Hey all,

Come tax-return time, the wife and I will be pulling the trigger on a new boiler.  Our "builder grade" unit went out on us in February last year.  We had bought the house in January.  Sweet.  Anyhow, the existing boiler was installed in 2000, and I'm assuming there are some components that are still good on it.  The circulating pump, expansion tank, what have you.  I've been searching for a readable diagram specific to condensing HE units, but haven't had much luck . . .

I'm looking for guidance on what I'd need if I was starting from scratch installing the new unit (I don't entirely trust the work of the plumber who installed the old one, as it got quite a bit less mileage than it should have).  Also, which components can I most reliably scavenge and how can I test them to see if they're shot?

System details:

1000 sq. ft. slab-on-grade, masonry home in Denver, CO
Tubing is cast-iron in the slab and I've NO idea how big it is or how the loops are spaced
Single zone
Existing boiler is 106kbtu output, but we're installing 2" of foam on the exterior of the house and I hope to use an 80kbtu condenser for the replacement.
yes, the tax-credit is a partial motivator

We're doing a wall-mount to save space in our tiny utility room


We're currently running our wood-stove as primary heat and staying reasonably comfortable - thank God the wife is as much a glutton for punishment as I am.  We figure if we can maintain 55* with the floor, getting above that with wood, we'll be happy, though selling the house down the road will be easier if the purchaser can use the boiler alone without destroying it.

If you are replacing a boiler the "trim" is very important of little cost and the boiler size you mention will heat three of your houses!

The thing to be concerned about first is the heat load after the all-important perimeter insulation has been installed. Near boiler piping and pump size is next. You should have a professional do the heat load, specify the boiler and draw the piping schematic before proceeding. And don't forget the companion water heater. If you have a 95%AFUE boiler you will want it to do all the work it can.

Actually, you can do a better job of calculating the pre-improvement heat load than any estimating tool if you have at least a few months (whole heating seasons are better) to correlate against weather data. From that you can using reasonable duty-cycle derating models of the boiler and use it's specifications to determine it's average as-used operating efficiency, then calculate BTUs of output per heating-degree day, to apply to the appropriate outdoor design temperature. The error for the whole-house/single zone number will be well-within 10% of reality. Most estimation tools overshoot by at least 25%. Did the seller provide any utility-use information (preferably with dates on 'em to correlate with weather data)? With that info you could place a very confident upper bound on the boiler output requirements that may be lower than what an estimation tool would deliver on your POST insulation-improvement heat load.

Either way, I'm guessing that the current boiler's 85K output is 2.5-3x oversized for the PRE-improvement heat load, and would likely be 4x+ oversized after nearly doubling the R-value of the entire wall area. This means you're probably looking at the very smallest of the line for any of the commonly available condensing boilers out there, and even those will be at least 1.5-2x oversized for your design-day heat load. (This is true for me as well.) I agree with BadgerBoilerMN's recommendations- adding an indirect-fired HW heater off the boiler is definitely the right thing to do, as it will improve the as-used AFUE of the inevitably oversized condensing boiler by giving it more load to work with. The combined system is more efficient, recovers more quickly, & has a longer lifespan than a standalone tank, and will be as efficient as an on-demand tankless, but with nicer flow characteristics for a lower installed cost & lower maintenance. (There are tax incentives for those too.) Size the tank for the biggest tub you have to fill- if you don't have a big spa or soaking tub a ~30gallon unit will usually do. Even the smallest condensing boiler's output is 1.5-2x that of a typical standalone 40-50gallon tank, and it's first-hour gallons rating will be similar to the bigger standalones.

And by going with an indirect using a direct-vented condensing boiler, you're no longer incurring the induced infiltration of the open flue of a hot water tank and the combustion+ dilution air of the atmospheric drafted cast iron boiler, which adds up to several percent of the typical annual heat loss.

Let a pro figure out the piping & pump sizing issues, etc. You may/may-not be able to use the existing circulation pump with a newer-smaler boiler, but a decade old pump is already at 50% of typical lifespan, uses more electricity than newer-better-more-appropriate designs, and generally isn't worth saving. The real money is in the boiler, and re-using anything more but the existing radiation & distribution plumbing is just begging for an early service-call. Any money you might have saved up front will only end up costing you double later.

FWIW, unless wood is free where you are, at current prices burning NG at 90%+ efficiency may be cheaper than heating with wood at 70% efficiency in a wood stove. If your existing boiler is rated 80-82% AFUE on the nameplate and is 3x oversized, it's really operating at 70-75% average efficiency due to cycling & standby losses. A 2x oversized condensing unit rated, say 92%AFUE might still operate at 92% or even HIGHER, depending on how low a temp you can run the water in the slab. With the thermal mass of the slab to work with it's cycling losses will be minimal (it can modulate down to typically 1/5-1/4 of full output when under partial load), and if the water temps can average 100F or less and still deliver the heat to the room (common, in slab radiant) the combustion efficiency will be over 95%, with very minimal standby & distribution loss. The fuel required would be cut by at least 25%, but could be as deep as 40%(!) compared to heating with your 106KBTU cast-iron beast. eg: If the old boiler burned 800 therms to heat all season it would only burn 500-600 therms with the new boiler. If you've cut the total heat load by 25% with the new insulation (also likely- but possibly more) you'd likely be under 500 therms for the heating season (maybe for the year.)

Oh geez, where to start?

First, the seller was the US Department of Housing and Urban Development. We got ZERO background or utility info. I got the year of install on the boiler by calling the number of the contractor on the unit. Thank heavens they were still in business and had computerized records that went back that far! Utility information should be public, though, right? If i call up our utility provider, will I be able to get back data on gas usage? We paid the gas bill for a month before the heat went down, and it was a pretty mild month.

On the water heater: When we purchased the house, the boiler worked and the water heater didn't. We replaced the water heater with a stiebel-eltron on-demand ELECTRIC water heater. The (pie-in-sky) reasoning was that we'd be able to eventually get grid-tied solar PV on our roof (we can eventually produce electricity, we ain't gonna produce gas). Almost a month to the day later, the boiler went kaput. I would have loved to upgrade everything at once, but we've got this mere year-old tankless heater plugging away . . . Yes, we're happy with the performance, by the way. It only annoys us when we're trying to run a bath but that's infrequent enough to not matter. The hot water sandwich is annoying, but when the galvanized pipe gets replaced (same time as boiler install), we'll pipe-wrap the hell out of everything. We live with a small footprint (no TV, few gadgets, no a/c, ya know; Luddite), and our summertime electric cost is a whopping $12 a month.

Piping into the floor looks to be 1.5"? Don't know if that's helpful information or not . . .

I had decided on an 80kbtu boiler precisely because it was the smallest size I was coming across online. I found one 60kbtu unit, but it's a buderus and very expensive. Are the "intelligence" and supposed quality of the German marque worth the extra price? I've come across some less-than-stellar reviews/rants relating to buderus units online. The place I've been talking to about purchasing carries Munchkin. I'm stoked to discover how much comfort we can squeeze out of our gas bill once the insulation is on the house! R1.2 to R11.2 is quite an upgrade (maybe not that good, windows and EPS vs XPS factored-in)! I'm also interested to see how a modulating boiler will change the character of heat. It seemed like the cycling was a bit harsh with the old boiler. It'd get up to 80, then drop to 65, then back up to 80 . . . You get the idea. I assume a modulating boiler will be able to keep things more even?

I believe we can run the water at pretty low temps here, as our first project on the house was to scrape up 1/4" of black mastic (Uncle Sam was kind enough to get rid of the carpet for lead paint remediation), acid stain & wax the floor. In the month the heat worked, it was SUPER comfy. That said, the PO had set the high-limit on the boiler quite high, which is probably partially responsible for the failure of the COPPER (no iron there) heat exchanger. I'm sure they were trying to get their heat through the mastic, pad and carpet. Why, I must ask, would a person insulate over their HEAT SOURCE!?

I've mentioned in another post how corners have been cut in recent years on this house. It continues to be a source of endless frustration. The shell of the place is rock solid, we just keep tripping over crappily-executed "repairs and improvements." What the hell though, the price was certainly right!

Funny you should ask, Dana: Wood IS free where I am! I'm an arborist by trade, so I bring home as much delicious hardwood as I feel inclined to haul, split & stack. It was lots of fun splitting my entire energy supply for the winter, but so-far so-good. Keeps my back strong, which keeps my wife happy ; ) "Goin' outside to pay the utility bill, honey." "Okay, I'll watch."

Woah, think we covered everything? Thanks for ALL the GREAT info! We're constantly learning in this place, and this site has been a HUGE help to the process!

charlie

ps- keep the ideas coming, by the way. I can never soak up enough data . . .

IRRC Munchkin makes a 50K unit as well, which looks remarkably like a Peerless Pinnacle T50 (a great li'l boiler) in Munchkin clothing:




Other popular candidates would include the Triangle-Tube Prestige Solo-60, and Burnham Alpine ALP80, all of which modulate down to the 15-20K range.  The lowest mod on the Munchkin T80 is 17k, compared to 16K for the T50, not a big difference, so if there's any potential future plan for an indirect hot water heater the T80 might be a better choice.  (With an 80K burner behind it you'd never run out of hot water- could probably take a 1.5-2 gpm shower all day, all night, all year if you wanted to.)  If yours is an older model Munchkin, check the documentation for the lowest modulation- that will determine how much cycling it'll have to do at part load (lower is better, but anything under 20K is OK.)  Whatever boiler you select, it's good to have local installation expertise & parts support (not that anything ever goes WRONG with that... ;-) )

With the appropriate thermostat designed for use with slabs to deal with the thermal lag issues (eg. Tekmar 54x series) and using the boiler's outdoor reset function to vary the output temp of the boiler you'll reap super-high efficiency with very steady room temps. 

Cast iron boilers weren't designed to run at the low temps used in slabs, and need more complex plumbing to both work well and to protect the boiler from death-by-condensation.  Copper tube boilers are somewhat more tolerant of low-temp operation, (was it a RayPak, by any chance?) but it's more likely that the heat exchanger failure was due to condensation from low temp return water than a too high high-limit. Controlling a slab with a standard thermostat used for baseboards, radiators, or hot air furnaces is a recipe for uncomfortable temperature over/undershoots. 

If you're trying to use the wood stove in conjunction with slab heat you could end up with some strangeness, however (in which case a slab-thermostat might be required.)

I have a neighbor in the arborist biz- he sometimes refers to himself as an "urban logger", considering that most of his work is inside the city limits of Worcester MA.  (His job got more "interesting" lately due to an Asian Longhorned Beetle infestation & quarantine zone, and transporting the wood outside of the zone is subject to fines, and even within the zone is frowned upon.)  He definitely has more free wood than he and 10 of his best friends would ever need...

Dana,

Thanks again for the good information! Our old boiler WAS a Raypak. Our home inspector described it as a "glorified pool heater" and said we should expect it to die. He just didn't say "in two months!" C'est la vie, though. I'll wager the original boiler was a cast-iron unit, as this house has a real Frank Lloyd Wright complex (with Bauhaus pretensions) and he LOVED those cast-iron monsters! I wish we knew the ownership history a little better . . .

The plumbing place we're planning on working with caters to people who are trying to update/restore/renovate Denver's aging housing stock. I had a sewer issue earlier this year and had to replace a lead & oakum joint. After a lot of back-and-forth with the Home Despot plumbing guy ("It's probably threaded. Yeah, I'm pretty sure"), the counter guy at this place took 30 seconds to find me the EXACT fitting I needed (rubber compression plug. I had no idea it existed!). Anyhow, they do boiler kits too, and if their general plumbing support is any indicator, I think we'll be in good hands! Doiturselfplumbing.com, by the way. Highly recommended. They're fighting the Scrape, which makes them worthy of support in my eyes.

On thermostats: How does one install a slab 'stat? Is it a wall thermostat with the "thermometer" on the floor? Drilled INTO the floor? Our existing unit is an ol' skool mercury bubble 2-wire 'stat, which must explain the obnoxious temp swings. Is a slab unit something that's difficult to find? And the added cost for the outdoor reset? Worth the price of admission?

We do plan to keep using wood. The more free, carbon-neutral energy I can utilize, the happier I'll be. I appreciate the conundrum your neighbor's in with the wood-transport ban. We're unique out here in that we exist in somewhat of an "island" ecosystem. We need to be afraid of pathogens transported across the plains from the midwest or across the divide, but we don't have any county-level quarantines to deal with. Once you understand the amount of damage that stands to be caused by moving firewood (thus fungal spores and bugs), it makes a lot of sense. We're contending with "thousand canker disease" around here. It only attacks Black Walnut -already pretty rare out here- and is INCREDIBLY hard to fight. We managed to dodge Dutch Elm's for the most part, but it's getting harder and harder to avoid imported pests and diseases! Back East it must be nigh impossible!

Worth the hassle to keep such creatures healthy, though. Consider 100 years of growth being destroyed by one truckload (one ROUND even) moving 5 miles! Pretty sad . . .

Sorry, /rant.

charlie

Ok, after some research, more questions:

Are the munchkin Contender models comparable to the T-range? There's an MC50 that appeals, as it's a wall-mount. Can I assume the internals are similar or the same? It's 92% versus 95% afue, but $1900 versus $2800 for the Triangle Tube Prestige 60, the other appealing wallmount on the sites I've been shopping. Dunno about local support for the TT, though . . .

As I recall, the Contender is the boiler the plumbing outfit was pushing. Should I be nervous about the pricepoint difference?

Charlie

The Munchin Contender 50 & 80 also modulate down to under 20K and are decent units with stainless steel heat exchangers.  Don't sweat the AFUE test numbers on 'em, since that's based on temps much higher than you'll be running them anyway. AFUE test protocol calls for 140F output, 120F return- temps that would actually destroy and violate the warrantee on a cast iron boiler, despite the ratings having been based on those temps. AFUE also calls for a specific duty cycle and max fire, IIRC.  Whether you get 92% vs 95% at 120F return temp at max fire matters not, if most of the time your return temps are under 100F and you're running at low-to-middle range modulation.  I expect ANY condensing boiler will hit or exceed 95%+ at 90F return at low fire which may be your seasonal average post-insulation, and making up the price difference with another 2-3% in fuel efficiency would take more than the lifetime of the boiler. 

The pricepoint difference is primarily a factor of the model change to the Peerless-style versions, and the slightly lower test numbers, old stock vs. new stock, etc.  As long as the seller & Munchkin will back it up with a warranty I wouldn't sweat it much.

Okay Dana,

First of all, thanks for all the help so far!

We got gas usage data for 2004 to 2007. Not day-to-day obviously, but month to month. What calculations do I need to do to crunch the numbers down to something useful? It looks like our usage tops out around 210 therms for January and bottoms (with the old gas water heater & no gas range-which we now have) near 20 for August. The table I was given also has mean temps recorded at xcel's well-head.

The label on our dead Rinnai indicates 135000 btu/hr for input and 109000 btu/hr for output. Does that mean it was rated at 80%? 1 therm converts to 100066 btus, correct? Can I assume my boiler accounts for 190 therms of usage in January? So convert the therms to btus in, apply 20% loss and I've got actual heat making it into the floor/house?

Am I on the right track? Sorry. It's been an eventful weekend and I'm a bit fried to work through it all on my own assumptions . . .

Thanks again for the guidance!
charlie

Every good heating design starts with a proper heat load analysis.

During the heating season you can ignore the hot water heater's contribution to the bill (it's "in the noise".)

If you can find data to correlate the average daily temperatures (not the high, not the low- the average) for a mid-winter billing period and calculate your fuel use for every heating degree day (use degrees below 65F outdoor temps.

Por ejemplo:

If the average temp for the day is 20F, that's 65-20= 45degree-days. If you had a bill for that one day, and you used 11 therms, that's 11/35= 0.31 therms/degree-day use.  (You'll have to add up the degree-days for the entire billing period, then divide by the therms used.)  A therm is 100,000BTUS, so the house is using 100,000 x 0.31= 31,000 BTUs/degree-day

Now, since RayPak's all run about 80% combustion efficiency, 20% of the that source-fuel energy went up the flue, and the heat load on the house is 0.80 x 31000=24800 BTU/d-d.  Dividing that by 24 hours/day yields  1033BTUs per degree-hour.

Now you need to look up the weather history for the typical lowest temp in the heating seasons (or in some ASHRAE or ACCA 25 year binned-hourly history for the 97th-99th percentile coldest hours at the city closest to you in altitude & temperature) an multiply that by 1033BTUs to come up with your minimum design-day boiler output. Let's say the 97th-percentile "design" temp is -5F.  That's 70F below your base-65F, so you'd need a boiler that can deliver at least 70x 1033= 72,310BTU/hr to be able to keep up when it's -5F out. (Which is an 80K mod-con if you have enough radiation to deliver the heat at low output temps, or a 90K ~80% RayPak)

If the boiler has an output more than 3x the design-day load your analysis delivered on the first pass, subtract 10-15% from it's efficiency numbers and re-calculate. (use 0.65 or 0.70 instead of 0.80.)

This analysis will be much closer to the whole house heat load than any estimator tool, since it's using the boiler in-situ to measure it, and requires no guesstimations or assumptions about how the house is built & insulated, etc.

Odds are your design day temp isn't much lower than that, and your fuel use per HDD is likely to be half the numbers in the example after you've insulated the place, but might be in that ball-park to begin with (or not- it's probably lower).  Even if your design-day temp is -15F, it doesn't affect the boiler sizing by much, since 10F one way or the other makes less than a 20% difference in your peak heat load. If you can't get the 25 year history binned hourly number, work with the coldest daily low you can find for the weather history you can find on wunderground.com's tracking for your zip code in the past year or two, and subtract 5-10F for some margin- it'll be good enough.

okay. Lots of number crunching done here. I used the weather data given in the table and then went back to correlate. The table's pretty close. January 2007 had a mean temp of 23* & the table lists 24*.

what I got: I'm using 534 btus/degree-hour, assuming the boiler operates at its rated 80% and that the previous owner set the 'stat all the way down at 65* (not bloody likely)
Dropping the efficiency decreases the btus used for heat, as more of the purchased energy went out the flue. Raising the assumed 'stat setting to 72 decreases btus/d-h as well

The all-time low for Denver was -29*, which would be 95 degree days
Even that day would have required a 50kbtu boiler, as the btu/hr requirement would've been 50788
A more realistic requirement (-10* - a REALLY cold night) is 40096 btu/hr

This means my 50k boiler will spend a ton of time modulated down to 20k, right? Is that a lifespan problem for a stainless exchanger? Why did they install such a BIG effing boiler?!

Feel free to point out anything fishy here. I plugged everything into a spreadsheet so it's easy enough to modify.

Thanks again!
Charlie

Yes,,even properly sized it spends most of it's time (and half it's fuel burning) at low-modulation, and that's a GOOD thing with a modulating-condensing boiler! It's highest efficiency will typically occur somewhere near 1/3 of full-fire, sometimes a bit lower, and often the lowest-fire. The stainless heat exchanger is designed specifically to operate under these conditions-it'll be happy.

As for the reason for the big F'n' boiler, in most cases it's a combination of not doing the math, not knowing the parameters, or the contractor wanting to be sure to never get the 4AM call from the customer saying the boiler is running flat-out and they're still freezing.

Using 65F as a base temperature heating degree-days is still right for a house kept at ~70F, because heat from appliances + humans + solar gain etc. typically meets or exceeds the requirements to keep it 70F or above when it's 65F outside, and the heating system never runs.  In VERY well insulated structures you may have to use a lower number, like 55F or 60F. (In super-insulated houses with R50+ walls, 25F.)

There was a time when 50k boilers were actually pretty rare (it's impossible to get standard oil-burners that run that low), and when fuel is cheap the cycling losses when installed in smaller houses are "don't care".  With modulating technology it's easier to get the efficiency out of it if the very lowest fire is still around the average mid-winter load or lower.

At -29F you'd be outpacing a 50K boiler (they're rated for input, not output BTUs). At 95% efficiency you'd only be getting 47,500BTU/hr out of it.  If the return water coming from the slab are ~125-130F it'll be running more like 88-90%, and you're in the 45K range for output.  If the return temps from are more like 90F you'll be doing great, but never 50K.

A point taken from another thread: If you're above 2000' you may need to readjust & derate the output. Above 5000' this has to be taken even more seriously.  Most manufacturers have "high altitude" versions or adapter kits for adjusting the fuel/air mixture properly for the lower pressure, but the definition of "high altitude" isn't standardized- it's in the fine print somewhere.

I've been reading this thread and have gone back and crunched some numbers as well--thought I'd share.


I live in Boulder CO and I've been keeping monthly gas/electricity consumption since moving here in late 2007. I also have a weather station and push data to Wunderground. To perform the calc I summed the degree days for 2009 and came up with 5500 degree days total--which is ballpark compared to other figures I've seen for this area. In Jan 2009 there were a total of 794 degree days; in Dec 2009 there were a total of 1146 degree days.


Using the formulas above I got 802BTU's/degree-hour and 502BTU's/degree-hour for Jan and Dec 2009 respectively. This seems weird because we used more therms in Jan 09 for fewer degree days (168 therms, 794 degree days) compared to Dec 09 (152 therms, 1146 degree days). Maybe this is from lowering the thermostat?


Assuming the coldest day of the year is -10*F (75degree-days) a max projected requirement of 60KBTU's/degree-hour using the inputs from Jan 09 (168 therms, 794 degree days). If I use Dec 09 numbers I get a max projected requirement of 38KBTU's/degree-hour using the inputs from Dec 09 (152 therms, 1146 degree days).


So how to size a boiler with such divergent numbers (60KBTU's/degree-hour vs. 38KBTU's/degree-hour)?

thx


PP

We like the LARRS Neotherm or LAARS Mascot, Both are modulating condensing units, with a turn down 5 - 1 so for the Mascott for instance the range is 25,000 BTU - 125,000 BTU.
Plus You now have the capacity for a wide range of conditions, Remember with gas appliance you will need to consider for de-rating due to your elevation. at 6,000 feet I think it is about 20%, but would want to double check.
Dan

your loss per degree hour can be changed radically by things like solar gain which can vary heavily month to month.

you definitely have to size for the higher loads. However, there is no telling by degree days that your heat loss of 60kBTU/dd is in fact your max demand. that's more like a max average. Imagine a situation where you had two cloudy weeks and two sunny weeks, the sunny weeks would understate your monthly average to the point that a boiler sized for that might be undersized for 2 full weeks. no good.

to size a boiler you need an actual design day heat load calculation.

A 2 week measurement period would be too short, but even were that the case, if it's taken during a mid-winter period the undersizing by that method would be small, and it would only be undersized for a very few hours for those two cloudy weeks, worst case. Typically the peak loads occur during early-AM pre-dawn hours when you're in bed, and even undersized for the peak load by a full 10% it WOULD catch up way before noon despite the darkest of clouds.

Unless the house was designed for high passive solar gain, using the boiler to measure heat load over a reasonably long period will be more accurate than any estimation tool, but that's not to say that sizing the boiler to exactly that number is the right thing to do. But even a 10-15% margin above that number will never leave you cold, the exception being homes with large amounts of southern glazing for huge passive solar gain, but also high heat-loss during darker periods.  Manual-J typically overshoots by more than 25% of real-world.

Tracking heating season monthly fuel use against degree-day data at my place is remarkably consistent, with less than 5% variation through the entire heating season. March has much higher insolation than January, with far more sunny hours, yet the number doesn't budge.  The number only moves in step functions concurrent with building envelope or heating system upgrades, not weather. YMMV.

In this case, it's a small house in a non-severe climate, to which substantial envelope upgrades are being added. I suspect that no matter how the heat load measured up using the previous boiler, the smallest mod-cons would still carry the load.

I make a living undershooting everyone else's heat load calculations by a good 20-30%... I'm not into oversizing.

but degree day back-figuring is not a proper boiler sizing technique. there are simply too many meteorological, mass, and construction variations to account for. the gas bill k-value on my leaky farmhouse (yes, I have lots of work to do) can swing quite significantly over 3 week periods in winter. he's seeing a 50% swing compared two months together. He should not be sizing a boiler with those numbers and I would put forth, no one else should either. you can do estimation

"YMMV" in this case is a pretty big caveat. "Peak Loads" occur at night, that's true, but ASHRAE Design temperatures are not peak loads either. they are design loads to be met on a more or less continuous basis.

Doh! I just realized you  were responding to ppmax, not treeguy303.

It would remarkable that one would see a 30-50% BTUs/HDD ratio between two monthly fuel use periods unless it IS a passive solar house.  Of several houses that I've tracked I've never seen numbers that varied that widely- even 10% would be a significant datapoint outlier. But taking the worst case number (after checking your data & arithmetic) and allowing a 10-15% margin is probably safe.

yes, but why go with "probably"? you're sizing a heat plant. shouldn't you, you know, actually size the heat plant?