Click here to Find a Contractor in your area.
Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit

About design day temperatures...

Elsewhere,<span style="color:#000000"> Mark Eatherton posted, in part, "... </span>based on worst case scenarios that rarely occur for more than 2% of the time." I have read that these design-day numbers are such that it is warmer 97.5% of the time, which agrees with ME pretty well.

Where I live in New Jersey, this temperature is 14F. but it gets colder than this

"fairly often." What I am unclear about these numbers is <span style="text-decoration:underline;">97.5% of what</span>? If it refers to an entire year, this is almost 9 days. If it refers to the heating season, which is what I had assumed, it would be less. But this would open up a new question: how long is the heating season?

So far this year, it has been under 10F twice, and one of these days, it went down to 7.5F according to my cheapo Radio Shack Chinese indoor-outdoor thermometer whose outdoor probe is in the shade all the time. I did not notice how many days it has been under 14F.

I suspect that this 97.5% number is for the entire year, so that if it went below that temperature 8 or 9 times, I should not be surprised. And it need not do this for an entire 24-hour day, but just a short time in that day. For me, it tends to be at night before 7AM.

 Is this clear and correct? If not, please enlighten me.


  • Brad White
    Brad White Member Posts: 2,393
    edited January 2011
    You hit on a good subject

    The percentage number is the number of hours annually at or above a certain temperature. "Hourly Occurrence" is the term and it is across an entire year.

    Energy codes are a bit inhumane in some ways. For example, our MA Energy Code says Boston's design temperature is 7 degrees F. Cold enough but historically and habitually, everyone knows it can get colder.

    Personally, I have yet to see these relatively minor difference figure be the difference in a residential boiler size and if used to size HW radiation and they increase a bit, it just means a lower temperature water is needed. With modulating boilers, to a degree, you have the boiler you need when you need it.

    That said, it is also understood that once a building is warmed up and we hope relatively tight, the temperature dipping for a day to ten degrees below design, probably will not produce the same drop in inside temperature, despite the theory. The reality is that lights, people, cooking, etc. plus the building mass as a thermal battery, will even out those dips. A few days at those temperatures, yes you will see an effect. You must, if all is sized precisely for the design temperatures. But any boiler has an increment and therefore, some margin, as small as a good designer can make it.

    For most locales, the design heating temperatures are grouped as 97.5%,  99%, 99.6% and the median of extreme lows.

    I have weather data (USAF) packets in PDF format for six NJ locations, if you can tell me which is closest to you, I will post.

    Atlantic City


    McGuire AFB



    Trenton (Trenton Mercer AP)

    As an example for Newark, the 97.5%,  99%, 99.6% and the median of extreme lows are, respectively:

    19F, 14F, 10F and 5F.
    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"

    -Ernie White, my Dad
This discussion has been closed.