Help with sizing a boiler
Last Post 06 Nov 2014 05:07 PM by Dana1. 49 Replies.
Printer Friendly
Sort:
PrevPrev NextNext
You are not authorized to post a reply.
Page 3 of 3 << < 123
Author Messages
Dana1User is Offline
Senior Member
Senior Member
Send Private Message
Posts:6991

--
04 Nov 2014 12:11 PM
Rob: $3000 of oil at last year's prices is only about 750 gallons (maybe even 700), not 1000 gallons. I used the 1000 gallons as part of a straw-man exercise.

Note, it's highly unlikely that the ancient oversized beastie boiler is delivering net efficiency in excess of 75%, and I ran the numbers assuming 80%.

The 12 BTU/ft number is roughly where MY HOUSE will be after I'm done with the planned upgrades, not Tom's building.

He's probably IS somewhere between 20-30 BTU/ft now. At 30 BTU/ft on a 2500' building (even with high ceilings upstairs) which would put him at no more than 75,000 BTU/hr, but no way is it over 40 BTU/ft- brick buildings rarely leak anywhere near THAT bad (and if it does leak that much, it's pretty easy to find & fix.) A It's just not a very big building. Take a good look at the street view pic, and estimate the true wall height.

But if the 175 is already uncrated and halfway installed we'll just have to wait and see just how low he can run the temp without short-cycling. The description of the convectors wasn't ultra-clear- it would be great if the fin spacing is tight enough to be able to get more out of them.

NRT.RobUser is Offline
Veteran Member
Veteran Member
Send Private Message
Posts:1741

--
05 Nov 2014 09:16 AM
Dana, tom gave the price they paid for oil, he was close to 1000 gallons at his specified price. For loads you guys are saying are appropriate here, his building would have to calculate out to 12 BTUs/sq ft at design. No way, as I know you know. But you used 2,500 sq feet again. He's 5,000 sq ft, not 2500... 2500 PER FLOOR.
Rockport Mechanical<br>RockportMechanical.com
Dana1User is Offline
Senior Member
Senior Member
Send Private Message
Posts:6991

--
05 Nov 2014 11:58 AM
My memory isn't photographic or long term had to look up what his clarifications on price were:

"We paid around $ 3.20 per/gal. last year so we went through around 937 gallons"

Fair enough!

Call it 5000 square feet. The output of the Solo-110 is over 99K, that's 19.8 BTU/ft (not 12 btu/ft) which could concievably be the heat load of this place.

But let's worst-case it a bit, one mo' time for the record:

The heat loss characteristics of the floor that is half or more below grade (even without insulation), is nowhere near as high as the space above it. The design condition in his location it is something less than 60F delta-T, not 80F. The ACCA puts the 99% temperature bin for Boston at +12F (www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/Outdoor_Design_Conditions_508.pdf ). Winchester would be a few degrees cooler than Boston, but above +5F, call it +8F, with an indoor design temperature tolerance of 68F.

If his reported 60F setback temp is fine (given that the place isn't use in the pre-dawn AM when the outdoor design temp occurs) that drops it to an even smaller 52F delta-T, which is substantially more margin.

Eyeballing the height of the steps in the street view pic, the lower section has maybe 800 square feet of U1 (worst case, with all the fans going inside and a high wind outside breaking up the air films) for a design load of 48K at a 60F delta-T, but as you say, with air films it's really roughly half that. But for argument's sake let's split the difference and call it 36K, in case my eyeball isn't well calibrated for that picture.

Add that to the ~48K you scratched out for the wall losses on the first floor. Looks like maybe 150 square feet of U0.5 windows & doors for another 4.5K, call it even 5K for the attic losses (probably lower), and you're at 93.5K.

The infiltration on brick buildings is usually pretty low, but there are also the internal sources to add up. It might be pushing the margins of the -110 if it leaks like a sieve, but the cheaper solution is going to be to fix the air leaks (&/or insulate the basement) not buy a boiler that's likely to short-cycle then patch up the system with buffer tanks or add radiation to limit the short cycling.

It wouldn't take a huge investment to get it down within spittin' range of a Solo-60, but if it's over the output of the -110 the cost of shaving the load down to 99K is miniscule compared to the cost of reworking the radiation to keep the -175 from short-cycling.
NRT.RobUser is Offline
Veteran Member
Veteran Member
Send Private Message
Posts:1741

--
05 Nov 2014 12:27 PM
I believe you've missed the buried wall loads, unless you are saying the lower level is 4 feet high (200 linear feet of wall, you've only got 800 square feet of load for that), and infiltration may not be half the load but it's not anywhere near zero, you're definitely in 5 digits of load there as well in this case at least and probably closer to 20k than 10kBTU/hr... Uninsulated slab floor in new england? that's vehemently non-zero too...

Between the three components we haven't put numbers on you've probably got another 30-50% of that load to add, way over a 110 and nowhere near a ballpark that the solo 60 should even be mentioned in... that would be, basically, building another building inside this one to insulate it to those levels.

I'm sorry, you're just not close on this one. They could get to a 110 but it would take some doing for sure. Minimum would be insulating the basement walls and floor. And at 1000 gallons/year... dunno. A heat pump or two might be more cost effective ways to get the energy costs down (though I prefer insulation when possible of course). Maybe not if labor is free...
Rockport Mechanical<br>RockportMechanical.com
Dana1User is Offline
Senior Member
Senior Member
Send Private Message
Posts:6991

--
05 Nov 2014 02:05 PM
If you allow a 52F delta-T at 6AM on the coldest day in January and some minor envelope upgrades you have quite a bit of wiggle room on the 110, even with the below grade loads. (Mind you I was positing 12K more of foundation wall load than your R2 estimate, which would probably more than cover the below grade load.)

The deep subsoil temps in Winchester is about 52F. Hard to say for sure just what the slab losses are without a soil survey, but they're "in the noise" of the wall-loss error bars.

There is almost no way that even a miniscule amount of foundation insulation wouldn't bring it in. But it's fair that to comfortably hit Solo-60 territory at a 70F interior design temp would almost surely need to insulate the brick-clad portion of the walls as well. There's (almost) no excuse for not insulating the foundation walls, or a least a large fraction of that wall area.

Clearly this needed a Manual-J all along (as suggested in the very first response.) The fact that we still don't know with clarity what the wall stackup is, it's hard to know how big the error bars on it, or what (if any) retrofits would be in order. The actual stack-up of the wall is a LOT better than his original "...concrete with no insulation on the outside or on the inside..." description, even though we don't know what that is. My mind's eye on the original description was an old painted poured concrete warehouse type building, not a brick veneer CMU or framed studwall.

At least we steered him off of something even more than 2x oversized for the actual as-is-unimproved building load, even if we weren't successful in convincing him that doing at least some load reduction then going even smaller boiler was probably a better course of action. Given his radiation constraints a 175-200K cast iron boiler with heat purge control might do as well on fuel as the Solo-175 since will likely be short cycling at condensing temps. (I'd love to be wrong about that.)
jonrUser is Offline
Senior Member
Senior Member
Send Private Message
Posts:5341

--
06 Nov 2014 11:07 AM
I'd like to see some discussion of the following logic (for non-high mass/slab radiator systems):

Even perfectly sized boilers short-cycle under very mild conditions
Fewer cycles leads to longer life (quantitative data?)
Fewer cycles lead to greater efficiency (quantitative data for mod/cons?)
Boilers are noticeably more efficient at higher firing rates
Radiator sizing and zoning are frequently non-optimal, causing non-condensing operation
Oversizing makes setback more feasible
Oversizing prevents cold customers when temps fall below design temp.
Buffer tanks aren't that expensive.

Therefore, most boilers should be installed with a buffer tank.
Dana1User is Offline
Senior Member
Senior Member
Send Private Message
Posts:6991

--
06 Nov 2014 11:41 AM
The RayPak firing rate  vs. efficiency graph is valid for water-tube finned heat exchangers, but does not correctly characterize mod-con heat exchangers.   The sweet-spot for mod-cons is typically between 20-35% of the max firing rate, at any return water temperature.

Mod-con characteristic firing rate curves, efficiency vs. return water temp:

http://www.mnshi.umn.edu/images/boilergraph-1.jpg

Finned water tube characteristic efficiency curve thermal efficiency vs. firing rate:


http://globalimageserver.com/FetchDocument.aspx?ID=42bcc2f1-d974-4b16-bff8-c47336fca90c

If there is sufficient radiation that a boiler won't short-cycle with 100F return water, there is nearly nothing to be gained by buffering the system to be able to operate with long burns at an even lower temp.  Since it will be operating at it's lower, higher efficiency range when the heat loads are low and already in the high-90s, dropping the temp even another 20F won't buy more than another  1-2%, and the value of that efficiency may be less than the increased amount of pumping power due to the longer cycles.

Buffer tanks are of use for extending the burn times when the zone radiation can't deliver the heat at the min-fire rate, which is perhaps typical of micro-zoned systems.

The radiation system here looks like it could do just fine un-buffered with the -110, but not with the -175. The cost of retrofitting sufficient buffer to that system is likely to be higher than the cost of just bringing the heat load within range of the -110, if it isn't already there.  This isn't likely to be one of those systems where a cheap 40 gallon hot water heater (unpowered) is going to solve the problem.

The HTP Versa or Pioneer series have favorable economics for many micro-zoned systems, since they are both modulating and inherently self-buffered.  It might be cheaper than a Solo 175 plus buffer tank too.
sailawayrbUser is Offline
Veteran Member
Veteran Member
Send Private Message
Posts:2270
Avatar

--
06 Nov 2014 04:03 PM
Fully agree Dana and well stated too!
Borst Engineering & Construction LLC - Competence, Integrity and Professionalism are integral to all that we do!
jonrUser is Offline
Senior Member
Senior Member
Send Private Message
Posts:5341

--
06 Nov 2014 04:03 PM
The sweet-spot for mod-cons is typically between 20-35% of the max firing rate, at any return water temperature.


Sounds like another argument for some amount of boiler oversizing + a buffer tank. The former to help avoid operation at high % firing rates (inefficient) and the latter to avoid operation at very low rates (also inefficient, although the boiler might not go much lower anyway).
Dana1User is Offline
Senior Member
Senior Member
Send Private Message
Posts:6991

--
06 Nov 2014 05:07 PM
You really have to do the math on it to know when you'll gain actual seasonal efficiency with a buffer tank. The armchair needs a notepad for fully analyzing it.

Turn down ratios rarely exceed 5:1, with 4:1 or and 3:1 being fairly common. The reason those manufacturers don't turn them down more is due to the rapid fall off in efficiency that occurs when the flow on the fire side of the HX becomes laminar rather than turbulent. In almost all mod-cons the lowest fire that it actually operates at will be in the sweet zone.

Between cycling losses and buffer tank standby loss it doesn't make sense to oversize by more than the minimum necessary for covering the 99% heat load. Undersizing by half and running it full blast most of the time would have an efficiency hit, but typical boiler sizing steps are large enough that 30-50% oversizing is more likely than being dead-nuts on the 99% condition heat load for sizing. At 50% oversizing these things still hit their AFUE test numbers if they're not short-cycling, and they will modulate most of the time in the lower 1/3 of the firing range.

By simply marrying every mod-con to a buffer tank you basically re-invent the high-mass boiler, which is fine. That's basically what you get as a pre-packaged unit with the HTP Versa, Flame, et al- a high mass boiler with a modulating burner, but with most of the control & pump aspects optimized, rather than forcing the designer to do that math for optimizing a mod-con + buffer solution.  It's not a bad way to go, but it's not the only way to go.
You are not authorized to post a reply.
Page 3 of 3 << < 123


Active Forums 4.1
Membership Membership: Latest New User Latest: hudson2000 New Today New Today: 0 New Yesterday New Yesterday: 2 User Count Overall: 34707
People Online People Online: Visitors Visitors: 101 Members Members: 0 Total Total: 101
Copyright 2011 by BuildCentral, Inc.   Terms Of Use  Privacy Statement