Electrical panel - remove load addition for fuel upgrades

[lixiliu]

There is a non-trivial amount of new loads added for fuel only HVAC upgrades, due to how we determined new loads. In practice, they should not be triggering capacity constraints.

More details to come

[shorowit]

Note there were some improvements in this area not too long ago. See #1517 and #1536.

[lixiliu]

Yeah, that does ring a bell now. It seems like we found this problem twice, thanks!

[lixiliu]

It looks like the PR above introduced the following changes. (fuel := not electric, for readability)

• New heating load = True if preexisting heating is fuel AND new heating is electric (and stand alone and in service); False otherwise
• New cooling load - same logic as new heating load
• New HP load = True if preexisting heating is fuel AND preexisting cooling is fuel AND new HVAC is a HP; False otherwise

New load characterization are used in two ways: load-based method cares if new HVAC load is being added, because it will use different discount factors in that case; meter-based method only adds loads it considers new (and overtime I have come to agree that the average users would only consider something as new if more load is being added than before even if replacing like for like)

I don't think the logic changes are nuanced enough. Consider these cases (and my thought process of what the values should be):

1. Home with HP (no backup) replacing to ER heating, it's electric-to-electric, but we need to account the ER heating as new because of the increased heating load. Similarly, anytime there is an equipment upsizing, we need to count the entire load as new.

• primary heating: pre-upgrade: none, post-upgrade: ER heating; is_new_load: True
• cooling: pre-upgrade: none, post-upgrade: none; is_new_load: False
• HP: pre-upgrade: HP, post-upgrade: none; is new_load: False
• For load-based calc: HVAC load: pre-upgrade: HP, post-upgrade: ER heating; has_new_hvac_load: True
• For meter-based calc: new_load: ER heating (because it is higher than existing HP heating)

2. Home with CAC + ER heating replacing to HP + integrated ER backup (backup and HP can operate simultaneously), HP should be considered new in this case if the total heating load (HP + backup) exceeds pre-existing heating load.

• primary heating: pre-upgrade: ER heating, post-upgrade: none; is_new_load: False
• cooling: pre-upgrade: CAC, post-upgrade: none; is_new_load: False
• HP: pre-upgrade: none, post-upgrade: HP + integrated ER; is new_load: True (Q: it is correct to lump integrated backup with HP here?)
• For load-based calc: HVAC load: pre-upgrade: max(CAC, ER heating), post-upgrade: max(HP, HP + ER heating) = HP + ER heating; has_new_hvac_load: True
• For meter-based calc: new_load: HP + integrated ER (because it is higher than existing ER heating). We should consider the integrated ER as part of a new system even if it is of the same capacity as the old ER heater.

3. Home with CAC + ER heating replacing to HP + existing ER as backup (backup and HP cannot be on simultaneously).

• primary heating: pre-upgrade: ER heating, post-upgrade: none; is_new_load: False
• cooling: pre-upgrade: CAC, post-upgrade: none; is_new_load: False
• HP: pre-upgrade: none, post-upgrade: HP; is_new_load: True *(BUT this depends in either calculation methods...)
• secondary heating: pre-upgrade: none, post-upgrade: ER heating; is_new_load: True (BUT this should be false because ER heating was primary heating before)
• For load-based calc: HVAC load: pre-upgrade: max(CAC, ER heating); post-upgrade: max(max(HP, ER heating), HP) = ER heating. If pre-upgrade = ER heating, then has_new_hvac_load: False since HVAC load did not change. If pre-upgrade = CAC, then also has_new_hvac_load: False since ER heating < CAC. (See example 2 if HP and backup can operate simultaneously)
• For meter-based calc: new_load: 0 (because backup and HP cannot come on simultaneously and HP < backup). (If HP and backup has simultaneous operation, then HP is considered new.)

I am starting to think that we cannot label individual HVAC loads as new in silo.

For the load-based calculation method, I think it makes sense to: tabulate the HVAC load first, and if the post-upgrade HVAC load is higher than pre-upgrade, then set has_new_hvac_load to True when applying demand factors.

For meter-based, I think the same logic as load-based applies (if the post-upgrade HVAC load is higher than pre-upgrade) when deciding whether HVAC has a new load to warrant a load calculation. The 2026 revision clarifies that new load is the net increase in load (in which case, it would just be post-upgrade HVAC minus pre-upgrade HVAC). But for 2023, it's complicated... But we should show that it's more conservative.

For 2023 meter based, if the post-upgrade HVAC load is higher than pre-upgrade, HVAC has a new load. If post-upgrade HVAC is cooling-dominant, sum all cooling components that are considered new (as the total new load to HVAC). If post-upgrade HVAC is heating-dominant, sum all heating components that are considered new.

Thoughts?

[lixiliu]

Another thing to update is the breaker space configuration for HVAC.

Currently we assume each branch can take up to 50A of MCA, but it is not correct.

MOP (Max Overcurrent Protection) treats motor load with a much higher margin of safety than other loads, i.e., = MOP = [2.25 x (Rated Current of Largest Motor)] + (Other Motor Loads) + (All Heater Loads)

This means, we can put more load on a single circuit from electric resistance heating than from a HP compressor before reaching the limit of MOPD (e.g., 50A breaker)

Source: https://www.titus-hvac.com/softwares/MOPulator/pdf/howwecalculate.pdf