220.84 - what defines "heating" and/or "air conditioning"?

[kwired]

What I find most interesting though is that when you have a MFD with, say, 40 or 60 units, you get to the point where the 220.84 calculation allows you to apply a smaller demand factor, resulting in a smaller distribution panel feeder & smaller service feeder (when compared to the 220.82 calculation)

So in essence, I have to add load (or allow for future load to potentially be added by accounting for it in the panel calculations) in order for the service feeder to be smaller. To me, that's reverse logic.

Also, if you flip this over to the mechanical side of things, the mechanical code doesn't allow you to account for cord & plug space heaters in the calculations for heat loading to determine the appropriate size HVAC unit for a given space. So this is a case, to me, where one hand (electrical) is taking something into account in order to be able to take advantage of the optional calculation, whereas the other hand (mechanical) can't allow that. Though I suppose if you just are making an allowance for it...when I have a unit that has a 100A panel but there's only something like 65-75A on it, adding a dedicated circuit for a 1 kW space heater won't affect that at all

Just seems like a "hole" to me....

By having more individual units - you still typically increase necessary service or feeder size, but allowable demand factors are not necessarily same ratio so to speak. A single 10 kVA rated unit will have a demand of 100% especially something that isn't capable of varying load level other then on/off. Now put in 100 of them and allowable demand factors exist because they assume that not all of them will be running at same time, they cycle as a thermostat calls for heat/cool, or possibly they even have variable output capability, and not all likely to run at 100%. Said units can potentially run at 100% load at times but are not likely to do so for very long when that does happen.

So the single 10 kVA unit will need full 10 kVA in the load calculation - but 100 of them that had say an allowable 50% demand factor (totally made up factor here), would still need more VA then the single unit, but only 50 times more instead of 100 times more.

 

[Floyd R Turbo]

I'm not disagreeing with you - I understand the basis behind applying demand factors.

What I'm saying is that the basis, to me, of 220.84 is that in a MFD where the units have actual loads that are heating and cooling, specifically, electric heating and/or air conditioning, I believe the intent of the code here is that there actually be permanently fixed electric heat in the form of baseboard heating or HVAC integrated heat coils or a dedicated compressor/condenser that is powered directly from each unit's individual panelboard.

If you don't have either of those, to me, it's clear that 220.84 should not apply. That's when you have to fall back to 220.82 which does have a section that allows you to take into account the same things as 220.84 does, but does not allow you to apply as hefty of a demand factor once you get to a certain number of units being supplied by the same feeder / distribution panel. But, I think that's appropriate, because the absence of those loads that would be kicking on and off / not running concurrently all the time means that your 'baseline' load level at a peak time is going to be somewhat more consistently higher...I guess it's hard to put into words, "higher" might not be the appropriate term - I guess the potential for it to be higher at times?

Anyways, it's obviously on a case-by-case basis depending on the size of the units and the fixtures in place within them (such as electric water heaters vs centralized hot water circulation, washers and/or dryers, etc) but where the 220.84 calculation really comes into play in a MFD is the service size. If you can't use that section for the optional calculation because of the absence of unit-powered heating or A/C, the 220.82 10 kVA + 40% number is almost always going to be higher than the across-the-board % you can use in 220.84. This could result in a service size calculation difference of 400-600A easily, and I'd be willing to bet that a few years of utility data would show that the extra amperage in such a case was indeed not necessary.

So while I get the rationale behind adding in an allowance for a space heater circuit so that one could use 220.84, to me that is kind of a strange way to approach it. And I'm still really surprised that there are examples out there where this flew with the AHJ, I never would have come up with that as a solution...

 

[jumper]

OK, I had been ignoring this thread, because the title did not interest me. But at Derek’s request, I will chime in.

This very discussion came up about a year ago during a meeting of the “Electric League of the Pacific Northwest's monthly Code Committee.” http://electricleague.net/ Most of the attendees that day agreed that the very small load associated with the FCU is not going to qualify as an “air conditioning” unit under the context of this article. The air conditioning takes place in a common area, and its load goes on a house panel. No individual unit supplies power to the A/C compressor. So no individual unit can say that they have A/C, and if they don’t have heat you can’t use this article.

Then one of the engineers present told us about a solution that he proposed to the City of Seattle’sAHJ. He said he would include in the total load for each individual unit a fictitious amount to cover a non-existent space heater. He proposed that that could allow him to use the optional method. The city agreed. I have since used this same concept for a couple projects, and the AHJ allowed it. It has been a while, and I don’t recall the amount of load we assigned to this non-existent load. I think it was on the order of 1000 watts.

Jeez Charlie, when I said I wanted your opinion I thought you were gonna support me, not throw under me under the buss and explain to everyone how I was wrong.......jk..

Okay, I remembered the story incorrectly. Oops.:ashamed1:

I think the solution you all came up with was well done IMO. Common sense prevailed over blind obedience to the code requirements.

I prolly would have went with 1500 watts just 'cause it is the same as SABC and laundry.

 

[Mgraw]

This is another example where code has not keep up with technology. I would go to the AHJ and explain the situation. There are AHU's, 2 ton for example, that only draw 120 watts. There are two pipe and three pipe systems that can supply heat to one unit and cool to another. There is no electric heat because newer units can provide heat in minus 20 degree or lower range. These units have been around for several years now yet nothing in code addresses this.

 

[Floyd R Turbo]

I think the solution you all came up with was well done IMO. Common sense prevailed over blind obedience to the code requirements.

I have to agree here that the reality is that the demand factors in 220.84 really should be able to apply - and that's probably why the AHJ allowed it. Good to know, it's a discussion that I'll likely have with one in the future.

This is another example where code has not keep up with technology. I would go to the AHJ and explain the situation. There are AHU's, 2 ton for example, that only draw 120 watts. There are two pipe and three pipe systems that can supply heat to one unit and cool to another. There is no electric heat because newer units can provide heat in minus 20 degree or lower range. These units have been around for several years now yet nothing in code addresses this.

The counter argument here, of course, is that the presence of electric heat or higher-draw A/C units allows for a higher diversity factor to be applied since the overall connected load would be higher with these units.

Remove those devices, and you remove the load, so you can't continue to apply the diversity when the load that allows is (creates the diversity) is not there.

I would like to know about the 2 ton A/C unit that only draws 120W, the last job I designed that had 1 ton units, these were each on the order of 1400W electrical power draw each. In that example, I could only put 8 units on a 208/3 225A distribution panel. With that same example, going to a 3-pipe centralized system, I'm at the same 8 units per distribution panel even though there is ~40A less connected load. That's really not where it makes a huge difference though, it's the service where it might start to make a difference.

 

[Mgraw]

I have to agree here that the reality is that the demand factors in 220.84 really should be able to apply - and that's probably why the AHJ allowed it. Good to know, it's a discussion that I'll likely have with one in the future.

The counter argument here, of course, is that the presence of electric heat or higher-draw A/C units allows for a higher diversity factor to be applied since the overall connected load would be higher with these units.

Remove those devices, and you remove the load, so you can't continue to apply the diversity when the load that allows is (creates the diversity) is not there.

I would like to know about the 2 ton A/C unit that only draws 120W, the last job I designed that had 1 ton units, these were each on the order of 1400W electrical power draw each. In that example, I could only put 8 units on a 208/3 225A distribution panel. With that same example, going to a 3-pipe centralized system, I'm at the same 8 units per distribution panel even though there is ~40A less connected load. That's really not where it makes a huge difference though, it's the service where it might start to make a difference.

Look at Mitsubishi multicity line. I have put 8 2 ton AHUs on a 2 pole 20 amp breaker.

 

[kwired]

I have to agree here that the reality is that the demand factors in 220.84 really should be able to apply - and that's probably why the AHJ allowed it. Good to know, it's a discussion that I'll likely have with one in the future.

The counter argument here, of course, is that the presence of electric heat or higher-draw A/C units allows for a higher diversity factor to be applied since the overall connected load would be higher with these units.

Remove those devices, and you remove the load, so you can't continue to apply the diversity when the load that allows is (creates the diversity) is not there.

I would like to know about the 2 ton A/C unit that only draws 120W, the last job I designed that had 1 ton units, these were each on the order of 1400W electrical power draw each. In that example, I could only put 8 units on a 208/3 225A distribution panel. With that same example, going to a 3-pipe centralized system, I'm at the same 8 units per distribution panel even though there is ~40A less connected load. That's really not where it makes a huge difference though, it's the service where it might start to make a difference.

I think he was talking about the air handler only drawing 120 W. Don't recall the BTU capacity of most the smaller units in the school project I did, but they were all under 100 watts. Guessing most were at least one ton units.

 

[kwired]

I'm not disagreeing with you - I understand the basis behind applying demand factors.

What I'm saying is that the basis, to me, of 220.84 is that in a MFD where the units have actual loads that are heating and cooling, specifically, electric heating and/or air conditioning, I believe the intent of the code here is that there actually be permanently fixed electric heat in the form of baseboard heating or HVAC integrated heat coils or a dedicated compressor/condenser that is powered directly from each unit's individual panelboard.

If you don't have either of those, to me, it's clear that 220.84 should not apply. That's when you have to fall back to 220.82 which does have a section that allows you to take into account the same things as 220.84 does, but does not allow you to apply as hefty of a demand factor once you get to a certain number of units being supplied by the same feeder / distribution panel. But, I think that's appropriate, because the absence of those loads that would be kicking on and off / not running concurrently all the time means that your 'baseline' load level at a peak time is going to be somewhat more consistently higher...I guess it's hard to put into words, "higher" might not be the appropriate term - I guess the potential for it to be higher at times?

Anyways, it's obviously on a case-by-case basis depending on the size of the units and the fixtures in place within them (such as electric water heaters vs centralized hot water circulation, washers and/or dryers, etc) but where the 220.84 calculation really comes into play in a MFD is the service size. If you can't use that section for the optional calculation because of the absence of unit-powered heating or A/C, the 220.82 10 kVA + 40% number is almost always going to be higher than the across-the-board % you can use in 220.84. This could result in a service size calculation difference of 400-600A easily, and I'd be willing to bet that a few years of utility data would show that the extra amperage in such a case was indeed not necessary.

So while I get the rationale behind adding in an allowance for a space heater circuit so that one could use 220.84, to me that is kind of a strange way to approach it. And I'm still really surprised that there are examples out there where this flew with the AHJ, I never would have come up with that as a solution...

Just a quick look so maybe I missing something, but 220.82 only applies to a service or feeder to a single dwelling unit doesn't it?

If you have a single central compressor unit, that is the only part of such system that art 440 applies to as the AHU's don't have a hermatic compressor in them, then you have only one "air conditioning load", and with no electric heat it is also the heating load. May depend on what part of country you live in but here, those we installed for my school project draw more when it is cold outside - that is when they are running at higher speed to soak up as much heat as possible to bring inside. Only time there is a high draw during cooling season would be when you initially start them up when the interior of building hasn't been cooled in a long time, but once you stabilize inside temp they slow down the compressor(s).

One may be able to apply demand factors to air handler units - but they are not much of a load in the first place.

My school project one 20 amp 208 single phase circuit could supply a pretty large area of the building's air handlers.

 

[Strathead]

I am not offering a solution, but some observations...

The intent of 220.84 is to size the service or feeder feeding a common point that feeds several units. As such, an AHJ could just nix the ability to use this table. If he/she was a reasonable sort, then I would think first off, the rooftop heat pumps would have to be on the same feeder as the units. If the units fed from one distribution panel and the rooftop units fed from a different panel, then 220.84 could be used only for the feeders where both of those loads originate from. Secondly, I would want to know just how much those rooftop units vary in amperage per the demand of the units. My guess would be that they would already be sized down for multiple unit demand so the intent of 220.84 would not be applicable. Unless it could be shown that the multi unit HVAC has a similar demand curve to that of individual units, I would expect the AHJ not to allow that table. I am curious what others have to say about this observation.

 

[MyCleveland]

Strathead

Agree with your statements....
How can anyone manipulate the language of 220.84 by using a 200w BLOWER motor as an AC load.

This is why the standard calc exists.

 

[kwired]

I am not offering a solution, but some observations...

The intent of 220.84 is to size the service or feeder feeding a common point that feeds several units. As such, an AHJ could just nix the ability to use this table. If he/she was a reasonable sort, then I would think first off, the rooftop heat pumps would have to be on the same feeder as the units. If the units fed from one distribution panel and the rooftop units fed from a different panel, then 220.84 could be used only for the feeders where both of those loads originate from. Secondly, I would want to know just how much those rooftop units vary in amperage per the demand of the units. My guess would be that they would already be sized down for multiple unit demand so the intent of 220.84 would not be applicable. Unless it could be shown that the multi unit HVAC has a similar demand curve to that of individual units, I would expect the AHJ not to allow that table. I am curious what others have to say about this observation.

Part of the whole point of using the VRF is high energy efficiency. It will use less energy as a general rule then multiple smaller units will use. If you were to do a load calculation for individual compressors at each tenant you will likely have a higher overall load calculation - this before factoring in any electric heating that may be used with either system.

The one I worked on had three compressor units per individual system. If you go there on a fairly moderate day, you may only have one compressor of the system running, and it will be at a low speed. Go there when it is 95 degrees outside and they may all be running, but they are also changing speed pretty frequently. Go there when it is zero deg F outside and they are all running at fairly high speed nearly all the time.

 

[Strathead]

Part of the whole point of using the VRF is high energy efficiency. It will use less energy as a general rule then multiple smaller units will use. If you were to do a load calculation for individual compressors at each tenant you will likely have a higher overall load calculation - this before factoring in any electric heating that may be used with either system.

The one I worked on had three compressor units per individual system. If you go there on a fairly moderate day, you may only have one compressor of the system running, and it will be at a low speed. Go there when it is 95 degrees outside and they may all be running, but they are also changing speed pretty frequently. Go there when it is zero deg F outside and they are all running at fairly high speed nearly all the time.

I can't tell if you are consenting or dissenting. I feel that your statement confirms what I am saying. The VRF variance basically takes the place of the electrical usage variance thereby defeating the code section's purpose

 

[jumper]

The VRF variance basically takes the place of the electrical usage variance thereby defeating the code section's purpose

I think I may almost but not quite understand what you just wrote.

 

[Strathead]

I think I may almost but not quite understand what you just wrote.

Let me try again:

The code section's purpose is to allow a smaller feeder because multiple dwelling units won't all use power at the same time.
The code section requires electric heat or cooling for each space.
As I stated before this could only be applicable to the feeder to feeds both the dwelling units and the VRF, if was allowed to be applied at all.
The intent of a VRF is to allow less heating or cooling energy because multiple dwelling units won't all use energy at the same time.

 

[jumper]

Let me try again:

The code section's purpose is to allow a smaller feeder because multiple dwelling units won't all use power at the same time.
The code section requires electric heat or cooling for each space.
As I stated before this could only be applicable to the feeder to feeds both the dwelling units and the VRF, if was allowed to be applied at all.
The intent of a VRF is to allow less heating or cooling energy because multiple dwelling units won't all use energy at the same time.

Way more clear.:thumbsup:

 

[kwired]

Let me try again:

The code section's purpose is to allow a smaller feeder because multiple dwelling units won't all use power at the same time.
The code section requires electric heat or cooling for each space.
As I stated before this could only be applicable to the feeder to feeds both the dwelling units and the VRF, if was allowed to be applied at all.
The intent of a VRF is to allow less heating or cooling energy because multiple dwelling units won't all use energy at the same time.

I understand your logic, but don't think NEC sees it entirely the same way. First if you have a smaller system especially if it only has one compressor I think you have to assume that single compressor will be heavier loaded more often then a system with multiple compressors.

slightly different but similar - I get service calls for central electric heating (typical is a 20 kW air handler heating whole house). I will get call that it doesn't seem to keep up when it gets really cold outside. Come to find only 5 kW of heat is still functioning - had been keeping the place warm until it got really cold outside and it just can't quite keep up anymore.

The VRF system doesn't run at maximum load too often, but you still must be able to deliver that demand when it does.

 

[Floyd R Turbo]

Thanks everyone for the replies/discussion, it's certainly an interesting topic.

Just a quick look so maybe I missing something, but 220.82 only applies to a service or feeder to a single dwelling unit doesn't it?

Actually you are correct here

The intent of 220.84 is to size the service or feeder feeding a common point that feeds several units. As such, an AHJ could just nix the ability to use this table.

Correct as well

I was incorrect in my thinking that if 220.84 did not apply, one could fall back to 220.82. You have to fall back to Part III, which mean more pen-to-paper but likely a similar result I'm guessing. Good catch.

How can anyone manipulate the language of 220.84 by using a 200w BLOWER motor as an AC load.

This is why the standard calc exists.

This is kind of my thinking...1) there is no high-load AC compressor or electric heat either directly connected to each unit panel or at least connected at each distribution panel (still a compressor dedicated for each unit) so 220.84 wouldn't apply.

Part of the whole point of using the VRF is high energy efficiency. It will use less energy as a general rule then multiple smaller units will use. If you were to do a load calculation for individual compressors at each tenant you will likely have a higher overall load calculation - this before factoring in any electric heating that may be used with either system

Agreed, and kind of the point of not being able to use 220.84. With a system with a compressor and electric heat coil for heating/cooling, you're going to A) have a higher connected load and B) have loads cycling on and off, giving you the ability to apply diversity. When you move to a higher efficiency system that essentially strips all that out from any distribution scenario and puts all the load on the MDP or a house panel, you don't have any (big) loads to diversify (besides electric ranges and dryers). But 220.84 says you need ranges and AC or heat. Take one away, and you can't use it.

So, now it looks like I'm back to Section III. Unless I can get the AHJ to buy off on the 1500 VA heater circuit for the heater that will never get installed idea. In reality, I'm not sure it will make a huge difference

 

[MyCleveland]

" I'm not sure it will make a huge difference "

It will...could a 1500W heater really heat the space ?

I would have the ME/MC provide me a MODEL AC load for each apartment size.
Use these numbers to mimic an AC load, similar language exists for adding in an electric range/oven.

 

[Floyd R Turbo]

I don't think that mimicking a load for purposes of the electrical calculation would be proper, I can get the heat load from an ME no problem, but that doesn't mean I can apply said heat load to the effective electrical equivalent within the space when there is no device (load) there providing that heating/cooling. Or did I misunderstand what you were referring to?

 

[MyCleveland]

Take a look at 220.84 A (2) exception
When have apartments without electric cooking, you allowed to ADD in a model load that will then give you access to derating table 220.84.

I have used this in past when developer selected GAS ranges for apartments.

I would consider modeling AC or TOTAL electric heat load in same manner.
This I would be comfortable approaching a AHJ with for discussion.

If I were the AHJ, I would not consider a 200W blower motor or 1500watts of electric heat as meeting the minimum for access to table 220.84.
Unless ME/MC tells me 1500watts will heat the ENTIRE space.