Thermal Mass of Metal Ducts & It’s Effects
I’m trying to decide how and what type of ducts to run for a 1 Ton MVZ air handling unit described below.
Background:
My home is a 2000 sf 2-story with vented attic in San Jose California, zone 3. The floor of the attic is foam sealed and insulated with Owens Corning R38 pink insulation batts. I’m replacing my old central air furnace and AC with a Mitsubishi 12kBtu MVZ air handler installed in the upstairs mechanical chase (conditioned), with overhead supply through ducts in the unconditioned attic. The MVZ will heat/cool 3 small bedrooms (total of 430 sf, 4.7k/4.5kBtu cooling and heating load), a small bathroom (45 sf), and the master bathroom (143 sf). I have no option except to run the ducts in the unconditioned attic. Attic temperature on hottest summer days is ~130-140 deg F, and probably near freezing on the coldest winter days.
Question:
Does the thermal mass of a metal duct system ever become a factor when trying to choose between flex vs. metal? I’ve never seen it discussed, and online search turned up nothing on the topic. I’m worried a little about the large thermal mass of the metal ducts in an unconditioned attic. Of course the metal ducts will have the best pressure drop performance, but I’m more worried about thermal inertia and higher heat capacity of the metal ducts vs. light weight flex duct, in this very hot and cold space.
IF my duct runs were long, and I had limited available static pressure, I would go with metal period. The MVZ will be only a few feet from the bedrooms (it’s located right at the corner of each of the 3 bedrooms). For noise reasons alone, I think I will just use flex for these.
But the master bath will require about 25 ft of duct and the other bath about 15 ft. The thought of using metal duct in a hot unconditioned attic just seems like it will add a lot of inertia to the thermal load, but I do still need to calculate the effect of flex vs metal on the pressure drop. The MVZ has 0.3, 0.5 and 0.8 IWG available static pressure.
As I finish writing this, I realize that the answer is probably that the thermal mass of the ducts only effects the transition time at the beginning of a heating or cooling cycle. It will take longer to reach equilibrium, but at steady state the energy efficiency of the metal ducts will be better than the flex.
Just curious what the experts have to say.
Thanks ahead for any thoughts on the matter.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
" I realize that the answer is probably that the thermal mass of the ducts only effects the transition time at the beginning of a heating or cooling cycle. It will take longer to reach equilibrium, but at steady state the energy efficiency of the metal ducts will be better than the flex."
That's true. The heat capacity of 28 gauge steel is about 1500 J/(square meter degree C). If you have 10 square meters (~100 square feet) of duct surface area, and a 40 C (72 F) temperature change from attic temperature to A/C supply temperature, that's 600 kJ of heat energy that you must remove from the duct, which is 570 BTU. At 12,000 BTU/hour, that means it takes 3 minutes of run time to cool the metal off, assuming the A/C has been off long enough that the duct wall heated up to the attic temperature.
Suppose your A/C runs for 1 hour every 2 hours. That means you are adding 3 minutes of run time to each hour. So that's a five percent increase in energy use.
But that scenario is unlikely--on the hottest days when you have that big temperature difference between the attic and inside, you will likely be running closer to continuous. On milder days, the attic won't be that hot, and I'm not sure the duct would get all the way to attic temperature in an hour anyway. And furthermore, I think the MVZ has the capability of modulating to run lower on mild days instead of cycling on and off, at which point you don't have an issue at all.
I agree with Charlie, although I'd consider shorter run-times than 1 hour every 2 hours. On the other hand, note that with shorter off times, insulation of the duct has an increasing effect (reducing the amount of heating up).
Higher thermostat hysteresis has efficiency benefits.
For bathrooms, consider that the door is usually open and AC ducts may not be needed.
At 0.122 BTU/degree-F per lb. the specific heat of steel is almost too low to matter here, and the likely impact on net system efficiency will be "in the noise", of measurement accuracy, given all the other factors (air leakage, actual vs. optimal refrigerant charge, etc.)
Flex duct that isn't stretched perfectly tight will have a larger impact on efficiency than the amount of "abandoned" energy there is in the steel that might be lost when (and only when) the system is operating at a VERY low duty cycle.
Thank you all for your replies. As I wrote my original post, I was working with a contractor to have the pink insulation batts installed (because they talked me into it). But I cancelled that and ordered cellulose blow-in instead, after sealing the attic. Now I plan to bury the ducts deeply, which may move my concerns about thermal mass of the metal ducts even further "into the noise".
Thanks Jon for that comment about the bathrooms not needing ducting. I hadn't thought about that, but may take your advice on the small bathroom. The master is too large and is not well insulated in the ceiling, so it definitely needs some conditioning.
I just finished up the Manual D Speed Sheet and was surprised to see the equivalent flex duct sizes were the same for all but one bedroom supply branch, and no difference for the supply trunks.
The return trunk was adjusted, 10" flex vs 9" metal.