Sizing Conundrum (Heat Loss)

Tony_23

Infiltration is the most elusive figure to nail down, IMO.

A GSHP or sealed combustion appliance will reduce the infiltration loss of a building. Remove the negative draft of a conventional vent boiler and the infiltration of outdoor air will slow dramatically. ManJ accounts for this neg draft in the infiltration factors.

Is the proposed GSHP a water to air or a water to water ? If it's water to water you'll need a hydro-coil for AC.

Personally, I would install a W-M Ultra :)

As far as a GSHP, 8 tons sounds right w/o doing the on-site calc myself.

I wouldn't worry about the what-if of the system needing more BTU's to "catch up" after a break down, worst case scenario of a week of design days after a lengthy (3 day) breakdown is sooo remote. Remember, design conditions are only 5% or less of a season (typical). Also, all the appliances and lights, etc. contribute to the heating of the house.

I also wouldn't be too concerned with only taking heat out of the Earth w/o putting any back in the Summer. Pretty tall order to do any harm that way, what with global warming and all :) (Sarcasm)

gasfolk

10 tons seems big

We are just homeowners with a strong need to get our new heating system right the first time. We like our engineer, our plumber, and our contractor, and we like The Wall and really appreciate all we have learned here so we could ask the right questions at the right time--before things get started. We hope you can help us again and that we can share your advice with our team.

The geothermal engineer says "Ten Tons." This engineer is very good and very experienced (decades). We remain puzzled by sizing issues.

Here are the heat loss estimates:

120 MBH The engineer, using good software (MJ7).

92 MBH Our Manual-J, using very similar good software (MJ8).

44 MBH Our HDD calculation from a January bill, unadjusted for solar gains.

72 MBH HDD calc adjusted for 6-hour passive (solar) heating (i.e. gas use over 18 hours).

60 MBH Boiler clocking, 8 to 9 A.M. on a 37F morning, adjusted to 95% design temp.

68 MBH Calculation from maximum observed supply temp and total area of all heat emitters.

During winter days, the house heats up to 68 (from solar gains) and the boiler shuts down, so we adjusted the HDD calculation for this. We measured our radiators, radiant floors, and baseboard and estimate an output of 68 MBH at 125F (without adjusting downward for radiator covers)--they would require 160F supply to reach 120 MBH, a temp we have only seen in the boiler loop. Our Tekmar 352 injection control is set at a heating curve of 0.8 and a Maximum Supply Temp of 140F, so at that maximum supply there could have been 91 MBH, but we have never seen supply temps above 125F (at a 0.8 heating curve it looks like we wouldn't max out at 140F until minus 10F outside). Last year, we kept the thermostat at 63F, dressed a little warmer, and were quite comfortable. This is a 1920’s, 4000-sq. ft., (3200 above ground) brick house with a converted gravity system and lots of cast iron. Storm windows were added in the seventies, and we just added R-28 open-celled foam in the roof, radiant floor in a bathroom. We can only get 120 MBH from our software if we assume “loose” construction and design for “mean extreme” temperatures (minus 10F).

We have been told the following, and they all seem like a reasonable points:

1. If the heat ever goes off, it will take a lot more BTUs to bring the house back to temperature, so we can’t go by usage.
2. Our house is brick, so we are benefiting from solar gains, but if we had a spell of freezing, cloudy weather, we would need more BTUs.
3. We should increase heat emitters to handle the design load.
4. We may have to run supplemental electric heat for 3 or 4 weeks per year.
5. We can install a smaller system, if we assume responsibility for (under)sizing.

If 10 tons is the answer, then we cannot afford that and changing our cast iron radiators. As homeowners, we also would not be comfortable asking for a smaller system.

Can anyone help us understand where our thinking is off track? Is geothermal sizing typically much larger than usual? Advice on The Wall prompted our work on the envelope first. We would like to replace our boiler this year, but would there be any value to further measurements of our current system over the coming winter?

Thanks for any comments or advice you may offer.

gf

Constantin

Whoa....

.... I am a mere homeowner myself, but I think you have uncovered a number of interesting conundrums.

For one, I don't think that most heat loss programs properly capture the thermal flywheel effect of homes. They simply assume the worst-case condition, which is a week of design-days w/o sunlight, etc. It's one way to cover the tuckus of the installer, right?

On the other hand, considering the huge cost of geothermal systems, here is what jumps first out of my head: I would install a Geothermal system that has 4, maybe 5, tons of capacity. That way, if you ever want to cool the place, you'll probably have enough capacity to cover the cooling load.

I would supplement the geothermal system with a condensing, modulating boiler like a T50 or T80 from HTP. It'll do a great job of heating the DHW and can run on propane in case you're far away from a natural gas source. That's likely to be a far less expensive option to run than installing 6 more tons of geothermal heat. Plus, the T80 could probably cover the house if ever there was an issue with the geothermal system.

With a geothermal system that covers your homes load 90%+ of the time, your operational costs ought to be very low. Meanwhile, the gas-fired appliance provides a fallback and coverage for the truely cold days. Your thoughts?

gasfolk

That's a very good plan...

Thanks for the reply Constantin,

We like your plan, and even after a day of pondering it, we see no downside and many benefits: 1) Redundancy, 2) Capacity for A/C, 3) End of discussion--the source of the heat-loss discrepancy becomes academic, 4) Substantial cost reduction. If the cost isn’t prohibitive, would you recommend sizing the earth-coupling loop for 8 tons for future expansion, better efficiency, and/or a little insurance?

We will probably go for a 5-ton system, because the 10-ton is not affordable. The sad consequence of Manual-J’s built in over-sizing seems to be a selective burden on systems (such as geothermal) where installation cost varies linearly with sizing, and without authoritative evidence to support a reduction in size, we totally understand our engineer’s commitment to the Manual-J numbers.

After asking many people (geothermal folk, engineers, ACCA, Building Science, and HVAC), who couldn’t/wouldn’t help us, we found our geothermal engineer, who suggested (correctly, we believe) solar heating as a significant, often unrecognized factor in the Heating-Degree-Day method.

Can you list the above methods of heat-loss calculation in the order you would rely on them? Most require only simple math (we’re not talking 18.03 here), so what should a homeowner strive for? Which is the gold standard?

Finally, can you or anyone comment on the magnitude of the sizing effects from 1) Manual-J oversizing, 2) Elimination of combustion makeup air, 3) Passive solar heating in the northeast?

Again, many thanks for your thoughts,

gf

Constantin

Thanks for the kind words...

I too considered the possibility of a ECR-DX system until I had a lot of trouble getting the local ECR rep to return my phone calls. Also, considering our location and the local prices of electricity vs. other fuels, going Geothermal didn't seem like it would provide that much of an advantage over gas or oil.

One thing I would discuss with your engineer is the thermal resistance of the soil around your home and what the long-term effects could be if you're only pulling heat out of it (i.e. not pumping the same amount back in during hte summertime w/AC). If the conductivity isn't all that high, you may need to oversize the system as you allude.

On the other hand, considering how well a small condensing gas boiler is likely to handle the entire load, you have considerable leeway... They modulate across such a wide range of firing conditions that you ought to be covered no matter how much or how little demand there is for the boiler.

If I were in your shoes, I would assume that an actual measurement during design-day conditions is about as accurate as you can get. Extrapolating from known but non-design-day-conditions and consumption works, but is fraught with more error. For example, usually it is quite simple to measure fuel consumption. On the other hand, getting the actual combustion efficiency nailed down can be more challenging in hydronic systems.

Simply assuming that the AFUE rating is accurate is optimistic in water-based systems. Otherwise, Mike T. in Swampeast, MO wouldn't have been able to reduce his fuel consumption with a home quite similar to yours by 43.6% when the AFUE difference was only 14%. He's using a Vitodens to heat his home, is very happy with it, and has a long report on the unit in the "On the Job" section.

As for Manual J, I have heard that the 7th edition oversizes by as much as 25%. However, I have not yet had any opportunity to compare firing times on my Vitola to the predicted heatloss. Several participants here and elsewhere have stated that the Manual-J assumptions are a bit simplistic in the light of more modern insulation materials like Corbond or Icyenene.

Eliminating combustion air heat losses by going direct vent has several benefits. For one, you pre-heat the incoming air. Secondly, there ought not to be any CO-poisoning potential. Thirdly, it reduces the exposure of the interior of your home to outside air... in my house it bugs me that I cannot hook my Vitola air-intake to the outside, so I have to treat and insulate my utility room as if it was part of the outdoors. (ARGH!)

I also noticed a lot of solar gain in our home during the wintertime. When it was 40°F outside, many rooms remained at 66°F+ even though the heating system was only set to 60°F. With all the thermal mass that your home offers, it ought to flywheel pretty nicely.

Mike T., Swampeast MO

That 43.6% Reduction

Still isn't as good as it can be.

Based on gross efficiency (or is that the gross inefficiency) of the old boiler, should be able to get a touch more than 50%.

Will have a second outdoor temp sensor installed on the South side of the house this season. During periods of higher solar gain, will switch over from the North side sensor. This will depress the supply temp closer to what is actually needed when the sun prevents the house from loosing as much heat.

Brent_2

geothermal

I think you are missing something here, unless I am misreading your posts. You need to look at the heating output capacity of a geothermal water to water unit. I'll assume you will have a closed loop geothermal system. With a 10 ton system I would assume you would have two 5 ton units. A 5 ton unit will put out somewhere around 41,000 btu. That means a 10 ton system will have a capacity somewhere around 82,000 btu. Depending on your actual design conditions the output will change slightly. The max supply water temp that most manufacturers suggest is I believe 125°.
You can check out the data sheets for whatever manufacturer you are thinking of using for more info.

Brent

Constantin

Interesting...

... I had no idea that the actual tonnage capacity is different from the air-side of the business. In other words, I thought a ton was a ton, i.e. 12kBTU/hr of heat removal capacity.

... with the information that has been put before us, I'd be inclined to point them to installing a Munchkin T80 and to have the near-boiler piping installed in such a way as to be compatible with the future addition of a GSHP. Once the owners want AC, they can then tie the GSHP into the heating system.

Constantin

Also...

... have you given any thought to installing the optional 2nd supply water temp sensor to make the system run on a ΔT basis? Looked like an interesting way to make the boiler internally and externally reset.

gasfolk

The geothermal would be a Standing Column Well (SCW) design. Our geothermal engineer sized the house at 120 MBH, and since then we have been debating 10-ton versus 8-ton installations. The proposed equipment, water-to-water GSHP, comes in 3, 5, and 10-ton capacities.

This is very interesting. Can you offer a reference for this information? Thanks,

gf

gasfolk

Thanks for your thoughts Tony. The systems under consideration would be water-to-water, standing column well (SCW) GSHPs. Can you estimate the magnitude of the effect of the reduced infiltration? We have been using Wrightsoft Right-J, so would this reduction already be reflected in our heat loss estimates above?

Thanks,

gf

gasfolk

We had the same problem with ECR, and perhaps they are still working out their marketing approach. Nevertheless, we found that the hydrogeology of our area is excellent for standing column well GSHP--40 feet of glacial till over bedrock--so we are looking at a system similar to the one we read was installed at Henry Wadsworth Longfellow's house on Brattle Street.

On boiler efficiency estimating, we have used the equation on page 29 (the 32nd page of the file) of Henry Manczyk's paper www.energy.rochester.edu/efficiency/optimal_boiler_size.pdf. We had a nice chat with him last year, but we never got around to asking for his sources. Any thoughts on the equation's validity and usefulness?

Mike T., Swampeast MO

Didn't know that was an option. Would LOVE to run it via delta-t; particularly in moderate weather.

Any reference?

You're fortunate to have their HQ so near...

gasfolk

Thank you Brent, you are right. The manufacturer's data sheets show that at an entering water temp (EWT) of 50, the output of a 10-ton unit will be up to 102 MBH; and a 5-ton unit, up to 51 MBH.

It looks like a 5-ton unit will not meet our load 95% of the time unless further shell improvements are feasible.

Bob Eh?_2

Constantin/Gasfolk

The ratings on heat pumps are the epitome of "Nominal". The relationship between source loop temperatures and flow rates on the input side, the load loop temperatures and flow on the output side, and the "Lift" (differential between source and load temperatures) has a huge impact on net output and efficiency. Add in the refrigerant's absolute efficiency curves at the low end and high end of the temperature range and the charts get to be quite complex.

Go here http://www.climatemaster.com/download/Commercial Literature/LC366.pdf for a sample.

Generally these units are more efficient in smaller sizes (besides being able to stage them) and with low temperature emitters (LWT requirements over 125* are pretty futile).

I get the distinct feeling that his application would require a supplemental heat source anyway due to that factor (the Emitter temps are too high). Personally I would go with 2-3 2ton/1 five ton and a really small modulating boiler (like a T-50 Munchkin) and set up indirect DHW with a preheat tank off the boiler loop and the Munchkin providing boost to the DHW tank.... I suspect the savings on the source loops would handily pay for the Munchkin. Just make sure that the net HP output is high enough to provide enough cooling when you get to that part of the event.

Bob

Constantin

The Vitotronic 200 manual...

... lists the usage of a 2nd, strap-on thermometer (same as the one that comes with a mixing valve) as a optional accessory. It self-recognizes it, you simply have to tell the Vitotrol the desired ΔT temperature difference.

I guess the question is whether the Vitotrol in the Vitodens is as expandable as the Vitotrol 200 and/or if you're using a 4-way or a 3-way mixing valve in conjunction with your Vitodens. Either way, I'd give Jim McCarthy a call and see what he says.

gasfolk

MikeT.

Please forgive the basic questions: When the sun sets, how much difference will you expect between your north and south side thermometers? If you can run by delta-T, will you still need outdoor setback at all? If one has TRVs (we don't), does outdoor setback save in the supply piping?

BTW, your "On The Job" review is very impressive. Do you recommend your data-logger? Are any cheap data-loggers worth the effort?

gasfolk

Thanks Bob..

...for the reference. Climatemaster must have recently revamped its website and added a lot of material we couldn't find before.

Because of the supply temperature limitations, we will likely add radiant under the entire first floor, which should also cast some heat into the basement (though we won't be counting on that). Down the road, we can add radiant over insulation on the basement floor, but that's for later.

It looks like we may have to pony up for a 10-ton geothermal unit. We understand there is a significant downside to using two 5-ton units (more expense and not much gain in reliability). We haven't yet seen the relative cost/complexity of combining a 5 and a 3-ton.

Thanks again,

gf

Mike T., Swampeast MO

Actually my outdoor temp datalogger is on the South side and I have another sensor (not datalogged) on the North. 15°F is about the peak difference I've seen. It starts rising in mid-morning, peaks around solar noon and drops off again in the afternoon. By sunset the sensors have nearly identical readings.

Will have to do some research on that second supply temp sensor. Nothing mentioned in the Vitodens manual and no coding adjustment that I'm aware of. If it works on the Vitodens, it appears undocumented. Should it work, suspect that it will "auto-detect". Don't really need the fixed delta-t setting when the boiler is fully modulating--my idea was to drop it in via a relay during periods of very low heat loss when the boiler is forced into "pulsing" operation. With a fixed and quite high delta-t should be able to reduce the number of pulses. That "pulse" mode (as I call it) is however, extremely efficient but I'd like to reduce the number (currently 10 per hour).

YES, reset is EXTREMELY useful with TRVs. With a decent reset curve, you can keep system flow fairly constant regardless of the weather. Without reset, flow would drop GREATLY in moderate weather as you'd still have full-temp supply.

Regarding the datalogger: It's actually a wireless weather station (model WS-2010 from LaCrosse Technology) with computer interface and a number of additional remote-reading sensors. Upper temp limit for the sensors is 168°F, so only suitable for lower temp systems. My only problem has been with the outdoor temp/humdity sensor. Like the rain and wind sensors it is solar powered. Have had two replacements and its range is EXTREMELY limited. The ONLY place I can get it to work on the house is about 15' away from the base unit on the South wall. The other solar sensors have significantly higher transmission range.

As far as dataloggers go, it was fairly inexpensive. Accuracy (both absolute and relative among the sensors) is very good. Battery life for the non-solar sensors (2 x AA) is about 10 months.

Tony_23

Wrightsoft J

I don't know if it accounts for it or not.

I have experienced 20% reduction in load changing to sealed combustion from conventional, consistently. I have to believe a GSHP would act similarly.

Constantin

Well, here is my thought process...

Mr. Manczyk's paper is a great start, though it seems to be focused on larger cast-iron (CI) systems. For example, he makes no mention of modulating, condensing boilers, which these days can usually ramp up their output from 20% to 100% with no loss of efficiency. Such a single modulating boiler is likely to have a better efficiency than the 5 staged, single-stage boilers shown in his report.

These low-mass, modulating, condensing boilers are available today in many different residential sizes (perhaps they weren't in 2001 when the report was written) and should efficiently cover your home's energy needs whenever the GSHP can no longer deliver the oompf it needs to. That will still leave the GSHP in charge of most of the heating load by your calculations (51kBTU out of 65-68kBTU). The modulating boiler would thus simply ensure that the radiators get the hot water they need w/o the need to resort to radiant floors and the like (though they are comfortable!). A Tekmar staging system could probably control the whole enchilada, though I'd ask hydronicsmike to be sure.

The drawing below is but a straw-man to consider. It shows a combined primary-secondary setup which is a classic way to ensure that both heat sources see very low water temperatures which allows them to operate at peak efficiency while feeding the Balance of Plant (BOP).

I would intentionally undersize the GSHP to 5 tons or less due to its enormous installation cost and rely on the boiler to cover the differences as they occur. In fact, I would take a look at what your heat gains are in 1% or 2.5% conditions and size the GSHP from there. That way, it can cost-efficiently cover your potential AC needs in the future.

gasfolk

Good points...

Manczyk’s paper won’t apply or be useful to folks who have already made the jump to modulating equipment, but for those with fixed output systems, his equation on page 29 allows a better estimate of actual efficiency under partial-load conditions than anything we have seen before. We would guess that most homeowners (like us) are also saddled with markedly over-sized, cast-iron equipment. The challenge of determining actual heat loss is made a little easier by having a better approximation of partial-load efficiency.

For instance, the system we plan to replace is a 300 [correction 350] MBH-input, 280 MBH-output, commercial boiler with a "steady-state," thermal efficiency of 75.6% (commercial boiler, so no AFUE). At design temperature it will run at 70 / 280 = 25% load, much lower in shoulder months. From Manczyk’s equation, the observed efficiency at 25% load drops to 53%, and for most of the shoulder season (i.e. loads of 10% or less), efficiency drops below 30%. We find these numbers to be useful and part of what drives our decision to upgrade. It also appears to quantify the distinct advantage (through better load matching) of modulating equipment.

A frustrating aspect of this sizing process is reading repeatedly of many identified influences on heat loss, but finding not even ballpark numbers on the magnitude of their effects: 1) Manual-J over-sizing; 2) Boiler efficiency under partial load; 3) Solar heating and the flywheel effect; 4) Reduced infiltration losses by eliminating combustion (geothermal) or by drawing outdoor supply air (direct vent) (see Tony’s posts below); 5) Increased effectiveness of foam over fiberglass insulation. Heat loss defines boiler sizing, installation costs, emitter temps, etc., etc. Heat loss is the Holy Grail of comfort. We hope the engineer in all of us is looking for ways to quantify these estimates, and that’s why we referenced Manczyk’s equation. What can we all do to improve the others?

We are awaiting installation costs for the 5 versus 10-ton geothermal, so we shall see how big that difference is (we expect it to be big).

BTW, thanks for the schematic. One question: should we reverse the pump in the BOP loop?

Many thanks for your good ideas and comments,

gf

Brent_2

Why do you think there is a significant downside to using two 5 ton units? You are gaining staging and if 1 unit fails you still have the other. The installation cost would be a little higher but it shouldn't be that bad.
I was also interested in why you are using a standing water column for your geothermal. Is there a reason you are not using a vertical closed loop system? Will you be using the same well for your domestic water? Have you figured out what size pump you will need to pump the water out of the bottom of the well?

brent

Constantin

I'm a bit confused...

... but then again, my wife would claim that is my normal state of mind. Up in the second paragraph you indicate that the present boiler is a 300kBTU/hr input, and 280kBTU/hr output boiler with a thermal efficiency of 75%... as the thermal efficiency indicated by the 300/280 split is on the order of 95%, right?

I also agree that the standard heat gain and heat loss calculators used by residential contractors are not well matched when it comes to estimating heat gains and losses from well-built, high-mass, etc. homes. I presume that Geoff McDonell is using something more advanced in his commerical/industrial applications than Manual-J to calculate and shift loads throughout the plants he works on.

There is hope on the horizon as the DoE has released the algorithms, etc. of their most advanced simulation software into the public domain. The question is who will write a GUI program easy enough for the marketplace to accept in lieu of the many Manual-J solutions. Also, I suspect that only enthusiast contractors would make the switch at first since Manual-J has served them so well, and for so many years.

Anyway, I look forward to hearing more about your project as it advances. The sketch up above has been corrected. Cheers!

Bob Eh?_2

To say nothing of having half load on the source pumping and compressor reducing the electrical load and far less short cycling (especially if you go to cooling). There is a significant difference between pumping 15 and 30 gpm!

Bob

gasfolk

Oops...

Yes, that should have been an input of 350 MBH on our current boiler. And I thought I was good with numbers.

Best.

gasfolk

Brent and Bob EH?...

Perhaps we misunderstood, but our geothermal engineer seemed to discourage using two 5-tons units, suggesting unnecessary complexity and no benefit.

We are very interested in your suggestions, because to date, most of our efforts have focused on improving the heat loss, which drives the other decisions. It is a big step from the engineer’s 120 MBH to our range of 60 to 70 MBH. And if (as Tony suggests below) we may get a 20% benefit from reduced infiltration of combustion air, then a 5-ton unit may handle the full load. Perhaps we could install a 5-ton unit (with the existing boiler as backup, for now, and leaving space for a second unit later), then re-measure actual heat loss, then either add a second 5-ton, a 3-ton, a modulating boiler, or nothing. Your thoughts?

We are just homeowners, so here is what we understand about standing column well GSHP: most of the time it functions like a closed loop. However, if a period of consecutive, near-design-day temperatures causes the loop temperature to substantially fall (during heating) or rise (during cooling) then a 10 to 15% draw and dump of water from the well will recharge the temp and the system’s efficiency will return toward baseline. We understand this recharging function addresses concerns for cooling (or warming) of the earth surrounding the loop. We found Deng’s thesis (http://www.hvac.okstate.edu/pdfs/Deng_Thesis.pdf) interesting reading.

We have city water, but we thought to use the well for our sillcocks and sprinkler system (if we ever can afford one). We were leaving pump and other specs to our geothermal engineer. Are there issues here that a mere homeowner would be wise to review?

We really appreciate your comments.

Thanks,

gf

Constantin

Just be aware...

... that if your home is in the People's Republic of Cambridge, that the inspection department may throw up a roadblock as far as sillcocks with water from a well is concerned.

I capture our rainwater in a cistern for the same purpose but I have no sillcocks with cistern-water. The reason being that the inspection department insisted our cistern-fed sillcocks to have large warning plaques attached to them (non-potable water, do not drink!) and that the valve-handle be removed during periods of non-use, lest a 3-year old run up to the sillcock and drink w/o knowing better.

I'd call the potential gastrointestinal fireworks a learning experience, but I guess the water department takes its job very seriously. Thus, the cistern will simply feed the soaker-loops embedded in the ground around the house via a small effluent pump hooked up to a sprinkler control, which is a fine solution also. This way everybody is happy as the garden still gets watered as needed, the water inspector can sleep well at night, and I don't need to apply funny plaques to the home.

As for your suggestion to keep the boiler and interface a GSHP into the loop to see if and when the boiler fires, I think it's an excellent way to minimize your risks. Just remember that unless you boiler is a cold-start model that the stand-by losses may be substantial. And even if the model is a cold start, you'd have to subtract the "warm-up" gas consumption from your calculations as modern, low mass boilers like the condensing models I have been writing about have almost no start-up losses.

So, if you want to keep the old boiler for the next season, I'd invest in the money to put a data-logger on the boiler to see when it fires once the GSHP is in place. Correlate that to the outdoor temps and I think you'll have a good idea where (and if) the need arises to have supplemental heat, how much heat is needed, etc. On the other hand, if your calculations are correct, I'd simply consider putting in anything from a HTP T80 Munchkin to a Viessmann Vitodens WB2 6-24 and focusing on other aspects of the project.

Bob Eh?_2

All of the above + .....

WRT to your engineer....... Really? Hmmmmmmm.....

Before I would go any further I would want to know....

Do you already have the well? Do you have any solid evidence that indicates that the natural flow through the ground (Water borne or just heat flow)will be enough to keep heat migrating to/from the well? Do you have any reason to believe that the water itself won't be laden with iron/sulpher/etc. that may be really hard on an open system?

Please correct me if I am wrong.....

You have an existing hydronic system

It is working (but may be expensive)

You are doing upgrades to the building envelope that will reduce the heat losses(heating)/gains(cooling)

Heating:

Step one: Do a space by space (room x Room) heat loss analysis of the for the entire building. Preferably after the envelope upgrades with a blower test.

Step Two: Figure out how much emitter you have in each of the above spaces. You will need charts to get the output at each water temperature for each emitter

Step Three: On a space by space basis figure out how hot the water needs to be (this will be an average temperature).

You are really taking a big risk if you do it on a whole building basis as it doesn't account for under radiated space.

If any of those turns out to be over 110-115* (and those aren't very conservative temperatures... these won't do it on a day below the typical "design day" for instance) you are going to have to add radiation or use a supplemental boiler to get you there.

The short summary is you need X?BTU to heat the space (worst case) and that has to occur at temperature Y?

From there you need to calculate the DeltaT for each area and figure out the peak temps required and recheck the load that can be met by heat pump technology.

A similar process will be required fo cooling load analysis (Presumably HydroAir so you can also use this for ventilation requirements).

If this doesn't ring a bell in the current approach to the problem might I suggest contacting someone like ClimateMaster to get a second party in to review this project?

Bob

gasfolk

Don’t worry...

We are not doing anything until the numbers make sense to us.

Water quality of artesian wells in the area is reported to be very good, but you never know till you test your well. We understand there is a 2 to 5 % chance the well will not produce at all.

We have already done almost all envelope upgrades, including Icynene in the roof. We are likely to add radiant on the first floor, but predominantly to take the chill off some cold ceramic. We understand the 125* supply limit. If there is a reliable way to determine heat loss of a structure, we can certainly plug in the few remaining planned upgrades.

120 MBH heat loss makes no sense to us based on our experience with the house (summarized above). We have done our own, complete, Wrightsoft Manual-J and a room-by-room assessment of heat emitter sizing. Heating-Degree-Day, boiler-clocking, and maximum-supply-temp methods are necessarily “whole house,” though the latter can be converted to room-by-room with the heat emitter sizing.

Can you recommend a reliable method of determining the heat loss of the rooms? This is exactly the reason we started this thread.

Thanks for giving us good ideas to ponder.

gf

Constantin

Some additional random thoughts...

...could you be so kind and tell us what temperatures your observations were extrapolated out to? In other words, what are your alleged design-day conditions?

The reason I ask is that HVAC-Calc uses ASHRAE 2.5% conditions, as best as I can tell. Thus, winter design-day conditions are listed as 9°F in my area. Yet, last winter, we had a week at -15°F. The BTU/(ft² x hr) difference in our case is about 25% between the two conditions. Your real-life gas consumption measurements should be extrapolated to real-life temperatures, BTU requirements... or you might end up short.

Since you have upgraded your thermal envelope, you might be surprised to see just how low your supply temperatures may be to keep the place comfortable... Our house is allegedly going to get by with 110°F hot water on a design day... and depending on the sizes of your radiators, you may get close to that also.

As for room heat losses, the only reliable way I can think of to do it would be to measure the amount of water going into the radiators, the incoming vs. outgoing temperature, and using a TRV to control the room temperature. Sure sounds expensive!

On the other hand, a simple way to see if any rooms are under-radiated is to open all the radiators full blast, set the home on as one zone, and then ride the thermostat up until a reference room is comfortable. Then walk through the house and note any temperature differences. Adding TRVs to your standing radiators is another way to ensure even room temperatures while accomodating environmental variations such as wind on one side, slight under-radiation in one area, etc.

Bob Eh?_2

If you don't ask......

I hear you regarding heat loss calculations. Wrightsoft is certianly decent, HVAC-Calc is too... none provide a realy good scope on what assumptions they make to derive the load estimates and what "Safety" factor they throw in. Mind you, outside is controlled by God and he isn't talking either.

I finally decided that I would use the following procedure...

Don't account for Solar gain.... Can you decide when the sun comes up?

If you are pretty much there in terms of envelope upgrades do a blower test..... That will establish the base infiltration rate you must account for.Unfortunately I gather you didn't do one before the upgrades so you won't be able to figure out how much you gained.....

Do make sure that you have enough ventilation to be healthy and safely operate fuel burning appliances that aren't direct supply vent.

Do the calculations at the ASHREA design parameters for your area
Do them again at the historical limits but only with the infiltration baseline

If that spread is < 20% you are likely in the ball park with the ASHRAE load number...

Design the heat source to provide this at the peak temps your emitter system requires......

Make sure your heat pump capacity has enough cooling capacity to do that protion of the event and trade off the cost effectiveness and redundancy for the balance of the load..... Given that you appear to potentially have a Primary/Secondary setup it isn't rocket science to have multiple sources..........

Find a contractor who is good with tools to make it happen!

Do bear in mind that an under run will probably result in a trip to divorce court which would have bought a lot of Fuel ;-) Someday you might want to sell it to someone who doesn't like running around bundled up like the michelin man.....

Bob

gasfolk

Tony...

Can you give any details on the changes you have made to get that 20% reduction in load? Was that comparing fuel use before and after? Were your changes to the venting only, or are there other possible contributing factors? That's a great figure, which helps us a lot. Thanks,

gf

gasfolk

The design temps were indoor 70 F, outdoor 9 F for the Manual-J estimates (the engineer’s and ours) and 2 F for the others (except the “maximum supply temp” sizing, which was observed during a recovery from setback one chilly January early A.M.). Weather bin data indicates temps and times as follows:

7 F 46 hours

2 F 15 hours

-3 F 3 hours

-8 F 2 hours

Was that week at –15 F sustained at that temperature? Would you design a system for –8 F if historically that temp occurs for 2 hours out of the year? Would the house’s ability to flywheel thermally influence you? How about –3 F for 3 hours? 2 F for 15? Come to think of it, the house has electric baseboard in the basement (3500 watts) and attic (2880 watts), which should close a small, temporary gap.

We are sorry we haven't heard from Jerry Scharf on this topic.

Best!

gasfolk

[Reposted below]

Tony_23

That 20 %

is typical reduction in fuel useage at the same avg daily temp with the home tstat set at the same setting, with similarly rated equipment. ie, a 90% FA furnace burning room air vs. a 90% FA sealed combustion.

These are usually what I would label "average" homes when calcing a heat loss.

I would think a GSHP would act similarly as it doesn't burn anything

Another point is modulation. I have seen great increases in comfort with some noticeable reductions in fuel useage with modulating boilers. In particular, I replaced a Trianco Heatmaker with it's successor, the Teledyne Endurance. These are virtually the same with a point of difference in AFUE. The customer called just to say he was now able to set his tstat 2deg lower and his useage was down 15-20%. You don't get that percentage with just 2deg.

Remember, these are my personal experiences in my customers' homes. I haven't written any books, etc., nor do I desire to. But, I wouldn't BS you either.

gasfolk

How is this for a plan of action?

1) Run the current boiler another year and set the Tekmar supply limit to what? 110 F? 120?

2) Get the data-logger and sensors Mike T. and Constantin recommend and collect some real data.

So, how many sensors, and where would you put them?

Any other suggestions?

Thanks.

gf

gasfolk

Addendum

Any thoughts on what is a reasonable estimation for solar heating in January? If we can firm up that number, we could better adjust our *adjusted*-HDD result. We belive this result provides a maximum heat loss, because (as described below) it assumes that no fuel was burned during daytime.

"72 MBH HDD calc adjusted for 6-hour passive (solar) heating (i.e. gas use over 18 hours)."

This figure assumes that, for six hours each day in January 2005, every Btu of heat loss was matched by solar gains. We also adjusted for improved boiler efficiency from the assumption that gas was being burned in fewer hours (higher partial load). That is why our HDD result went from 44 MBH to an adjusted 72 MBH.

Constantin

Depends how quickly you want your payback

If you like, I can dig out the company that offered the ECR installation. Perhaps they can add a proposal with a DX for your review. As the supply piping on a DX is a lot smaller than on a standing well, the approach may be more cost effective.

If you decide that a GSHP is out of bounds financially, I'd simply install a low-mass, condensing and modulating boiler in the 80-100kBTU/hr range and be done with it. You can still add a GSHP later on if the energy prices and/or desire to have AC warrant it.

Come to think of it, here is something you could try that wouldn't be 100% nutty: We know that your energy loss cannot be more than 120kBTU/hr. We also suspect that it's below 80kBTU/hr. I earlier suggested a GSHP + a new boiler in a combined pri-sec piping arrangement.

Well, nothing prevents you from keeping the old boiler as a second stage to a T80 Munchkin. Pipe both boilers in a combined pri-sec. Let the old one fire off whenever the Munchkin cannot keep up with demand... If and when you decide to get a GSHP, simply unhook the old boiler and add the GSHP (and switch the staging priority).

The only key to determine whether this will work energy-effectively or not is whether the old boiler can cold start or not.

If you still want to datalog, I'd look into an integrated datalogging solution like Mike has used so that you can correlate the periods of demand from the boiler to outdoor events like temperature and wind. If you have OR, I'd also measure the supply and return temperature of every circuit (i.e. 1 sensor on the supply, 1 sensor on every return). That'll help you determine what ΔT you're running.