Choosing size of Indirect Water Tank

[Robert Grenley]

Again, forget the first hour delivery- this is a first 15 minutes time.




Again, a 127MBH AHRI net-water boiler is ABSOLUTELY GROTEQUELY oversized for your house, unless you sleep with all the windows open or the house is more than 12,000 square feet.

Oversizing a space heating boiler that badly has a fairly significant (negative) impact on the as-used efficiency. DON'T DO IT! You probably don't even have enough radiation in the whole house to even emit that much heat at an entering water temp of 180F, which means it will be prone to short cycling into even lower efficiency. (Do you have 300 feet of 8-10" tall fin-tube baseboard, or equivalent?)

A typical 2500' house in Seattle has design heat load less than 30,000 BTU/hr (30MBH) , and a better tighter house that size can have a heat load under 20,000 BTU/hr (20 MBH). If you have a heating history on this place, run this napkin-math to get a handle on your actual heat load. Total up the amount of radiation in the house, zone by zone if multi-zoned.

If you're willing to share information on zone-by-zone radiation, exact meter reading dates & amounts, and a ZIP code (for more accurate weather data) it's possible to sketch out what boiler capacity might make sense here. Don't rule out modulating condensing boilers, which can be cheaper to install and easier to get a reasonable fit to both the radiation & heat load than Godzilla-sized cast iron.

Forget the recovery rate too- with a boiler right sized for the space heating load it's too slow to make a significant difference for the first 15 minute hot water capacity numbers.



The existing storage temp is about 140F, and the total volume is 75 gallons. Simply boosting the storage temp to 160F adds only about a minute to the showering time. (See response #11). If you're not installing drainwater heat recovery it's going to need the 115.

A $1000 up-charge is better spent on drainwater heat recovery, since it will increase the apparent capacity to more than that of the 115 gallon tank, has at least 3x the service life, uses half the amount of gas for showering, and won't increase the standby loss.

I didnt see this post until after my post below was written and posted.
The person who assessed my house for the boiler size, even before we started talking at all about the indirect water tank option, recommended the US Boiler Burnham Series 3 Cast iron boiler Model 306...Input 175 MBH, Heating Capacity 146 MBH, Net AHRI rating Water 127 MBH, AFUE% 84.

My 1910 house, completely renovated in 1987, is under 3800 sq ft...with just over 1250 sq. ft. on each of 3 floors.
Our oil burning boiler is serving 7 zones plus 2 bathrooms that have radiant water heat beneath the floors.
The windows on the weather size of the house are all new double pane, and the house is well insulated.
The windows on the other side of the house are mostly single pane leaded glass.
We have a total of 22-25 cast iron radiators (without taking off all the radiator covers, I can't tell if some of them are very large or doubles, hence the number of at least 22 and maybe 25...these are the old fashioned, refurbished heavy radiators).

I was trusting his recommendation, as I was referred to his firm because of his long (32 year) experience with boilers, of which several people in Seattle simply don't work with boilers.

You are saying that the gas boiler he has recommended is vastly oversized?

 

[Reach4]

Copper drain pipe and copper tubing wrapped around the outside. Minerals are driven out of the water at a high temp 120*. Recovery you have 40* entering and 55*- 65* out not hot enought to drive out the minerals.

Cool. Softened water might change that, since the minerals should mostly be soluble.

However any corrosion eat-through reaction should speed up with temperature.

 

[Sponsor]

 

[Dana]

I didnt see this post until after my post below was written and posted.
The person who assessed my house for the boiler size, even before we started talking at all about the indirect water tank option, recommended the US Boiler Burnham Series 3 Cast iron boiler Model 306...Input 175 MBH, Heating Capacity 146 MBH, Net AHRI rating Water 127 MBH, AFUE% 84.

My 1910 house, completely renovated in 1987, is under 3800 sq ft...with just over 1250 sq. ft. on each of 3 floors.
Our oil burning boiler is serving 7 zones plus 2 bathrooms that have radiant water heat beneath the floors.
The windows on the weather size of the house are all new double pane, and the house is well insulated.
The windows on the other side of the house are mostly single pane leaded glass.
We have a total of 22-25 cast iron radiators (without taking off all the radiator covers, I can't tell if some of them are very large or doubles, hence the number of at least 22 and maybe 25...these are the old fashioned, refurbished heavy radiators).

I was trusting his recommendation, as I was referred to his firm because of his long (32 year) experience with boilers, of which several people in Seattle simply don't work with boilers.

You are saying that the gas boiler he has recommended is vastly oversized?

Yes, the gas boiler he recommended is VASTLY oversized for the space heating load.

My house is 2400' of 1920s vintage 2x4 construction 1.5 story bungalow retrofitted with cellulose insulation in the cavities, maybe R15-R20 average in the mostly 2x6 cathedralized ceilings, plus 1600' of (retrofit insulated to ~R17 c.i.) basement, clear glass 1980s vintage storm windows over the original 6/6 wood sash single panes. At 0F outdoors, 70F indoors, the heat load comes in right around 40MBH for the whole shebang. If a boiler were in the basement (inside the thermal envelope of the house) with 146MBH of output it would theoretically not lose ground until it was -175F to -200F outside. (It hasn't gotten that cold in the 25 years I've lived there, so I can't say for sure. ) Even with your single-pane windows it's unlikely your design load at +25F outside is over 50 MBH unless the house is REALLY air-leaky (and that could be fixed.)

The high thermal mass radiation is probably what drove the boiler sizing to such a ridiculous level for a cast iron boiler, to ensure that it (over) heats fast enough to avoid chronic entering water temperatures at the boiler that would cause destructive corrosion from exhaust condensation on the heat exchangers. But that's the RONG way to treat the problem. The original system was intentionally oversized for the heat load for the house even when it was all single pane with little or no insulation. Even without upgrading the house it was common to paint the rads with silver or bronze paint to reduce the heat emittance or put covers over then to keep the occupants from roasting.

A right sized condensing gas boiler would take advantage of continuous low temp operation, putting the heat of vaporization of the condensing water taken out of the exhaust into your heating system, delivering much more stable room temperatures (and pretty amazing efficiency.) With a 50K load and a 146K cast iron boiler the duty cycle on boiler only hits 35% at design condition. AFUE testing assumes about a 35% duty cycle average over the entire heating season. The system probably overshoots the setpoint temps on the thermostats by more than a degree, then it's a long wait for the next cycle.

With 7 zones it may be short cycling the oil boiler on zone calls, if it's in the same order of magnitude output of the recommended 6 plate Burnham. With a right size condensing boiler with the outdoor reset dialed in it takes long enough for the zone thermostats to be satisfied that zone calls will almost always overlap, and the average duty cycle will be over 75%. The radiators will ALWAYS be warm, but just warm enough to heat the place. There are many 80 -100 MBH condensing boilers out there with 10:1 turn down ratios, so the min-fire output can be under 10MBH, which is important for comfort & efficiency when it's cut up into 7 zones.

If you can share oil fill-up dates and amounts (wintertime fill-ups only) it's possible to infer the actual heat load from oil use. If you're on a regular fill-up service that stamps a "K-factor" on the billing slips, a few wintertime K-factors is enough information. Otherwise, take couple of stabs at it using LoadCalc or CoolCalc. Both of those tools will overestimate reality, often by quite a bit unless you select for the tightest possible construction options and fairly aggressive assumptions about R-values/U-factors, but you have to really screw it up for them to come up with a number that's 2x reality. Any number spit out by those tools should be considered a hard upper-bound, and the boiler need not be upsized from there (and should not be upsized if the boiler is non-modulating.)

BTW: Installing l0w-E exterior storm windows over those leaded glass windows and other single-panes brings the performance almost up to current code minimum for a fraction of the cost. The hard-coat low-E coatings used for those types of storms cuts very little out of the visible spectrum, but reflects the deep infra-red back to the indoors (where it belongs.) Even the Larson low-E windows sold through box stores are pretty good, and come in a range of powder-coat painted trim colors to work with a wide range of exterior cladding & trim color schemes.

 

[Dana]

Cool. Softened water might change that, since the minerals should mostly be soluble.

However any corrosion eat-through reaction should speed up with temperature.

I have a few pieces of original copper drain in part of my house that's now 97 years old, and some copper pipe more than 50 years old. It was the old sections of lead drain that eventually failed. A drainwater heat exchanger is just copper drain, copper pipe, in a funny configuration. The warm-side output of a drainwater heat exchanger (even a pretty big one) will always be under 90F in this application, probably under 75F in mid-winter. The warmest part of the drain will probably hit a bit over 100F during a long gusher shower, but never more than 105F (unless you normally look like a cooked lobster). The duty cycle of that drain is about 1% (assuming one 14-15 minute shower every day.)

An indirect is usually good for 20-25 years. The higher temperatures induce liming & corrosion, and the duty cycle at those high temperature is nearly 100%.

So yes, a drainwater heat exchanger will outlast an indirect several times over.

 

[Reach4]

I have a few pieces of original copper drain in part of my house that's now 97 years old, and some copper pipe more than 50 years old. It was the old sections of lead drain that eventually failed. A drainwater heat exchanger is just copper drain, copper pipe, in a funny configuration.

Then there is this recent thread: https://terrylove.com/forums/index.php?threads/brass-to-copper-to-brass.85508/

This was fed from a closet flange, and I think the users might have tried to save water too much.

I used to have copper pipes and drains. No problem with those 50 year old pipes either.

 

[Robert Grenley]

I called the oil company and they don’t use K factor, or perhaps they don’t share?
But we used 1200 gallons of oil last year, and 1250 gallons of oil the year before.
Zip code 98119.
If their calculations are correct.

 

[Dana]

I called the oil company and they don’t use K factor, or perhaps they don’t share?
But we used 1200 gallons of oil last year, and 1250 gallons of oil the year before.
Zip code 98119.
If their calculations are correct.

Not all fill up services bother with a K-factor, though it's more common than not in the northeast.

Gallons per year isn't going to be good enough. The exact wintertime fill up dates and amounts are what matters. The shoulder seasons have way too much inherent error from things like solar gain, and an extremely low duty cycle, etc. Winter is when the duty cycle is highest, for the lowest idle time/standby loss, and also the time with the least amount of solar gain, and fewer days that have no heat load, or even a cooling load.

 

[Dana]

Taking a stab at it (knowing that the result will be GROSSLY exaggerated).

Looking at the data from the Boeing Field weather station on degreedays.net, between 1 Sept 2017 and 1 July 2018 it logged 3100HDD (base 65F). From 1 Sept 2018- 1 July 2019 it logged 3134 HDD.

So the boiler used (1200 +1250)/(3100 +3134)= 0.383 gallons per HDD. A gallon of #2 oil runs about 138,000 BTU/gallon, so burned at 85% efficiency (your boiler probably isn't that efficient, but maybe on it's best day ever it was), that would yield 117,300 BTU/gallon. So the 0.383 gallons delivered (117,300 x 0.383=) 44,926 BTU per degree-day.

In 24 hour day that would be (44,926/24)= 1872 BTU per degree-hour.

With a 99% outside design temperature of 28F (the 99th percentile temperature bin at Boeing Field) and a presumptive heating/cooling balance point of 65F (the degree-day base), it's (65F-28F=) 37 heating degrees...

...which makes the presumptive heat load 37 degrees x 1872 BTU per degree-hour = 69,264 BTU/hr, absolute worst-case, including all standby & distribution losses. In reality that number is significantly higher than reality, since very low numbers of degree-days in Sept-Oct and May-June means that MOST of the oil used during those months was standby loss, and your oil burner might only be delivering 80% efficiency.

So, call it a design load of 70K worst-case, even though reality is likely less than 60K. Per ASHRAE the optimal oversize factor is 1.4x the 99% load, so for a 70K load the biggest boiler that would make sense has a DOE output of 1.4 x 70K= 98,000 BTU/hr. If it's (more likely) in the 55K range, the biggest boiler that makes sense would have a DOE output of 1.4 x 55K = 77,000 BTU/hr.

Run a CoolCalc or LoadCalc load calculation- see how it stacks up against the crude annual fuel use analysis. Those tools typical oversize from reality by 15-35% (more if you're not aggressive about the inputs). But one thing is certain: The heat load CAN NOT be higher than the source fuel input BTUs that actually heated the place. (The rare exception would high-R houses with lots of thermal mass and passive solar gain designed in- which is the opposite of your house.)

There are several ~100,000 BTU/hr modulating condensing boilers out there with 10:1 turn down ratio that would almost certainly be able to run in condensing mode on your high-mass zone radiation without short-cycling at condensing temperatures despite the micro-zoning. (eg: HTP's UFT-100W, Lochinvar's KHB/WHB-110, Navien's NHB-110, IBC's SL 14-115 G3, there are others.) Any of them would be more appropriate than a 6 plate Burnham Series 3. Of that series the biggest that makes any sense from a whole-house load point of view would be the 304, with a DOE output of 88,000 BTU/hr, but since it isn't inherently protected from condensation from cool return water it's better to go with a mod-con.

At both internet and distributor pricing the UFT-100W or NHB-110 is several hundred USD cheaper than a Burnham 306 (or 304), can use cheap plastic venting, and doesn't need low return water protection features designed/built into the near-boiler plumbing.

 

[Jadnashua]

Ideally, your boiler output would match the heat lost in the house and run constantly. THis is the most efficient and most comfortable. Things get worse and worse as the thing turns on and off.

We can't reach that ideal state, nor, would one that was too small, be able to recover from a deep setback very quickly, but on some systems, a big setback isn't all that efficient, at least when comfort plays a part. In reality, most any boiler you have will likely be oversized per that ideal. The goal is first to have enough so the house can stay warm when you want it, and not be too big, which costs more up front to buy and in comfort levels that drop since it easily overshoots, and must shut itself down.

Think of boiling water on the stove...to get it to boil faster, a huge burner may help, but once it's boiling, you turn it down, often, way down to maintain. You don't need a monster boiler to bring the temperature up quickly, only one that can easily maintain without excessive cycles. Plus, you don't have the ability to release all that much energy, so a big boiler is then guaranteed to need to cycle so it doesn't overheat itself since it can't radiate it into your home...just not enough radiators.

lt's sad that so few boiler people actually have faith or understand how to size things. It's much easier to just use some rule of thumb that is guaranteed to oversize things. Then, they'll never get a call after installation saying the boiler can't keep the house warm. It's much easier for them to ask...did the old boiler keep things warm, and if the answer is yes, just replace it with a similarly sized unit. The math isn't hard, but it does require a bit of understanding and extra work not that many will do. It's not really very complicated.