Calculating and Designing a New HVAC system

[Dana]

I'm sorry, but personally I think the calculations are crap. First, your talking about a garage, and I assume you are going to be heating it low until you go out there and want to turn it up. There's no way in CT winters, you can heat a garage from 45 degrees up to 65 degrees with just 43,000 btu. I'm in Michigan with a single car garage shop that is well insulated and about 12x30 and I have a 80,000 btu gas furnace. I keep it at 45 and turn it up to 65 when I go out there and that will heat it up within 10 minutes. Yours is 24x50. That's 4x larger, and your wanting to use half of the heating btu? It will never work. I'd be going with 120,000 btu and that would be a gas furnace at a minimum. As for your bonus room above the garage. Whatever your cooling calculations tell you, double it. With all the heat from the roof beating down directly in that room, you will need much more cooling and airflow in that room to compensate. I would spray foam the ceiling in that room if I had to do it again. This is from experience in our bonus room above the garage.

The heat load is only the heat load, not a equipment sizing spec for what it takes to to rocket the temp up quickly by 20F. It's important to know what the steady state loss numbers are to appropriately size the system.

As for calculations. When we purchased our 2 story, 1800 sq ft house, there was a 3 ton central air system. No matter what the house wouldn't stay cool. We replace the furnace and central air and the calculations said the 3 ton would work. I told the hvac contractor no way, but he insisted. After it wouldn't cool the upstairs, he actually told me it would help if we changed our shingles on the house to a lighter color. Uh huh, yeah, right. They came back and put in a 5 ton unit. That helped, but what really helped was installing thermostats in each bedroom and a zoning board and dampers. I also added hydroponics blowers to the upstairs runs. The hvac guy said there's no way that would work. Well, I cool the upstairs in the middle of summer to a meat locker if I want without affecting the main level. So, sorry, but the calculations are crap. Get someone with real world experience in there to tell you what you need.

Sounds like your house had the air handlers & ducts up in the attic above the fluffy stuff(?), otherwise the SRI of the shingles wouldn't make a heluva lot of difference. The parasitic duct gains as well as parasitic air-handler driven infiltration need to be included too. It's possible or even likely that your ducts aren't well balanced, driving at least a half ton of infiltration load when the AC is running, along with ton of parasitic gain from the ducts & air handler being outside of conditioned space. Unless you have massive amounts of unshaded west facing window there's no way you have more than 3 tons of raw cooling load in an 1800' house- it's almost certainly the system design/implementation that's killing it. The basic load is most likely less than 2 tons.

 

[Taylorjm]

The heat load is only the heat load, not a equipment sizing spec for what it takes to to rocket the temp up quickly by 20F. It's important to know what the steady state loss numbers are to appropriately size the system.



Sounds like your house had the air handlers & ducts up in the attic above the fluffy stuff(?), otherwise the SRI of the shingles wouldn't make a heluva lot of difference. The parasitic duct gains as well as parasitic air-handler driven infiltration need to be included too. It's possible or even likely that your ducts aren't well balanced, driving at least a half ton of infiltration load when the AC is running, along with ton of parasitic gain from the ducts & air handler being outside of conditioned space. Unless you have massive amounts of unshaded west facing window there's no way you have more than 3 tons of raw cooling load in an 1800' house- it's almost certainly the system design/implementation that's killing it. The basic load is most likely less than 2 tons.

The air handler is in the basement which is in conditioned space. House was built in 1977 so it has had the same ductwork runs since then for heating and cooling. Pushing cold air from the basement up to the second story is almost impossible without shutting off all the downstairs vents and increasing the static pressure. Hench the booster fans on those runs. All the houses in the area were built the same way and the same time. You can tell when the summers start because you start to see houses with central air and room air conditioners in the upstairs windows. Your response of only needing 2-3 tons is why I don't take any stock in the calculators anymore. Experience tells me they don't work for everything.

 

[Sponsor]

 

[Dana]

The cooling loads and heating loads aren't very proportional to one another floor by floor, and it's often the case that ducts well balanced for cooling are out of balance for heating & conversely.

Even in cooling-only or heating-only mode it's often difficult to get floor-to-floor balance when operated as a single zone. But even taking that into consideration its sounds like a potentially poor duct design, and possibly an air-leaky house (&/or leaky ducts).

Are there return registers in every room that has a supply duct? An Energy Star duct system would have room to room pressure differences no more than 3 pascals (0.012" water column) under all air handler speeds and conditions, room doors open or closed. Without sufficient return paths rooms with supply ducts only get pressurized relative to outside, using "The Great Outdoors" as part of the return path, with outdoor air getting sucked into rooms being depressurized by the air handler. Even when the supply ducts aren't properly sized for the cooling cfm requirments, proper returns do make a difference.

It is possible to get this right, but the HVAC system can't make up for any deficiencies of the home's thermal envelope, especially when operating multiple floors as a single zone.

Solar gains through windows tend to be higher on upper floors than lower floors due to generally lower shading factors. Air-leaky attic floors & basements also introduce an uncontrolled stack-effect infiltration & stratification issue just to make it more interesting/complicated. With a large stack effect cold air is drawn in to the basement & lower floors in winter, and the warm air is rising to the upper floors (and leaking out into the attic by myriad paths). When that is happening the upper floor doesn't need as much cfm from the heating system as the lower floor for both floors to remain comfortable. Often houses built before air-tightness specs for the building went into the IRC had the duct systems over-supplying the first floor so as not to overheat the top floor. During the cooling season the stack runs in reverse, but isn't as powerful, but that means hot humid air from the attic is seeping into the upper floors, which are already getting more solar gain than the lower floors, as well as some direct conducted gain through the attic insulation, all of which calls for a higher cfm for the upper floor than the lower floor. This can't be fixed 100% with simple tweaks, to just the house or just the HVAC system, but there are several generic fixes that help in both the cooling & heating season.

The path to better comfort levels starts with air-sealing the house, followed by fixing any deficiencies in the insulation (including the basement walls & band joists.) This includes temporarily pulling up any existing insulation in the attic to be able to seal the seams of partition wall top plates, and all electrical, plumbing stack & flue penetrations etc.

If the attic insulation is all l0w density fiberglass from the 1970s or 1980s, that stuff is translucent to the infra-red radiating down from the hot roof deck, and also not very air retardent. A quick fix is to blow 3-6" of cellulose on top of the fiberglass, which forms a much more air retardent and IR-opaque layer, restoring the performance of the fiberglass to it's label ASTM C518 tested labeled level. Most newer fiberglass blowing wools have additives in the glass to reduce IR translucency, but none are as opaque as cellulose. This is a fairly cheap & easy DIY, but only after the air sealing has been properly addressed.

If the upper floor windows are clear glass double panes (or single panes with clear glass storms), ANY low-E replacement glass (or low-E storm windows) will reduce the solar gain on the upper floor, lowering the cooling load.

Before diving into it or paying for a lot of band-aids it's worth taking the time to review Nate Adams' take on home comfort and basic HVAC (or slightly deeper HVAC) in his short videos & free downloable chapters from his book. Nate started as an insulation contractor in Cleveland OH, but quickly figured out that people weren't getting the comfort they were paying for by attacking it piecemeal, and often had to replace existing equipment with properly sized equipment and re-doing a lot of poorly thought out poorly implemented weatherization & insulation. His track record is pretty good. He even coined a term for what it sound like you're experiencing, namely the "Gulf of Disappointment", where a lot of work has been done, but comfort has not (yet) been fully achieved.

BTW: One characteristic of grossly oversized systems is that the furthest rooms (in this instance, the upper floor) don't get sufficient cooling simply due to operating at too low a duty cycle. Adding duct boosters doesn't help the duty cycle- it only delivers more cooling when the thing is actively running, resulting in bigger temperature swings during & between cycles. Does the 3-tonner run continuously for an hour or more on hot afternoons, or does it satisfy the thermostat on the first floor too soon and cycle off?

Most 1800' houses would have a cooling load of about 1.5 tons (give or take a half ton), very few would need three:

The cluster of homes slightly smaller than 2000' run between a ton per 600' (=three tons for an 1800' house) for the very worst performing house, to a ton per 1400' (1.3 tons, for an 1800' house), with the median being about a ton per 1000' (1.8 tons for an 1800' house.)

Bigger is the opposite of better. Replacing the 3 ton unit with 4-5 tons of AC wouldn't cool the upstairs any better, and would probably just make it worse.

 

[Taylorjm]

Sorry, but you can run all the calculations you want. In real life, many times they don't mean anything. The house has had insulation added to the attic. All the windows were replaced with low E glass. Yes there are returns in every room. One thing that you were incorrect on though, you said replacing a 3 ton unit with a 4-5 ton wouldn't help. In reality, your forgetting that when you go from a 3 ton to a 5 ton rated air handler, you are going to a larger blower with more cfm capacity. The duct design is typical in the area, basically a trunk with runs off it. Not much to mess up. Ducts were scoped and make sure nothing was disconnected. Roof was recently replaced and all the soffits were replaced with perforated and baffles were installed and proper venting to make sure the attic is breathing. You can rely on calculations all you want, and just like my neighbors, you will be putting ac units in your windows.

The 3 ton unit would run continuously for the entire day just to try and maintain the temperature on the main floor. In fact, when the outside temp would reach 90 or above the unit would never reach it's temperature of 70 on the main floor, which means the upstairs would be unbearable hot. Moved to a 5 ton with the larger blower before the zoning and booster fans, and the unit would actually be able to keep up the main floor temperature and cycle off after a few hours. The upper floor was better, but still had at least a 5-7 degree temperature difference. Moved to an emc zoning board with dampers and blowers in the upstairs runs and thermostats in all the rooms. I probably could have just put the upstairs on a single zone except the kids like to keep their bedroom doors closed so putting a thermostat in the hallway wouldn't be accurate, so went with thermostats in every room. There are a few zones that do not have dampers and stay open to make sure there's enough cfm running through the system plus I have a sensor right above the coil that will turn the compressor off if the plenum temp gets too low because only one zone is calling. It's never activated and the coils have never froze up.

The part I love is when the hvac people say "it's more efficient for the unit to run all day than to cycle on and off and that's how they are sized". Um, yeah. Sorry but that doesn't make any sense. In their mind, it's more efficient to run the unit 12 hours a day, then to zone it and let it serve the rooms that need it and cycle on and off as needed but only run say 6 hours a day. How is that more efficient? Oh, then the conversation turns to "oh, I meant it will be more comfortable to have it run all day" Well how can it be more comfortable if it's never reaching the set temperature and turning off? If I set the thermostat to 70, that means I want it 70, I don't want the unit running 12 hours a day to keep it 72. If it's 72 in the house, it's not working correctly or efficiently or making me comfortable. lol

 

[Taylorjm]

Buy the way, I'm not trying to act like an a$$hole or anything. I'm just telling you what I've seen happen in homes in our neighborhood and what I did to solve it. I'm sure it's not 100% correct, but in our case, it works. I'm not trying to say anything your doing or that all calculators are wrong, I'm just saying that sometimes, people need to look at what works for the home instead of trying to make the calculator a one size fits all type of solution.

 

[Dana]

Right sized modulating systems can run 24 hours per day and still save money. A Manual-J is just a starting point, not an equipment spec. Math works. Measurements work too.

As stated previously, it appears you have been limited by an inadequate duct design. It sounds like some of those limitations have been overcome by blasting it with a more powerful air handler and smarter controls, and somehow with a lot of micro tweaking it almost works.

Blower door testing & rectifying the air leakage in the house along with testing the duct leakage & duct balance probably would have been a better place to start, along with directly lowering the loads by fixing deficiencies of the house. Ducts that aren't mastic-sealed can leak quite a bit of air even when they aren't disconnected.

Improving attic ventilation has nothing to do with heating or cooling loads. Attic ventilation is about purging moisture (moisture that can end up in the attic from stack effect drives in winter in a leaky house.)

FWIW: A few years ago I consulted on a circa 1960 "mid century modern" house in central MA with ~2000' of fully above grade conditioned space and 2000' of walk-out basement. It was equipped with 200,000 BTU/hr of heating equipment and ~8 tons of cooling equipment and was fighting comfort issues both winter & summer. It now has a 60,000 BTU/hr 2 stage condensing furnace and a 3 ton air conditioner and has NEVER been more comfortable. Yes, there were building envelope improvements, but they weren't radical. There were two Manual-Js performed- initially on the "before improvements" picture, then the "after" improvements version. The oversize factor of the equipment installed was more than ASHRAE's 1.4x, but well under 2x.

When the owners had me over for dinner over this past winter holidays they commented it had been transformed from a Holy Hell Hole they were ready to dump to now the most comfortable house they have ever lived in (and that included their prior home which had all radiant heat, which was pretty cushy, to be sure!)

I'll stick with doing the math and follow where it leads. Clearly YMMV.

 

[Taylorjm]

How can you say that attic ventilation has nothing to do with heating or cooling loads? If an attic isn't ventilated properly, the heat is going to radiate and increase the temperature of the rooms below it. That's just plain thermodynamics and heat conduction. Even though heat rises, if the heat doesn't have anywhere to go, it will eventually reach the attic floor which is the living area ceilings.

So, if the issue in my home is inadequate duct design, are you recommending the home be torn apart to replace the ducting just so your 3 ton unit in your calculations will work?

 

[Dana]

How can you say that attic ventilation has nothing to do with heating or cooling loads?

I can say it because that is NOT the purpose or function of attic ventilation. It has some effect on heating cooling, but it's much smaller than those who don't do math generally think.

If an attic isn't ventilated properly, the heat is going to radiate and increase the temperature of the rooms below it. That's just plain thermodynamics and heat conduction. Even though heat rises, if the heat doesn't have anywhere to go, it will eventually reach the attic floor which is the living area ceilings.

In homes with reasonable pitch to the roof the roof deck and attic temperatures between poorly vented and well vented are fairly self-limiting due to exterior convective cooling of the roofing. Yes, peak attic temps might be 20F hotter in a completely unvented pitched roof, but not 50F hotter.

With IR opaque insulation types that have at least a measurable amount of thermal mass, such as 10-12" of cellulose the roof deck temperature isn't being radiated through to the ceiling- the attic temp can peak at 130F and still not be delivering a huge cooling load to the ceiling. With somewhat IR translucent old-school fiberglass the highest temperature in the insulation layer is an inch or two below the surface, and hotter than the average air temperature in the attic, insulating the ceiling against a higher temperature, with less R-value between the hot layer and the ceiling.

Heat doesn't rise- it travels from hot to cold. (Go ahead and put your hand under the bottom of a hot pan- do feel the heat flowing down much?) Hot air is less dense than cold air, and will stratify in an enclosed space, or convect if air is allowed to flow. So yes, there is some convective cooling of a soffit-to-ridge vented attic, but but effectively none in a gable-vented attic. The cooling benefits of attic ventilation are highly over-rated and often over stated. The purpose of attic ventilation is again MOISTURE related, not heating or cooling related.

So, if the issue in my home is inadequate duct design, are you recommending the home be torn apart to replace the ducting just so your 3 ton unit in your calculations will work?

Again, the 3 ton unit was for an all new system, with properly balanced ducts. It did not require that "...the home be torn apart ...", but yes, it took some amount of demolition & building/rebuilding to get there for that Mid-Century Modern house. But since your ducts are already fully inside of conditioned space it would likely be easier deal with than that house.

Other than generic issues like air sealing and duct balancing it's hard to make firm recommendations for your house.

Again, are there defined and adequate return paths for every doored-off room with a supply duct?

Has anybody even measured the room to room pressure deltas (with doors closed)? There are about a half dozen ways to provide better returns to get the room-to-room pressures under control.

Has the house been air sealed? A single round of blower-door and IR-imaging directed air sealing can make a large difference (and mitigates against wintertime dry-air problems considerably in your climate.)

How much fluff (and what type) is in the attic?

Care to describe your second floor window types? Any roof overhangs to shade the south side? Any west facing windows?

You can figure out a lot of what does/doesn't affect the loads (and comfort) playing "what if" games with Manual-J tools. I'm not crazy about either LoadCalc or CoolCalc, but they're both better than a WAG. The input parameters on those tools are somewhat limited- there are much better load calculation and energy modeling tools out there, but they cost money, and aren't particularly more useful in the hands of newbies (or HVAC contractors prone to putting thumbs on scales.)

Take a look at some of the case studies on Nate Adams' blog site. Some of those home comfort fixes were far more extensive than others from a "...home be torn apart ..." perspective than others, but he fixes the obvious (to professional home performance types) and does the math, and generally gets it right without turning the house into a construction/demolition zone.

 

[Jeremyt]

Dana:
The garage doors are new (installed in August 2018). The exposed stem wall is minimal because its 12 inches at it most exposed portion, but there is a water table trim board overlapping the stem wall. So in reality theres only 4 inches or so of exposed stem wall to the outside. Theres about 8 inches of overlap from the water table trim board and its about a half inch gap from the back side of the house sheathing to the concrete stem wall.

taylorjm:
Yes the plan is to keep the garage around 40-45 unless I am planning on spending a considerable amount of time out there. Half of the back wall, and half of the ceiling is abutting a conditioned space, (half the ceiling joins to the bonus room, and half the back wall joins to the living room). Maybe this has something to do with the calculations where they are? I can imagine that knocking the ceiling from 1200 sqft to 600 sqft and the back wall from 500 sqft to 250 sqft can only help. I can't imagine a 120,000 btu heater in that space being efficient or beneficial. The boiler in the 2800 sqft house was a 120,000 btu unit. That heated the entire house, and the domestic hot water supply. It wasn't efficient because it was old and lacked a normal service interval, but it always had plenty of house heat and hot water..

 

[Taylorjm]

A 120,000 btu boiler is completely different than a forced air furnace. Going by what a house has isn't comparing apples to apples. Plus you have to remember that everything in the garage will be 40 degrees. Your trying to heat an area that's 50' long, your going to need something that blows hard and hot to get the air circulating through a space that's 50' long. I have another house with a 20x20 garage and a 45,000 btu gas forced air hanging heater that works, but it's something you need to start up and go back inside the house for 30 minutes before the room will go from 40 degrees to about 55 degrees so you can work in there comfortably. So that's 1/3 the size of your garage. So I guess it's up to you what size unit you go with, but also remember, if you are thinking about a heat pump, the temperature of the air exiting the unit isn't near as warm as a gas forced air furnace, which means it's going to take even longer for your garage to heat up.

 

[Jeremyt]

Dana you have a direct message.

 

[Dana]

Dana:
The garage doors are new (installed in August 2018). The exposed stem wall is minimal because its 12 inches at it most exposed portion, but there is a water table trim board overlapping the stem wall. So in reality theres only 4 inches or so of exposed stem wall to the outside. Theres about 8 inches of overlap from the water table trim board and its about a half inch gap from the back side of the house sheathing to the concrete stem wall.

At a 20F temperature difference, say 45F indoors, 25F outdoors (the January mean temp in your area) every square foot of exterior exposed 8" stem wall is losing more than 15 BTU/hr. It adds up fast. Installing even R8 (a couple inches of EPS) cuts that to about 2 BTU/hr, almost an order of magnitude. Where ever you can insulate it, it's "worth it".

taylorjm:
Yes the plan is to keep the garage around 40-45 unless I am planning on spending a considerable amount of time out there. Half of the back wall, and half of the ceiling is abutting a conditioned space, (half the ceiling joins to the bonus room, and half the back wall joins to the living room). Maybe this has something to do with the calculations where they are? I can imagine that knocking the ceiling from 1200 sqft to 600 sqft and the back wall from 500 sqft to 250 sqft can only help. I can't imagine a 120,000 btu heater in that space being efficient or beneficial. The boiler in the 2800 sqft house was a 120,000 btu unit. That heated the entire house, and the domestic hot water supply. It wasn't efficient because it was old and lacked a normal service interval, but it always had plenty of house heat and hot water..

Wall & ceiling area with conditioned space on the other side do not have any heat loss or gain, and have to be zeroed out in the calculations. (When it's under 50F in the garage there would be a measurable heat flow from the conditioned space toward the garage.

A 120K boiler in the 2800' house is ridiculously oversized for the space heating load (and maybe even oversized for the radiation?), likely oversized bo be able to deliver sorta-reasonable domestic hot water deliver with an embedded tankless coil (?) and could cover the additional loads of the garage & bonus room if desired, but it won't air condition those spaces. If it's an oil or propane burner it'll be more expensive operating cost than heating with a heat pump, but if it's only going to be 45F in the garage most of the time, an 80- 100K hydro-air coil could blast it up to temp pretty quickly no matter what the outdoor temp.

 

[Dana]

A 120,000 btu boiler is completely different than a forced air furnace. Going by what a house has isn't comparing apples to apples. Plus you have to remember that everything in the garage will be 40 degrees. Your trying to heat an area that's 50' long, your going to need something that blows hard and hot to get the air circulating through a space that's 50' long. I have another house with a 20x20 garage and a 45,000 btu gas forced air hanging heater that works, but it's something you need to start up and go back inside the house for 30 minutes before the room will go from 40 degrees to about 55 degrees so you can work in there comfortably. So that's 1/3 the size of your garage. So I guess it's up to you what size unit you go with, but also remember, if you are thinking about a heat pump, the temperature of the air exiting the unit isn't near as warm as a gas forced air furnace, which means it's going to take even longer for your garage to heat up.

The temperature of the exit air isn't what's going to make the difference. The delta-T of the incoming to exit air, and the cfm is what determines the BTU rate, not exit air temp.

That said, at residential room temperatures most ductless mini-split heat pumps have exit air temps comparable to that of condensing gas hot air furnaces, and when it's sub 50-F in the garage they have substantially more capacity & efficiency than shown in an AHRI test.

When the entering air temp is 40F many fossil-burner furnaces are susceptible to damage from acidic exhaust condensation in the heat exchangers. (Units designed for industrial applications are somewhat better protected from those issues, by design.) A hydro-air solution off the boiler wouldn't have that issue if designed correctly.

Bringing the garage up to temp quickly isn't always important as it seems. Even if sized at only 1.4x the design load (the ASHRAE residential spec) a weekend car hobby mechanic can turn up the temp before they go to bed on Friday and the place will be just fine before breakfast, any night of the year.

A single zone 4-way ceiling unit big enough to heat half the garage would have plenty of throw in a 50' wide space. They are used all the time in commercial & retail applications. Even wall units with enough would usually not have a problem, but a bit of forethought has to be put in to where to mount them for maximum benefit.

 

[Jeremyt]

I went through the garage calculations last night, and I finally saw the infiltration tab. I set that to tight instead of average and it brought the BTU demand down 5k or so. I have attached screen shots. I also measured all the bonus room dimensions, and did some calculations for that. Those are also attached. Its calling for 7k in the bonus room on CoolCalc, but only 4k on LoadCalc. Again a vast difference in the figures. In the bonus room, there really is no walls. Its cathedral ceilings to a 4 foot tall knee wall, so the majority of insulation up there will be R30.

 

[Dana]

Again, look at where they differ. The CoolCalc attributes about 3500 BTU/hr (half the total ~7k) to walls, compared to LoadCalc's ~1200 BTU/hr. That's a huge difference in just that one factor- take a look at your input selections for wall type and wall area on both tools to resolve the differences.

CoolCalc is showing over 1000 BTU/hr for infiltration, compared to zero from LoadCalc. In under the "construction" tab in LoadCalc it's showing "-----------" . Try selecting different tightness levels there, see what it shows. LoadCalc's window losses look a bit higher than CoolCalc's, but that's probably because of how limited the window choice options are in LoadCalc (one reason why that tool overestimates for houses with pretty good or even code min low-E windows.)