Supply ventilation – energy use impact
I downloaded BEOpt and EnergyPlus and have been experimenting. Some of the outputs are surprising at first but make sense when I think more deeply, However, I have not been able to make sense of the results I am getting for different ventilation options. I am trying to determine if there is some physics advantage to supply ventilation or whether the results I am getting are modeling errors.
I have four cases: no ventilation, exhaust, supply, and 70% efficient HRV. The later three all use the same ventilation rate in cfm. The software reports the source energy used per year for various categories. The source energy for heating comes out:
none:40.9
exhaust: 52.12
supply: 47.79
hrv: 44.79.
The relative values make sense for none, exhaust, and hrv. But why wouldn’t supply be about the same as exhaust? In each case, you are bringing in the same quantity of unconditioned air. The fan vent energy is the same for exhaust and supply. (It goes up dramatically for HRV, making the total energy for HRV higher than for supply.) The HVAC fan energy is close, with exhaust being slightly higher than the other three. I would have expected supply to be higher in that category if the software is modeling a CFIS system.
The software automatically determined the following heating capacities:
none: 29.53
exhaust: 30.99
supply: 27.73
hrv: 30.74.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Reid,
The most common type of supply ventilation system is a central-fan-integrated supply ventilation system. (To learn more about these systems, see Designing a Good Ventilation System.)
I don't know if the BeOpt assumptions are based on energy monitoring data from actual homes, or whether BeOpt is based on calculations. BeOpt may be making good assumptions, or its assumptions could be flawed.
That said, here is one possible explanation for the numbers you cite: For much of the year, a central-fan-integrated supply ventilation system operates without any additional fan energy. It uses a fan that is already required for other purposes (space heating and cooling). In contrast, an exhaust ventilation system cannot operate without fan energy.
BEOpt is reporting vent fan energy, HVAC fan energy, and heating energy separately. If it was modeling CFIS, I would expect vent fan energy to be much lower than exhaust only, HVAC fan energy to be much higher, and heating energy to be the same. It is reporting equal vent fan energy, slightly lower HVAC fan energy, and much lower heating energy. That leads me to believe that it is modeling some other configuration of supply ventilation than CFIS. I cannot envision a configuration that would produce those results. I am inclined to attribute this to some bad data somewhere in the software. However, since the results are favorable, if there is a real system that would produce those results, I would be interested in it.
I found a conversation on the BEOpt forum on what appears to be the same issue. The response there indicated that this is due to an interaction with duct leakage. I will try to get more clarification in that forum and report back here if the answer would be interesting to this readership.
After some more detailed analysis, I was able to answer the main part of my own question. The difference in heating energy between supply ventilation and exhaust ventilation is due an interaction with infiltration rate and spot ventilation like bathroom fans and range hoods. When no spot ventilation is occurring, supply ventilation and exhaust ventilation both increase infiltration + mechanical ventilation by the same amount (although by less than the mechanical ventilation rate). When spot ventilation comes on, however, that further increases the infiltration + mechanical ventilation for exhaust type but decreases it for supply type. The software is counting the spot ventilation as part of the mechanical ventilation and reporting the net flow. Thus, when spot ventilation is occurring, supply ventilation comes out more efficient than exhaust ventilation.
In each case, the model assumes that the whole house ventilation system operates continuously at the ASHRAE rate. Under that assumption, supply is superior to exhaust. The conclusion might be different once one considers how each would actually be implemented. For example, if an exhaust system is implemented using bathroom fans, it would be simple for the control to factor in at least the bathroom fan portion of spot ventilation. Also, a system that operates intermittently at a higher flow rate, such as CFIS, might not show the same behavior.