost electricity state where natural gas and geothermal has too high of an initial cost and the electric costs wash out the long term savings. An air source inverter mini-split has a lower initial cost way to get high performance heating but the cooling savings is where it's at in a moderately hot and lake humid state.

In short, especially for heating loads: I want to know the furnace size, square footage, outdoor temperature, insulation level, and air leakage of any home that a natural gas furnace heated home is experiencing a situation where it can't hold a steady state temperature. I'm having a really hard time, based on my own experience, that a home air sealed reasonably well and has at least R13 in the walls can't maintain its heat with the furnace that is installed to manual J spec. ...also keep in mind oversizing to some degree typically ends up being built into the incremental sizing of the equipment unless the manual J is sitting right on the numbers.

Man j is seen as the best way to size and it is good but... It also has the GI-GO problem. (Garbage in garbage out).

When using the recommended numbers in our area it doesn't work for cooling. The design temp here is 99* however the last 5 years we averaged over 100 days with the temps over 100* with many of those days being 5+ hours over that temp and it will be upwards of 110* for a few hours. We had nearly a month where it was 115* to 117* every day.

The issue is that those high temps are just part of the equasion. The issue is that it will be 90* by 9 am then 100* by noon then it will still be mid 90s just before midnight with a low temp of 80-85 at around 5:30 am.
So when a unit is sized by man j much of the summer the ac will be running pretty much non stop all day and it won't cool the house to the desired temp till after midnight when the system finally can catch up.

Like mentioned many homeowners here with houses that have man j sized units were hot in their houses through the summer. Many have been up sizing their equipment because of it.

As to the devil of over sizing that is always brought up it isn't as bad as you may be led to believe. There have been DOE research that has shown that over sizing doesn't have an appreciable impact on operating cost unless it's grossly oversized and that nearly never happens. If you have a fear of over sizing you can go two stage systems or thermostats with adjustable temp settings to change from 2* to or 4* delta t (differential temp)

Allot of the same goes for heating as well and man j is still a good resource but it isn't the end all and be all and it even says that in its manual. It does need some updating for its design temps in certain regions. I researched the temp here for the 30 year period and it was a good chunk higher than the man j numbers ie their numbers aren't being updated...

Elcam,

You are exactly correct that oversizing an AC unit doesn't cost any more to cool than a correctly sized unit.

But ONLY if you are talking temperature . . . .

The condensation of latent heat requires that the unit actually run longer if the system is to remove humidity. That is the reason AC works - to remove moisture and as an added benefit, the air is cooler.

Steve

Further discussion re prior post.

There are large differences on heat/cooling depending on environment.

One of the worst places I had to contend with was an area in the Philippines that was subject to high humidity (80% almost all day - for months) and high winds. The design criteria was to maintain 50% humidity to protect sensitive electronics in a small "insulated" building with about 4-5 humans operating (mostly watching) the equipment. The both inside and outside walls were covered with foam sheets to prevent anyone outside listening to the goings on inside. VERY well insulated - perhaps R60 or so.

Two sturdy doors, with weather stripping, separated by about 5 feet comprised the entrance. But again, a lot of air leaks.

The combination of human water off gassing (mostly exhaled water, some perspiration) and the incredible imposed outside humidity load was daunting. The outside temperature was not bad - in the upper 80 - low 90s F. The electronics "heat" load was small as it was almost all solid state (no vacuum tubes). Probably a maximum of a few hundred watts (all CMOS technology).

But each person, at rest, contributes about 100-200 watts and five people is a substantial heat load in a 400 sq ft space. Also, no windows. A concrete block structure that one would think would have no air infiltration. Wrong! In retrospect, there were gaps in the manner to which the ceiling fit on top of the walls and this resulted in a large amount of water vapor infiltration. The significant winds essentially enlarged these small cracks as the windy side was pressurized and the down wind side had a slight vacuum.

Today, a small ERV would be helpful to remove moisture.

The smallest ac "through the wall unit" was found, but it cooled off the area too quickly and the inside became a soupy, wet, clammy and smelly mess. The electronics was literally failing as condensation would occur.

The solution came from a Navy submarine where tiny AC units are used to remove moisture from crew quarters where high density sleeping areas are. I don't recall the specifics, but it ran essentially all the time and removed a constant stream of water from the "shed". The problem I had was converting the power supply to accommodate the unit - subs are even stranger that aviation power supplies.

Frigid cold or blazing heat can be easily contended with, but high humidity at low temperature loads is tough.


Steve

I usually figure with R60 that they would have prevented or retroactively resolved the air infiltration issue. Normally using foam involves taping as well as offsetting seams including sealing(gaskets/foam/liquid applied flashing/etc.) any areas that meet other building materials. It sounds like something went wrong with the design.

..but either way with an equipment damaging moisture problem, a dehumidifier would have mitigated that when cooling loads are low. In my opinion, an ERV doesn't make much sense if your primary latent gains are through infiltration and internal loads, if it were to replace a ventilating system that was bringing in the moisture, that would make more sense but that doesn't sound like the issue. Granted it would remove the human factor/perspiration issue of multiple people in a small space whenever ventilation is lacking.

MN - you have never seen the Navy build a shed! Foam sheets when sloppily applied, do little to block wind - especially when it it 20+ mph almost constantly.

Steve

Hello stevehull,

I read your description of thermal mass with great interest. I have about 85275 pounds of sand that I use as a radiant slab in my office. I think it stores about just shy of 400,000 btus of usable btus. I'm in the process of trying to figure out exactly how to estimate the value of this in order to calculate the lowest usable water temperature I can operate at.

I'm thinking what I may do is this winter set-up a buffering hot water tank and see how low I can set it's thermostat while tracking the temperature along many points of my system.

Short of that I don't think there is why to estimate.

AC oversizing penalty has 4 major factors:

1: Poor humidity removal during part load conditions. Those humid days in the 80's people will turn their AC down to 70 or below to get humidity relief. A smaller AC will pull the humidity without overcooling as much.

2: Uneven temps in between cycles. A cold blast of air followed by a stuffy humid feeling. Happens each cycle, customers end up fiddling with the thermostat constantly.

3: Ductwork unable to deliver needed CFM. A big one, system will only deliver the amount of cooling the ductwork can handle. Customer stays comfortable because all the capacity isn't needed to begin with. They just pay higher utility bills.

4: Cycling efficiency. It takes 5-10 minutes before pressures stabilize and the system is operating at it's rated EER. oversized systems cycle more often and spend less time in the efficient operating zone.

Actually, I found using heating/cooling degree days PLUS utility data was very accurate for determining my house's loads. For a one month example,

123 therms = 12,300,000 btus at 882 degree days is about 14,000 btu per degree day or 581 btu per degree hour. My design day outdoor temp here in North Carolina is 20*F. So 70*F - 20*F is 50*F delta. Then 50*F * 581 btu / (degree hour) = 29,000 btu/hr at design temps.

Of course it is not exact as it does not include internal heat gain or the fact that those therms also heat water. But I refined the numbers by running a simple linear regression of two years data. The results matched what I expected from using a stopwatch to time how long the system runs on extreme days. The Manual J (with all assumptions made to overstate the load) came back 50% higher at 42,000 btu/hr. At the time, the system output could not go higher than 32,000 btu, and it heated our house without any trouble on an actual design day.

If I had time I would make a simple webpage that could run these calculations for different areas. But I have some toddlers to run after now. Cheers!

An easier way is to clock system run time on a design day. Do so after thermostat has been at the same set point for at least 1 hour. In most houses you will find the furnace runs less than 50% (aka 2X the size needed).

My old furnace was an 88k, never ran more than 5 minutes before hitting setpoint unless it was recovering from setback. New 44k furnace runs about 20 minutes out of 30 at 17f design conditions.

I wish I still had the info from my laptop. I took the local temps and did my own calculations to come up with what man j uses. It was quite a bit different than the published temps they have. Here the book design temp is 99*. When I ran the local temp data it was 105*. Which is why so many houses here are hot in the summer. It's very common to up size units on relativly new houses. A friend of mine does quite a few. Also some city's have started to spec design temps that are higher than in the man j calculators as they know they are off.

As for the muggy days in the 80s. Well here there are so few that it doesn't matter. It goes from nice and the windows open to mid 90s like a light switch then summer gets here and it's over 100* for 8 hours a day with highs up to 117 some summers and often for weeks on end with those highs.
Man j is a great tool but you have to keep a skeptical eye on it and don't believe the numbers provided are correct for your locality. It even says as much in its own documntation.

Our old system was a 3 ton (dying compressor)/100k furnace which just wasn't enough in the summer. The new one is a 3.5 ton 16 seer with a 100k two stage furnace 96%. Should be a night and day difference this summer. I also had to move 3 registers. The one in the kitchen and dining room were in really bad spots and there was only one in the living room. They didn't understand much about airflow in 1960 or leikely didn't care...

What still surprises me is the gauge of the wire on furnaces now. They are using 18 ga wire from the j box in the unit to the board. That means a max current of about 8 amps. Much lower than the older models. The efficiency increases have made quite a difference in amp draw. Also the change to scroll from recip compressors. However I would have gone with a lower seer as we will be moving. Higher seer doesn't help home value and spending more on super high seer units never pays back unless your Elec is super pricey or you are solar.

Odds are VERY HIGH the 3 ton systems was only delivering about 2 tons. This is why the house was hot, it wasn't because the 3 ton wasn't big enough. To get the full 3 tons ductwork and system charge has to be right. 90% of the time they aren't right and system performance suffers severely. Hopefully they get the 3.5ton right and it delivers it's rated performance. If you notice a HUGE difference, you can bet the old system wasn't delivering anywhere near it's rated capacity/efficiency.

100k furnace is nuts for a house in Texas that's under 3,000sqft. Unhook the 2nd stage, you wont be needing it. Yes, you basically just overpaid for a 70k single stage. Surprised you spent the $$$ for 16 SEER/96% AFUE if you plan on moving. The 96% will NEVER pay for itself in TX. the 16SEER MAY pay for itself in a few years, depends on how much more it was vs the 14SEER.

Let me guess, the contractor used 500sqft per ton sizing. It's worked since the 1960's, why change now?

Our old system was probably in the 2 ton range or little better due to a bad compressor. I forget what the pressures etc were last time I checked it.

The system I put in was actually the cheapest system I could get. A 14 seer was actually more as would have been a 3 ton system. The system all in worked out to 3k. I put the unit in a different place as the old spot will be going away for a master bedroom remodel. I used to do commercial refrigeration and a friend of mine has an hvac company. I could have gone with a 13 seer as there are a few the warehouse had that were within the grace period but I didn't want a recip compressor.

Oh and yes we do need the 100k btu... The old one was a 100k and a couple years ago when we had a week of 17* for the high it was running pretty much non stop. Also the furnace I used was the cheapest that didn't need combustion air.


All the duct work came out.. All done with 3' sections and poorly placed registers. M&M is close and they had some stuff that wasn't picked up I got a deal on. I hate doing metal duct in a 100* attic so my friend need up doing it... Already that hot up there this time of year.

How big is the house? Unless its 3,000sqft or REALLY leaky, a 100k furnace shouldn't have to run wide open with a 17f outdoor temp.

Are you on the high plains? If so, I'm sure the wind isn't helping.

The issue is its a pier and beam built in 1960 and back then houses were built poorly... I have added onto the house and done all the usual while doing it. Only thing I haven't finished is the insulating of the perimeter concrete walls and the rest of the plastic on the ground. That adds allot of heat and cooling load.
While not on the plains it is somewhat windy. Especially in summer... Funny how when the temps are over 100* people say it's not so bad because there is a breeze... And I have to remind them that the wind and heat is how a convection oven cooks so much faster...

The cooing problems here aren't so much the day time highs. It's the fact that it doesn't cool off at night. It will be midnight and it's still over 90*. The lows will briefly get to 80 to 85 but by 9 am it will be back to 90*. Now you know the main part of why we are moving. I don't like to be hot and it's hot allot here.

Thanks for these informations.

[QUOTE=Elcam84;49968 The lows will briefly get to 80 to 85 but by 9 am it will be back to 90*. Now you know the main part of why we are moving. I don't like to be hot and it's hot allot here.[/QUOTE]


When you mentioned 17 F highs I thought of Amarillo or Lubbock, but your definitely not on the high planes. I live near Charlotte, NC and I love the weather here, but I still miss those cool and crisp summer evenings we enjoyed in the Texas panhandle. We never needed AC after about 7-8 pm, just open a couple windows.

With manual j there's no heat gain from underfloor crawl space. Heat loss for winter is there but still minimal compared to other losses. I've been experimenting with sealing up my crawlspace. This winter I blocked all the underfloor crawl vents with rigid insulation board. It seemed to help, my gas usage was lowest ever, but we also had a ridiculously mild winter. I'm keeping the crawlspace blocked off thinking it couldn't hurt unless it gets too moist under there. My house was built in '58, its taken some doing but not impossible to make efficient.

The best solution for a crawlspace is to total encapsulation however there are some exceptions. Basically pool liner over the floor and just up the walls and foam insulation around the perimeter walls and piers. Lots of good info over at building science site.

I have about 90% of the floor covered and closed off the vents and monitored humidity and it does make a difference in heating and cooling. The floors used to be noticeably warmer in the winter and cooler in the summer with the vents closed.

If we end up with another pier and beam house in NC when we move it will be part of the negotiation that it has the crawlspace encapsulated or knock off $ and I will have it done. I'm not doing another one myself...

You still might want to look at your manual j numbers before hand. Closed and insulated might be the best, but in a southern climate the gains aren't much for the trouble. Money and effort could yield you faster returns on other areas vs what only affects me 3000bthuhr during 1-2 months of winter. I really can't measure the difference where my floor is insulated and where it isn't, but its also about outdoor ambient and even wind speed.