Some claim that modern furnaces have minimized off cycle losses to the point they no longer matter. Oversizing has become a comfort issue rather than an efficiency issue. Is this really true? What are the effects of the system cycling more frequently when it comes to ductwork losses in unconditioned space? Is it better to have 1 long cycle instead of 2 shorter ones? Some argue that modulating furnaces operate at a LOWER efficiency due to greater ductwork losses.

Tags: CPH, cycles, duct

Views: 2071

Reply to This

Replies to This Discussion

Hi Bob,

Many AC and heating contractors have told me the same thing over the years;  If you can do ANYTHING that can cause a unit to cycle on and off when it did not before, whatever life that unit had left, you just extended the life of that unit by 100%.


We were talking about the AC units every time.  I assumr the heat pumps would benefit from that also. 


Energy efficiency vs extended equipment life?  I think those go hand-in-hand myself. 


I assume the less amount of cycles for the heating unit would result in extended unit life in the heating  unit also.


But, then again, you know what happens when we assume.  :)

Want to spot a simpleton? It's a guy who thinks EE and comfort are disconnected. 

Want to spot a thief? It's a guy justifying selling a larger furnace. 

Cycling losses can be HUGE, because they are hard to measure they are often dismissed. I've had projects I thought would save $2-300 a year save $5-700, and the only place I can find the difference ties all the way back to being aggressive on sizing. 

If we don't get an "equipment undersized" warning, we assume it's oversized. This approach has also resulted in surprising satisfaction levels from customers:

"Can't hear the thing running"

"Never think about whether I'm comfortable because I always am"

"Doesn't blow hot and cold"


In my opinion you are trying to get very detailed about equipment that is designed to perform in a wide range of climates and environments.   Unless a furnace is operating at it design max, it will always be oversized for the load.  (Full capacity at 0 degrees vs 0 capacity at 65 degrees +)  

Oversizing is a consumers comfort issue because they live with it 24 hrs a day but they only pay the bill ( efficiency related ) once a month.  They are two sides of the same coin and the homeowner does not have to sacrifice one for the other.

The question about duct losses and system cycling is interesting.  A modulating furnace would be the most efficient IF the DUCT IS EFFICIENT AS WELL.  When you ask about "duct work losses in unconditioned space"  are you asking about leakage, poor or no insulation or poor fittings and layout?

If it is leakage - then the shorter a cycle the better since the system will only lose air when the system is running.

if it is poor or no insulation - then the longer cycles work better.  There will always be heat loss into the duct when a system starts up.  The greater the temp difference between the air temp and the duct temp, the greater loss into the duct materials and unconditioned spaces.  the system will operate more efficiently once every thing is up to the same temp.  Then there is no loss into the duct material and (if well insulated) the surrounding space.

Steady state is always more efficient but that cannot be achieved by looking at the furnace or A/C alone.  You need to factor in the building loads, duct work and location, people and their habits and on and on.

Each application is unique and the end result will be a compromise of comfort, efficiency and costs.  


I can offer some empirical information, general and highly variable, but borne out with regularity.

The "cycles per hour" metric we understand is applicable to on-off equipment of fixed capacity; at best and often rarely it is most closely sized to the load on a design day and is increasingly over-sized every day that is less extreme.

Modulating-fire equipment, properly sized and applied, "becomes the appliance you need, when you need it", at least down to its minimum capacity. 

For example, a home with a design heat load of 50 MBH might be served by an 80 MBH input/68 MBH output 85% efficient boiler, only because there are no smaller increments that will do the job. This is of course means that on the coldest design day, the boiler is 36 percent over-sized (68/50 = 1.36) and it only gets worse as it gets warmer, with increasingly more cycling.  If the house has limited radiation, it may cycle even more- there is insufficient radiation to deliver the heat produced and the boiler cycles off its high-limit controls. At best the boiler will operate about 45 minutes out of an hour at design temperatures. This is much as Steven Lewis described above.

An identical home next door, same heat loss, has an 80 MBH condensing boiler or, let's go for it, a 60 MBH input condensing boiler, because there are a few in these smaller sizes. The output would be 54 MBH at 90 percent efficiency, which is pretty close, but it will modulate to meet whatever load occurs, fairly closely. As the weather warms, modulation will take the output down from 54 MBH to say, 10.8 MBH, assuming a 5:1 turn-down ratio.  Only then will cycling occur.

Now,  the empirical observations I have had, even with a fixed capacity and modulating capacity boiler of the same input for similar houses:  One might assume that the normal calculation would be a proportion of AFUE or other efficiency metrics, that a boiler or furnace running at 95% efficiency would use 89.5% of the fuel that an 85% efficient appliance would consume. (0.85/0.95 =0.8947..)

The reality is, savings of 20, 25 or 30 percent are commonly found, based on the boiler alone, all other things being equal.  The differences, the efficiencies, I can attribute to modulation versus cycling. Less standby loss and ramp-up time when fuel is burned but not directly delivered.

To the last sentence you posted, Bob, "Some argue that modulating furnaces operate at a LOWER efficiency due to greater ductwork losses"- I cannot get my head around why that might be. I was thinking the same issues that Steven Lewis mentioned in his post but am still not certain.

I would argue that modulating furnaces tend to operate at lower temperatures (moving more air at lower temperatures for the same heat output) and thus duct thermal losses would be reduced somewhat and be relative to outside temperature at some level, versus a constant temperature in bursts.  But if the ductwork is properly sealed (sealing first!) and insulated, the more constant operation points to greater space satisfaction, and that is where it counts.

Let me know if I misunderstood the point being made, please?

My main point was ductwork in unconditioned space. When a furnace cycles off the ductwork will "level out" to the surrounding air temp. You can feel this every time the furnace cycles and blows a bunch of cold air in the house upon start-up.

I'm not sure it it's better to have a higher or lower delta T across the furnace. If a house is 70 degrees with a 40 degree attic, is there more heat loss through the ductwork with 100f or 120f supply air from the furnace? If the ductwork looses 10% that would result in a 6 degree vs 8 degree loss. 6 degrees is 20% of 30f delta T when the furnace is blowing 100f air. 8 degrees is 16% of the 50 degree delta T with the 120f supply air.

Now if we could get builders to stop putting ductwork in unconditioned space to begin with, but that's another topic...

I see now what you mean.

Agreed that installing any ductwork outside the thermal boundary unnecessarily, should be avoided. 

But to take this a step further and by the numbers,  code minimum for insulation on ductwork outside the thermal boundary is R-8 (u=0.125 not counting air film resistance).

If one has 120 SF of ductwork with relatively low mass and specific heat and a 50 degree delta-T, that is 750 Btus per hour held constantly. Raise the delta-T to 100 degrees F. and you have 1,500 Btus per hour.

If the duct is moving say, 1,000 cfm, that is a temperature loss on the air of about 1.4 degrees F. steady-state. Not great, it is a loss, but is not that significant; it would be about 2.8% if considering a 50 degree rise across the furnace. 

An off-cycle start would tend to have that ductwork begin at equilibrium, say 120 SF at 1.5 lbs. or 180 lbs. of sheet metal. With a specific heat of 0.122 (Btu/lb./degree F.), that is about 2,200 Btus to get that metal up to temperature.  Sure, you would have an initial cool burst but it would not last long, especially with plenum controls prior to fan start.

I respectfully submit that it is better to have a lower delta-T across the heat exchanger, within reason. (Granted the penalty is the energy cost of moving more air at a lower temperature.)  The principal reason I state this is that lower temperatures promote cooler flue gases and thus amplify condensing, if that is the principle being used.

I think we will both agree that sealing the ductwork and adding insulation beyond code, makes sense where installation outside the thermal boundary cannot be avoided. Sealing applies regardless, inside or out.

Now consider what happens in a modulating furnace. The 750btu loss is relatively constant even though the furnace input can vary from 40% to 100%. Furnaces are generally GROSSLY oversized and the mod furnace will spend much of the winter cycling on/off at the 40% level. Only when temps hit extremes will the furnace actually modulate up from the 40%. Another issue with modulating furnaces is the furthest rooms from the furnace tend to see a bigger reduction in airflow and will be colder than the rest of the house.

Of course the correct solution is to tighten the house to where you can heat it with less than 20BTU per sqft and install a correctly sized furnace. I don't see builders doing either of those on a widespread basis any time soon...

Agreed, the loss is relatively constant (save for the differences in temperature, narrowing during milder weather and widening with cold, thank you).

As you note, improper sizing coupled with poor duct locations are often existing conditions with no cheap solution. A separate issue we agree.

Now, regarding the furthest rooms having a reduction in airflow- I respectfully submit that the type of furnace, modulating or not, is irrelevant on that issue. It is a function of air balancing and one hopes, proper air sealing. In fact, current furnaces using ECM motors give more flexibility and higher external static pressures to overcome duct resistance. If that, coupled with balancing does not work, then air leakage on the longer runs becomes a prime suspect in my experience.

Bob, I'm really enjoying watching you walk this path. You are questioning everything, and asking the right questions.

Wow Bradford, you are hitting every nail squarely on the head. It appears you are a very rare bird who has built his expertise and intuition by going back and actually measuring, rather than the preferred path in this industry of both regurgitating dogma and simply making things up as you go along.

To fill in some voids Bradford left around the periphery and further support his conclusions: 

  • I've seen the same savings and arrived at the same conclusions. There are huge losses in cycling. 
  • He didn't tie losses to comfort. There are additional losses in cycling due to discomfort. 
  • When it comes to furnaces, most "pumps" are grossly oversized to the "pipes" which leads to major pumping inefficiencies. Modulating furnaces have some hope of "pump" match at least when operating at lower outputs. 
  • Size of heat exchanger doesn't change. Lower burn rate allows better heat transfer as it effectively "increases" the size of the heat exchanger. 
  • Lower heat exchanger temps increases condensing efficiencies.
  • Colder supply air - PARTICULARLY through unconditioned space - has geometrically lower losses. Longer run times will not give these savings back unless the duct is leaky. 

"Another issue with modulating furnaces is the furthest rooms from the furnace tend to see a bigger reduction in airflow and will be colder than the rest of the house."

Incorrect. With 1/2 way decent duct, even airflow occurs across a broad range of static pressures. Too little airflow is so rare you are unlikely to ever see it, except in "fan on" settings for some poorly designed ECM furnaces.

You should be able to take esp down to .08 before control of airflow becomes an issue. I've installed equipment delivering .16 on low and received rave reviews for comfort, noise, and efficiency.

Ultimately, what we want is BTU not airflow. Get too focused on airflow and you aren't solving the right problem. 

Even in unconditioned space lower/longer is better. Gavin Healy and Dan Piekunka have measured duct loss through unconditioned space of properly insulated and air sealed duct. I don't have the exact figures, but by dropping delta T about 25 degrees the losses were cut 2-3 times. Longer run cycles clawed back a tiny fraction of the savings. 

Lower airflow IMPROVES comfort in the furthest rooms because duct temperature uniformity occurs and with it BTU delivery. It is this condition that allows uniformity of room delivery temperatures. As it is coupled with lower BTU output this means lower duct temperatures and significant energy saving as well. 

It seems Bradford is a wet head, so let's think about this in terms of radiators. If you rapidly heat near radiators and satisfy the thermostat, the far radiators never get warm and neither do far rooms. Put lower temperatures to the radiators, attempt to match loss and never shut off, and all radiators become the same temperature. Same concept with airflow, some differences being water is much easier to control than air, leaks are unacceptable, and easily spotted.

Now, regarding the furthest rooms having a reduction in airflow- I respectfully submit that the type of furnace, modulating or not, is irrelevant on that issue.It is a function of air balancing and one hopes, proper air sealing. 

I'd rather see those sentences read thus:

Now regarding the furthest rooms having a reduction in airflow- I respectfully submit that with proper air balancing, air sealing, insulation and sizing, the modulating furnace dramatically improves appropriate BTU delivery to all rooms under the widest load conditions. 

All good points and refinements, Tedkidd- and by the way, call me Brad. And yes, I am a confirmed wet-head but also a dry-head on commercial/institutional projects more than residential.

Anyway, agreed that the business is moving BTUs and not to put too fine a point on that ultimate objective, we have to deliver the BTUs by some means- being air or water; it's all mass flow at that point but now at variable temperatures too.

My point about balancing remains, not so much to delay delivery to the thermostat-holding space, but to deliver the BTUs at the appropriate rate to all spaces at the same time.  Your point is well taken though and there it goes, back to cycling. 

The constant circulation principle is one I favor highly. On a water system, constant circulation of water with TRVs on most radiators and the thermostat firing the burner has been a proven simple method in Europe for decades. Very even temperatures. Same can be said for air, but the delivery cost is a bit higher.

Still doesn't explain how you know these things that most never learn.

You did a lot of M&V, reconciled to projections, saw projections didn't correlate to reality, and spent a lot of time thinking about the "why". 

Why, curious? Was it all on your own time? 

One of my several skill sets (I really hate that term, it harkens to nursery school!), is commissioning. By day I am an HVAC design engineer and PM but oversee projects from conception to completion, so it comes with the territory.

My side work is residential energy audits and enclosure testing with IRT, plus general consulting on all kinds of HVAC systems. The two areas enhance themselves and I learn new things every single day which makes it all worthwhile.

Regarding the commissioning, going back years, I was always surprised when a good design (not just mine, but anyone's), did not yield the results promised or hinted at.  Mostly comfort complaints which is how they show up, but then once those would be resolved, the energy use would go up enough to defeat some of the purpose. You would go back to find broken belts, motors running backwards or not even wired sometimes.

When someone attempted to restore comfort on the fly, one would find as evidence bypasses left open, dampers left open, valves opened, pump VFDs set to 60 Hz. without a responding variable... whatever would solve the complaints.  One could tell that the TAB contractor had marked the original damper positions so the new positions were evident. These devices would be left in those positions forever, unless someone was monitoring energy use from a solid basic benchmark. Cannot manage what you cannot measure, after all.

The beauty of all this, is that it all ties together, design and performance. All about connecting the dots.

How about you? What brings you here and what makes this fun for you?
I say, "fun", but a mentor once told me, "any job you take needs one of three things. Money, Glory or Fun. If none are likely, refuse the job."

But fun is the best part, IMHO   :)



Latest Activity

Safe-R Insulation commented on Chris Laumer-Giddens's blog post Rock Wool Insulation for Floor of High Performance Tiny House
"Thanks for sharing this much information about roof wall…"
4 hours ago
Lindsay Bachman's 4 events were featured
20 hours ago
Profile IconChris Hoskins and Ken Goodrich joined Home Energy Pros
20 hours ago
Diane Chojnowski commented on Diane Chojnowski's event Better Buildings Peer Exchange Call: Home Improvement Catalyst: Keeping the Ball Rolling with Homeowners by Delivering More Value
"Hi Victor, All the BBRN Peer Exchange Calls are archived so you can listen anytime. Check out over…"
Victor Rols Lindsay commented on Diane Chojnowski's event Better Buildings Peer Exchange Call: Home Improvement Catalyst: Keeping the Ball Rolling with Homeowners by Delivering More Value
"Hi Diane, Unfortunately some of us are unable to partake in this webinar/ forum. Is there a…"
Ken Summers replied to Gregory Cisco's discussion Abandoning under-slab HVAC ductwork which fills with groundwater seasonally.
"My largest concern is moisture. I have seen mold forming on interior walls in the heating season…"
Seth Romme replied to Sonja Persram's discussion Energy Savings Guarantee for single family dwellings in cold climate??
"Hi Sonja We provide home energy optimization services and provide a 5-year heating and cooling…"
Lindsay Bachman posted events


  • Add Photos
  • View All

© 2017   Created by Lawrence Berkeley National Laboratory.   Powered by

Badges  |  Report an Issue  |  Terms of Service