To heat or not heat upstairs

[Gary in NJ]

I have a three-zone heating system; zone 1 is the general living area on the first floor, zone 2 is for the bedrooms on the first floor, and zone three is the second floor that has a great room and two bedrooms and and bath. The great room is 700 sqft and is connected by an open stairway to the 1st floor foyer.

My wife and I are now empty nesters - no one lives upstairs anymore - so there isn't much reason to heat the upstairs more than 60F. But my concern is if the upstairs is colder than the downstairs, then I will create an airflow from the 1st floor to the 2nd.

Is it a false economy to think by lowering the temperature on the second floor that I'll save money on oil? Will the rise of the hotter 1st floor air to the 2nd floor cause the the 1st floor heating zones to cycle more often in attempt to keep up with the traveling air.

I've always kept the two floors at the same temperate; 68 during the day and 62 at night, and I've never noticed that the upstairs is hotter then what the thermostat is set. So when the temps are the same there seems to be balance.

 

[Dana]

Lowering the temperature will save energy, no doubt. Lowering the temperature TOO much could lead to mold issues, but if the floor below is fully heated that's probably not going to be a problem. If you keep it at 60F it won't be a problem, but under 50F you'd need to monitor things for condensation or mold.

Warm air will rise, cold air will fall, so yes there will be both air rising and falling in the open stairway that increases with temperature difference. If the great room is kept at 55-60F and the first floor is at 68-70F there will be some convection but probably not at a velocity high enough to feel drafty anywhere downstairs.

Almost all oil boilers are oversized for the heating load, and WAY oversized for the radiation of a single zone. There may be boiler short-cycling issues to sort out (there may already be some), but with a better description of the zone radiation (size & type) for each zone, and the boiler's DOE output number we'd be able to assess the risk.

Is the boiler located in a basement under the first floor? If yes, is the foundation insulated? An oversized boiler serving only 1-2 smaller zones will be delivering a huge fraction of the heat as standby & distribution losses. If the basement is insulated those losses aren't fully "lost", but if the basement isn't insulated and you're only heating the first floor it could easily add up to half. If the boiler room is the warmest spot in the house in the winter it's time to think about doing something different.

If the house doesn't have air conditioning and you'd like some, a cold-climate ductless mini-split heat pump on the first floor can provide most or all of the heat for that zone, and save quite a bit on heating expenses, even in high cost electricity markets, and deliver very high efficiency cooling as well. The boiler would still be there as backup for when it's REALLY cold outside.

For a snapshot on what the actual whole-house heat load is (including distribution losses), running a fuel-use heat load calculation (winter fill-up dates & quantities only) can give you a fairly good approximation of what your 99th percentile load is. With your thermostat settings use base 60F weather data for the load calculation. If you're on a regular fill-up service that stamps a "K-factor" on the slips, a few mid to late winter K-factors would be another good data set. The units are K-factors are base-65F heating degree-days per gallon, so with the boiler's nameplate efficiency numbers it's easy to convert that in to BTUs per degree-hour. (A smart 5th grader can do that math.) If you like I can run the load numbers here for you.

 

[Sponsor]

 

[Gary in NJ]

Wow, what a well thought-out reply. Thanks. My basement is insulated and stays a steady 68F all year long. My house is 4,200 sqft (2,600 1st and 1,600 2nd floor) and the boiler (WM Gold P-WGO-4 Series 3) is 145k BTU, net I=B=R rating of 128k. Each of the 1st floor zones have about 50 LF of baseboard, and the 2nd floor as about the same 50LF. There is a 4th zone on the system for domestic hot water as well. I use about 1,000 gallons a year of oil, but I estimate that 350 of that is consumed for hot water (it's about 1 gallon per day). As long as I'm giving info, the 1st floor has 8' ceiling and the second 7-1/2' ceilings with knee walls (it's a cape style). All vertical walls are insulated to R-19 and there is significant insulation on the flat portion of the ceiling/roof, but perhaps only R-19 on the slopes. I'm also insulated between floors with R-28 as I didn't finish my second floor when I first built the house. The house is fairly tight with the exception of three sets of French Doors that are somewhat drafty and the air conditioning return is installed directly above the zone 1 thermostat (some years I remember to cover it...some I don't. This year I AM covering it).

Now, with all of that info, I believe that my boiler does short cycle during the winter. In the morning when the call goes from 62 to 68 on all three zones it runs for a good long time. But mid-day, if one zone calls to raise the temperature 1F - the boiler may run for just a minute. The system is set up with a single circulation pump. Not the way I'd set it up today....but it is what it is.

 

[Dana]

With the basement insulated it's more relevant to use the DOE output, not the IBR-net numbers, since the standby & distribution losses are still indoors, and actively heating the first floor. (Think of the distribution plumbing as a long skinny radiator. ) For your boiler that's ~140,000 BTU/hr.

A decently insulated 2x6 1.5-2 story 4200' house with at least R20 in the attic and clear glass double panes (or clear storms over single pane double-hungs) and an insulated basement would come in with a heat load of about 50,000-80,000 BTU/hr @ 0F, and proportionally less at +10F (or whatever your 99% outside design temp happens to be.) So you're probably in the 2x oversized range for the whole house load, which isn't terrible, if less than ideal. ASHRAE recommends 1.4x oversizing factor of the 99% load as the best compromise for non-modulating boilers.

The 100' of baseboard on the first floor can deliver about 50,000 BTU/hr of heat at a boiler temp of 180F. It probably doesn't need to be that high, but both zones 1 & 2 together only add up to a bit more than 1/3 of the boiler's output. Even adding in the 50' for the third zone you'd be at only about half the boiler's DOE output, so yeah, it's doing a lot of unnecessary cycling on zone calls AND for the whole house load. With an indirect water heater zone as the "priority zone" the biggest boiler that should ever be hooked up to 150' of baseboard would be about 75-80K of DOE output. That's enough burner to deliver a continuous full-flow shower, and unless you're showering for an hour at a time you'll never miss the drop-out of the heating system when it's serving the indirect.

Annual fuel use numbers aren't really very good for estimating heat loads without using a more sophisticated boiler model. There is a lot of error related to hot water use and solar gain in the shoulder season. Mid to late winter fill-ups reflect a higher duty cycle, low solar gain, and a lions share of the fuel going toward space heating. So if you have some K-factors or billing slips from the mid-winter to at latest April Fools day we CAN figure this out. With only 75,000 BTU/hr of baseboard and some standby/distribution loss heating it's pretty clear that the whole house load is no more than 80K, but it could be only 60K (even when fully heating the upstairs. When dropping the temps upstairs your oversize factor is probably going to be closer to 3x than 2x.

Given your limited zone radiation it's worth retrofitting a heat purging boiler economizer control (eg Intellicon HW+ , there are others) and setting the low-limit to 140F (any cooler and it risks copious acidic exhaust condensation in the flue). This type of boiler control maximizes the use of the available thermal mass in the system & boiler. On a new call for heat it purges heat from the boiler until it hits the programmed low temp, but allows the high temp to run as high as it needs to. When it "learns" the system based on recent burn history it anticipates the end of a call for heat cutting the burner early to park the boiler at a lower temp between calls for heat, thus reducing standby losses. Between the reduced number of ignition cycles and the lower standby loss the overall burner on time will in your case likely be cut by 10-15%, possibly more.

Many/most new cast iron boilers come with heat purging controls, ever since this bit of research at Brookhaven Nat'l Labs was published about a decade ago. See Table 3. At an oversize factor of 2x most boilers run 5-10% below their AFUE numbers even without the short-cycling issue. System #3 was the only one with heat purge control that was tested, dropping only 1% in as-used efficiency even at an oversize factor of 3x.

The fact that the basement cruises along at 68F indicates fairly high standby & distribution losses. A heat purging controller will reduce the basement temp by a few degree too, which lowers the actual heat loss. (You don't really care if it's 65F down there instead of 68F, do you?)

Is the air conditioning broken up into zones? Is it recent enough to use R410A refrigerant, not R22 or R12? Are any of the ducts or air handler on the other side of the insulation, say behind the kneewalls? (Sounds like it might be if it's cascading cold air onto the thermostat.)

 

[Gary in NJ]

I have two zone a/c in the house; upstairs and down stairs. Both systems are installed on the second floor in a utility closet and all of the supply ducting run in the knee walls. The return for the second floor is adjacent to the air handler and the first floor return is half in the conditioned space and half in the knee wall. The systems are 16 and 9 years old, so they aren't the most efficient, but in NWNJ the a/c only runs maybe a total of 6 to 10 weeks a year.

I'll do my research on a heat purging boiler economizer control. What I have now is a rather crude analog control and there's no doubt that there is room for improvement. All of my installed equipment is TACO, so I'm comparing the Intellidyne HW+ to the TACO SR-501-OR.

Thanks for the input.

 

[Dana]

The TACO SR-501-OR is an outdoor reset controller, not a heat purge controller. There IS a difference. If you had a right sized boiler (you don't) and more radiation per zone (not even close) it would be able to deliver some savings, but only up until the point where the temp was so low the boiler began to short cycle.

With your oversized boiler and limited zone radiation an outdoor reset approach would cause the boiler to short cycle even more. Even at a 180F output a 50' zone only emits about 1/6 of YOUR boiler's output. Most of the time you don't need the water to be that hot to heat the house, but the lower temperatures will likely reduce efficiency &/or the boiler's lifespan:

At 150F-out there is some risk of the return water being too cool, causing damaging levels of exhaust condensation, but at that temp the a 50' zone would then only emit about 1/8 of the boiler's output. Any time just a single zone is calling for heat it cycles on/off at a low duty cycle, with even shorter minimum burn times than you're experiencing at the higher temperature operation. During an ignition cycle it takes about 15 seconds for the combustion efficiency to approach the steady-state efficiency. If your minimum burns during a single zone call now is 1 minute, 25% of that minute is at a significantly lower efficiency. If the low limit is 140F (necessary for exhaust condensation control) and the thermal mass of the boiler + zone radiation is only slewing through 10F, instead of 20F or higher, and the combination of less heat being emitted (= more excess heat going into the thermal mass) means the temperature is slewing faster and hits the high-temp sooner. If the temp is low enough that the minimum burn time to about 45 seconds, 33% of the total burn (15 seconds) is at a lower efficiency. If the temp is low enough that the minimum burn time is 30 seconds, half of the total burn is at lower efficiency.

That's what makes short-cycling is so damaging to efficiency. Ideally you'd have 10 minute minimum burn times, less ideally 5 minutes, but that's simply not going to happen without more thermal mass in the system, and using that thermal mass effectively. A heat purge controller can't add mass but it makes the most of what is there.

At the beginning of a call for heat the water is pumped through the boiler without firing until it hits the low-limit temp. But when it fires it will allow the boiler to heat up to 180F or even higher, (your boiler is safe to operate up to 210F) and it takes more time time to get there. The burner turns off at the high limit, but doesn't re-fire until the boiler's temp reaches the low-limit again. If that's 140F low, 200F high it's 60F slew instead of 10F (in the 140F low 150F high outdoor reset case) but it takes more than 6x as long. If it was burning 30 seconds at a 150F high limit, it'll burn more than 200 seconds with the heat purge controller. That's still a short cycle (5-minutes would be better), but the inefficient ignition period is now only ~8% of the burn. It's also dramatically fewer burns before the call for heat is satisfied.

From the history recent calls for heat the the controller anticipates how many low-to-high burn cycles to expect before the call for heat will be satisfied, and how long it takes, so it can cut the burner early to finish the call with heat purged from the boiler, abandoning less heat in the boiler & distribution plumbing.

So, while there is more jacket losses and distribution loss during the burns, the average water temp of the boiler & distribution plumbing is lower than it currently is, maybe even lower than when running an outdoor reset approach.

 

[Gary in NJ]

I think its fair to say that would not have extracted that data from the brochures or installation manual. It looks like the HW+ is a worthy use of $180 and 30 minutes of my time to install. Thanks again.

 

[Gary in NJ]

I ordered the Intellidyne HW+. I'll be hooking this up between my TACO 404 zone controller and my Honeywell L8148A Aquastat. Any advice or hints to help ease the installation?

 

[WorthFlorida]

....Is it a false economy to think by lowering the temperature on the second floor that I'll save money on oil? Will the rise of the hotter 1st floor air to the 2nd floor cause the the 1st floor heating zones to cycle more often in attempt to keep up with the traveling air....

You may know it or not but with both floors set to the same temp, how long does the second floor zone run compared to the first floor zone? Other than the coldest days, the second floor may not run hardly at all since the heat is rising from the first floor. I would at least keep the doors closed to all the rooms on the second floor so only the stairway and hall area are being heated from the first floor. Set the second floor thermostat were you feel comfortable at and on the those real cold blustery days check the temperature of the closed off rooms so as Dana stated to minimize the chance of any condensation problems.

 

[Gary in NJ]

Dana,

I was giving some thought to calculating the actual load...and I remembered that back in the winters of 2012/13 and 2013/14 I kept some impeccable records of fuel use. At the time I was trying to relate outside air temperature to fuel consumed. The winter of 2013/14 was the winter of the "polar vortex" so we had a week of weather where the night time low went to -5 to -12 degrees. My data captures average daily highs and lows for the period between fill-ups.


Date In Tank | Added | Used | Daily | High | Low
20-Nov-12......580...........0............0......0.0........54......32
27-Nov-12......560...........0...........20......2.9........46......31
19-Jan-13........175...........0.........385.......7.3........44.....30
29-Jan-13.........50.......200.........125......12.5........31.....13
10-Feb-13........175...........0...........75.......6.3........37.....22
20-Feb-13.......100.......300...........75.......7.5........41.....23
2-Mar-13.........330..........0...........70.......7.0........42.....28
4-Aug-13.........160...........0.........170........1.1........75......45
13-Oct-13..........80...........0..........80........1.1........82......65
Season Total 1080

21-Oct-13..........70........500...........10.........1.3......68.....44
21-Nov-13.......435............0..........135........4.4
1-Dec-13.........380............0............55........5.5.......44.....27
30-Dec-13.......170............0..........210........7.2.......43.....25
6-Jan-14...........79........536............91.......13.0.......30......8
12-Jan-14.......560............0............55.........9.2......38.....20
1-Feb-14.........400............0..........160........8.0.......35.....16
28-Feb-14.......155............0..........245.........9.1.......36.....16
10-Mar-14........80........300............75.........7.5......38......17
31-Mar-14.......245............0..........135.........6.4......47.....27
30-Jun-14.........70............0..........175.........1.9........0.......0
Season Total 1336

 

[Dana]

Got a ZIP code, to be able to come up with the most relevant weatherstation dataset and 99% outside design temperature?

 

[Gary in NJ]

08802

KABE, KNJSTEWA2 or KNJGLENG1 seem like they'd work. I've seen drybulb temps of 6 to 10F for my area.

 

[Dana]

08802

KABE, KNJSTEWA2 or KNJGLENG1 seem like they'd work. I've seen drybulb temps of 6 to 10F for my area.

So looking at primarily winter use volumes, it looks like between 21 October 2013 and 10 March 2014 you burned through 836 gallons. At at 138,000 BTU/gallon and 85% efficiency that's about 98 million BTU (MMBTU).

Unfortunately degreedays.net can only look back 36 months, but with a bit more work we can dig out that info from weatherdatadepot.com. (Set the base temp to 65F the comparison years to 2013 & 2014, click on the month tabs to find the daily HDD for the last 10 days of October and the first 10 days of March.)

For 22 October 2013 through 10 March 2014 they they logged 4557 (65).

So that's 98MMBTU/4557 HDD= 21,505 BTU/HDD or (/24=) 896 BTU per degree-hour.

Using +6F (the 99th percentile temperature bin for nearby Phillipsburg) and the presumptive 65F heating/cooling balance temp you have 59F heating-degrees.

59F x 896 BTU/F-hr= 52,867 BTU/hr for the implied heat load (including distribution & standby losses) , which at a load/floor are ratio of ~12-13 BTU/hr per square foot of conditioned space is a credible number for 4200' of reasonably tight, 2x6/R19 construction house with an insulated basement. (You didn't spend three weeks in Belize in January 2014 with the thermostats set to 50F, right?)

If you were using deep setbacks to 55-60F and only heating the place to 65F during the day you could repeat the exercise using base 65F data then add 5F heating degrees to see what the load if kept at 68-70F, but it won't change by more than about 5-7K.

So with your boiler's DOE output of 145K and a load of about 50K you're nudging up on a 3x oversize factor even before you turn down the upstairs zone. With the lower temp upstairs it'll be more like 4x+ oversizing for the as-used load. The Intellicon is definitely going to be saving you some oil in addition to the savings from lowering the temperatures upstairs.

 

[Gary in NJ]

Looks like it's time to add on to the house...or leave a window or two open.

It's nice to know that the house is efficient. When we built the house in 2002 I remember that the plumber that helped me install the boiler questioned my calculations (which at that point are a guess because we haven't burned a drop of oil yet) as they pointed to a boiler in the 115k BTU range. He talked me into going bigger...so it's my fault for not trusting the math.

Hopefully I never have to replace my boiler, but if I do I know to get a much smaller unit.

My Intellidyne HW+ will be here tomorrow. I got it on eBay for $150. I didn't realize that Intellidyne is out of business, so finding a DHW sensor turned out to be an ordeal. In my research on the HW+ I learned that the Beckett 7212 was manufactured by Intellidyne, so I was able to locate a Beckett 7218 temp sensor. It should all be connected by the weekend.

Thanks for the help. After the winter I'll be seeking out your advice for replacing two under-counter kick units in my kitchen with a radiant system.

 

[Dana]

Somehow I'd missed that they've gone under too! There has been a lot of competition that showed up in the past decade, and once boilers started being shipped with smarter controls as standard equipment I guess the writing was on the wall!

Using ASHRAE's standard oversizing by 1.4x the 99% load, if the ~52K number is right that would call for a boiler with ~73K of DOE output, which would be an ~83-85K-in boiler.

Looking over your shoulder... With only 150' of baseboard a boiler with 115K of output would have had to run about 215F output to balance perfectly. If 115K was the input number, the DOE output would have been ~98K , which would balance with the radiation at about 200F output.

It's sad but true that many heating & plumbing guys use dumb rules of thumb such as "25 BTU per square foot" for newer construction, "35 BTU per square foot" for 2x4 construction when sizing a boiler, then pick the next size up, when actual heat loads are typically under 15 BTU/ft^2 for new construction, under 20 BTU/ft^2 for 2x4 stuff. That reliably heats the place, but they're not doing you any favors.

For new construction the right way to do it is for the architect/engineer to run an ACCA Manual-J or similar load calculation using fairly aggressive assumptions on R-values, window U-factors, and air tightness. Even a Manual-J will often overshoot measured reality by 10-15%, but upsizing by 1.4x from the Manual-J still leaves you at no more than 1.6x, which is the presumptive oversizing factor in AFUE testing.

 

[Dana]

. I didn't realize that Intellidyne is out of business, so finding a DHW sensor turned out to be an ordeal.

Are they really out of business? Their website still seems fully functional.

 

[Gary in NJ]

I read that info in several places. None of the suppliers (Supply House, Patriot Supply) on line carry their products anymore. They may show them, but there's no way to order them. The website is functional, but they haven't responded to two emails nor is there a telephone number. They may be reorganizing. I found one listing for the unit from a vendor called "sellingmystuff" - hardly a big retailer - and eBay had just three listed...now two. Sensors aren't available anywhere (one of my emails to Intellidyne).

It doesn't look good.

 

[Dana]

Sorry to hear it, but thanks for that info! (I think I still have one packed away somewhere...)

 

[Gary in NJ]

Dana, I did hear back from the sales department at Intellidyne, so someone is manning the ship.