Understanding Manual J Occupant Loads

28 Comments Read/write comments

Occupant loads in Manual J load calculations

People complicate things.  If, say, you want to sell a motorcycle, you place an ad and interested potential buyers come check it out, maybe take it for a little ride, and then one of them buys it.  But every once in a while, one of those potential buyers arrives with a small instability that gets amplified by some kind of cosmic resonance and next thing you know, you're scraping your motorcycle off the driveway and letting the would-be buyer use your shower.  And it just gets weirder from there.

But I'm not here to talk about that kind of complication.  No, this is about a small but important aspect of Manual J heating and cooling load calculations:  the effect of the people in a house on said loads.  I've mentioned in previous articles about how some people inflate the loads by getting this part wrong, but I've never talked about the nuances associated with occupant loads and how to do this part of the load correction correctly.  So, let's tackle that today and see if we can avoid destructive resonances, cosmic or otherwise.

Heating, cooling, or both?

People inside a house add heat to the living space.  If you count this in the winter, the heating load would be smaller than without occupants, meaning you may be able to get by with a smaller heating system.  In summer, people increase the cooling load, requiring more air conditioning.  Before I reveal the official answer from Manual J, let me point out that the peak heating load occurs at night, when people are asleep.  That's when it's coldest outdoors.  (OK, if you have teenagers, they may be playing video games at 3 in the morning, but teenagers are unreliable.  Sometimes they will have snuck out of the house and be playing video games at a friend's house.)

Peak cooling loads occur during the day, when the sun is shining and people are active.  So the answer is that occupant loads (and other internal loads like refrigerators, too) add to the cooling load, but Manual J doesn't let you take credit for them in the heating load.

How many people?

It's a simple rule.  The total number of people in the house, for the purpose of calculating the cooling load, is equal to the number of bedrooms plus one.  It is not the actual number of people who live in the house (unless that happens to be the same as the rule).  Nor is it the number of people you have over at your 4th of July party, the slide show you host after your trip to Portugal, or the party that your teenager has when you're out of town.

The Manual J number of occupants equals the number of bedrooms plus one

The Manual J load calculation should include 4 people for a 3 bedroom house, 5 people for 4 bedrooms, and 6 people for 5 bedrooms.  It could hardly be simpler.

How much load per person?

Since occupant loads affect only the cooling loads, each person in the house has the potential to increase two types of load:  the sensible cooling load (temperature) and the latent cooling load (humidity).  And in fact we humans do affect both.  Here's what Manual J prescribes for the amount of cooling load added by each person:

230 BTU/hr — Sensible

200 BTU/hr — Latent

The sensible component is easy to understand.  It's the amount of heat our bodies give off as we attempt to regulate our internal body temperature.  The latent component is due to breathing and evaporative cooling (sweating), of course, but also 45 of the 200 BTU/hr is from moisture-generating activities like showering, cooking, and water balloon fights.  OK, OK!  Manual J doesn't actually mention water balloon fights so if that's an important part of your indoor life, you'll have to adjust for that manually.

When and where?

Peak cooling loads occur in the early evening, so they should be assigned based on where people are at about 6 pm.  Normally, that's kitchens, dining rooms, living rooms, family rooms, offices, and other common areas where people hang out.  Each house is different, which is why Manual J says the room assignment should be "based on the judgment of the designer."

Room assignment for the people in a Manual J load calculation

In my house, for example, we have 4 bedrooms, so I put the 5 occupants in 4 different rooms.  One is in the basement family room, and the others are shown in the floor plan above:  2 in the den and 1 each in the living and dining rooms.

What about parties?

One of the most common justifications for inflating the number of people in the load calculation is that "these homeowners entertain a lot."  OK.  That's fine.  But guess what?  Manual J discusses those kinds of temporary occupant loads.  The solution is handle such temporary loads with either a supplemental cooling system (e.g., a ductless mini-split) or a system that can move excess capacity from unused parts of the house to the party zone.

To find the amount of temporary cooling load, estimate the number of guests and use these numbers for the sensible and latent components:

230 BTU/hr — Sensible

180 BTU/hr — Latent

The sensible is the same as the regular occupant load, but the latent is lower because your party guests don't usually take showers at your house.  If they do, they probably just crashed your motorcycle so handling that cosmic resonance will be more important than a little excess humidity.  You're trying to prevent your house from shaking itself to bits the way the Tacoma Narrows bridge did in 1940.

People are complicated but simple

The rules for handling the cooling loads from people in a house are straightforward and simple.  Manual J explains it all in just three paragraphs.  (Yet here I am writing the fourteenth paragraph of this article.  Hmmmm.)  Where it gets complicated is when people doing load calculations try to make up their own rules on the fly.  Time travelers, they say, have sensible loads two to three times normal, and even higher if they've just returned from an apocalypse.  Follow their logic and you may find yourself the basement, drinking a beer with a stranger who's wearing a pair of your underwear.

No, no, no.  Just follow the rules above.  Understand how to read Manual J reports.  And above all, watch out for destructive resonances.


Allison Bailes of Atlanta, Georgia, is a speaker, writer, building science consultant, and founder of Energy Vanguard. He is also the author of the Energy Vanguard Blog. You can follow him on Twitter at @EnergyVanguard.


Related Articles

How to Read Manual J Load Calculation Reports

Got Manual J? Don't Assume It's Correct

The 3 Types of Heating and Cooling Loads


NOTE: Comments are moderated. Your comment will not appear below until approved.


Wow Allison, "Cosmic resonance, peeling the motor bike off the drive" ?????? Did your medications expire?

Do you think Manual J people loads are sufficient to account for someone suffering from Hypothalamic Pituitary Adrenal Axis dysfunction or an occupant taking 4000-5000 mg/day of Niacin for Vitamin B3 deficiency?


Danny, I'm not that kind of doctor.  You'll need to ask a medical doctor about conditions like that.

Would it ever make sense to use an additional occupant load to account for pets? I'm thinking 100lb dogs. They rarely take showers, sadly, but I reckon their latent loads from panting are relevant. Call it one extra person per 100lb of dog?


James, that's a great question, and it also gets at a bit of inconsistency in Manual J.  First, I haven't found anything in Manual J about adding load for pets but what you say makes sense. 

Second, the inconsistency is that Manual J is muddying the waters between this calculation being an asset label and an operational label, as those terms are used in energy modeling.  An asset label rates houses with certain parameters being equivalent for more accurate comparison of one house to another.  An operational label attempts to model the house as it's actually lived in.  (Here's an article about that topic.)  The number of occupants in the load calculation is supposed to equal the number of bedrooms plus one.  That's what you would do for an asset label.  But Manual J also has a prescription for adding latent load due to plants.  That's what you would do for an operational label because there's no formula for how many plants you should add to each house.  They also give you some leeway on appliance loads, making that more like an operational label, too.

Ah, makes sense, that's a useful distinction to keep in mind.

200 Btu/hr (sensible or latent) is 200 Btu/hr divided by 12,000 Btu/ton which gives 0.02 tons/person. (note that I am being conservative by rounding up)
So if you have 4 people in your house, you are adding another 0.08 tons of sensible and 0.08 tons of latent. That adds up to 0.16 tons of load--for 4 people. Is that significant when it comes to sizing equipment at design conditions?

@ Roy, as we dial down envelope loads, internal loads (occupants, appliances, lighting) account for an increasing share of the total cooling load. In the context of homes built to today's energy codes, these internal loads only seem small when taken individually. In any case, I think it's important to encourage designers to be a thorough as possible when modeling a home's component loads.


Roy, of course it's important to understand the relative impact of things like this, and David Butler gave the perfect answer.  I'll just add this breakdown of the cooling loads in one small, efficient house we did the design for.  As you can see, internal gains in this house are the second largest contribution to the cooling load at about 30% of the total cooling load.

Manual J internal gains in the cooling load for a small, efficient house

Allison and David: You seem to be taking the common tactic of using fractions as opposed to absolute values when it better fits your narrative. If you are talking about real houses, which to me means those in the range of 2-5 tons total capacity, I would still claim that a couple of tenths of capacity for direct occupant loads is practically irrelevant in terms of sizing, especially when available equipment capacity increments are one-half or one ton.

@Roy, you could say that (re: too small too to be relevant) about pretty much every individual contributor to the design load. So why bother doing a load calc at all? My point was that the more we ratchet down the envelope loads, the bigger occupant loads and other internal gains become relative to the total load.

I'm curious about the source of these internal gains on your pie chart. Title 24 defaults for half a watt per square foot, so even in a 2000 SF house the default would be 1kW which is all of 1/4 ton. I'm working on my living room setup with a big screen TV and a home theatre sound system on, and the UPS powering all this and my Dish receiver says it's drawing 200w. My iMac on a different UPS says 56 watts. The "mini-IDF" in the side bedroom, 9 cameras, a Mac mini, and the router is 67W on its UPS. My smart meter is reading nearly 800w right now, but I think most of that is not an internal load because the hot tub's pump is running presently. I do have the lights off, but they're all LED, so even having several of them on would add 100-150W at most.

So are you showing us a house for which the total cooling load is sub-ton?

Internal loads have also been going down. High efficiency lighting, appliances, and electronics have all helped. A friend of mine who built a super-insulated home says that his heating/cooling balance point is still around 60 F outdoor temperature because of the accompanying internal load reductions. Should we ignore direct occupant loads? No, of course not. But if anyone is not sizing equipment properly according to Manual J, I don't see that being the problem. I would look at glazing first.

@Avery, the largest internal loads are typically in the kitchen and laundry (range, oven(s), refrig, dryer).

BTW, my own home, which I completed last year, has a design cooling load of just under 1 ton @ 100F outdoor / 77F indoor. It's in SE Arizona. I installed a 1-ton Carrier 18VS with four zones. The home has 3,300 conditioned floor area, although part of that consists of a fully-below-grade finished basement, which only has a small cooling load (mostly from egress windows).


Avery, yes, the sensible cooling load for this house was less than 11,000 BTU/hr.  The internal gains added up to 3,290 BTU/hr.  We typically use 1,200 for the kitchen, 900 for the living room, 500 for offices and laundry rooms, plus the standard MJ occupant loads.  So in this case, I think we had 1,200 + 900 + 500 + 3x230 = 3,290 BTU/hr, which is almost a third of the sensible load for the house.

David and Allison: This is interesting. I will admit that I have never done a Manual J load analysis. Could each of you take your super-efficient home case and give an actual breakdown of the design cooling loads (Btu/hr, not percent)? As for laundry (washers and dryers), they do consume a lot of energy, especially if you are all electric, but a lot of that energy goes down the drain or out the dryer exhaust, thus is not all internal load. Does Manual J assume that they occur at peak cooling load times?

@Roy, I uploaded my home's load breakdown here: http://optimalbuilding.com/files/kc_williams_rev4.pdf. As you can see, 'internal' is 2nd largest component of the cooling load, after windows of course. My windows would have been larger but I have engineered overhangs that fully shade all south glazing in summer, plus I modeled 'closed blinds' on all windows that get direct sun, and I only have one west facing window (I put my garage on the west side of the house). Interesting side-note: As of the 8th Edition of Manual J, windows loads are analyzed hourly based on orientation and shading. This means solar gains on east windows don't directly contribute to a room's peak load.

MJ recommends 500 BTUH for a laundry (accounts for residual heat from washer & dryer ops, which is fairly conservative) and 1200 BTUH for a 'standard' kitchen. I used those numbers plus I modeled 4 occupants @ 230, so internal gains = 2620, or 28% of my sensible cooling load. And yes, internal loads are assumed to be present during peak cooling hours.

Thanks David, those are interesting numbers. My previous house in Texas had a west-facing patio and it was miserable. The house before that in Minnesota had a north facing patio and it too was miserable since the sun moves quite a bit north of west in Minnesota late in the day in the summer time. My current house has an east-facing patio which was a requirement when we were house hunting this time. I don't think that anyone can deny that west-facing glazing is a disaster anywhere, so thanks for pointing that out. I am glad that Manual J figured out to use hourly window loads. But back to internal loads. It appears that you do not have any mechanical ventilation. Do you know if your house meets ASHRAE 62.2 minimum ventilation requirements? Also, you appear to have neglected latent loads which is probably OK in Arizona. However, I am curious, do you ever see condensate leaving your indoor coil drain pan?

This is a bit off-topic but I'm sure Allison won't mind :-)

I didn't neglect latent loads... It's shown at the bottom of the report (550 BTU/hr = 800 BTUH for 4 occupants minus 250 BTUH for infiltration). The infiltration is negative because the ASHRAE coincident wet bulb for my location is slightly less than the assumed indoor wet bulb in my model. We typically see condensate during monsoon (July through mid-September).

Regarding ventilation... my house, which tested out at 0.52 ACH50, definitely does not meet ASHRAE 62.2, but that's another conversation!

David: But you don't have any other latent loads like cooking, showers, or laundry?


Roy, the latent loads for cooking, showers, and other moisture-generating activities are built into the 200 BTU/hr latent load that's assigned to each person in the house.  Manual J says 45 of the 200 BTU/hr covers those things.

Allison: I guess that makes sense. I was thinking that 200 Btu/hr was a bit high for direct latent load from an occupant. But 45 Btu/hr/person seems low for showers, cooking, etc. That amounts to 1 pound of water per day per person. Hmmm

Allison wrote: "The solution is (to) handle... temporary loads with either a supplemental cooling system... or a system that can move excess capacity from unused parts of the house to the party zone."

Zone control isn't much help if a big crowd creates cooling loads in winter. One scenario I occasionally encounter is a home theater that can seat a dozen or more people. An appropriately sized mini-split is typically the best strategy when unusual occupancy loads are anticipated.

> "peak heating load occurs at night, when people are asleep... (therefore) Manual J doesn't let you take credit for (internal loads) in the heating load."

For the same reason, MJ also ignores solar gains. And it also ignores the heat stored in a home's thermal mass. These simplifications may have been reasonable in the past, but unaccounted for gains begin to loom large in tight, well-insulated homes. It's not unusual to see MJ to overstate the heat load by 50% or more in beyond-code homes with small envelope and infiltration loads.


David, both of those are excellent points.  We try to steer clients to mini-splits for supplemental entertainment loads, and we always try to steer them away from zoned systems.  On the heating load front, there's another issue, too.  In Aspen, for example, where some houses have all-glass, south-facing walls, not only does the heating load get overstated, but they actually have a cooling load in winter.  A mechanical contractor who works there told me some of those homes use their air conditioner on sunny winter days.

Allison..... do you know where your towel is?


I always know where my towel is, Avery!

I'm always fond of Hitchhikerisms. Perhaps it's partly because my father was a psychtherapist named Marvin. And yes, he actually did have a brain the size of a planet!

Are people located in the home the same for heating as well as cooling loads?


Howard, no, Manual J doesn't include internal gains in the heating loads at all.  I discussed that in the section Heating, cooling, or both?  Since people are typically asleep at night and not using appliances, internal gains don't get factored in.  But see David Butler's comment just above for a good argument about why, at least in some cases, it might be a good idea.

Add new comment