Speaking with Brian Curtsinger, Conservation Representative for Pedernales Electric Coop, he suggested a novel, rather counter-intuitive approach to zoning our house.

You can see a floor plan at the project file here:
http://www.greenbuildingtalk.com/ProjectAlbum/tabid/62/action/viewproject/projectid/1139134/Default.aspx

It's a shoebox shape, with long axis running east-west. The "night time" private spaces are at the back/top/north side of the box. There is a middle service corridor and galleria. The "day time" public spaces are at the front/bottom/south side. The machine room is near the middle of the service corridor that runs east-west through the house. We will run one (or more?) trunk lines though the service corridor, with short ribs off that spine to each room. Should make for quite efficient distribution I hope.

Before talking about the zoning idea, I want to ask first if zoning is even that important in a house with a really nice envelope and a geothermal system? Walls of ICF, roof/ceiling of 10" SIPs, low-E glass. Yes, there are a few areas that could normally be closed: Guest bedroom, Guest bathroom, Gym (note, all on the east side). I suppose we could close the master suite at night, but we like to let the animals roam. So, I wonder whether zoning in this situation will give us significant benefits in performance or energy saving?

Now Brian's idea was to zone the house east and west, instead of what others propose: north (night) side versus south ("day") side. Brian was curious whether the differential of rising versus setting sun would be more significant than north/south. Consider that even on our 100+ days, it's not all that unpleasant outside until about noon and thereafter, when the sun has burned off all the nighttime coolness.

Furthermore, east-west zoning would seem to fit with the space we won't use as much. AND, east-west would mean running a single trunk line in each direction. This would leave us space to duct the return air, which I think would be desirable.

And what are your thoughts? Thanks much.  --David

Regardless of how you're going to zone the residence,

With these key assumptions:

a.) heating/cooling the house with an Envision 4 ton 2-stage variable-speed unit

b.) using the IntelliZone zoning panel.

c.) duct work is designed well enough to ensure an ESP of .50" or less

d.) you don't want to use a by-pass duct or dump zone

Keep in mind a few key items:

1.) Each zone needs to be able to accommodate about 900 CFM. While it might be a little loud, you can take the FPS up to about 900. This will help deliver adequate speed when multiple zones are open.

2.) Your building heat/cooling load will dictate how many SF you have to dedicate to 900 CFM. A tight building envelope and high insulation will result in requiring more SF for each zone, which may end up being counter to what you want to accomplish from a zoning point of view. I.e., you probably won't be able to zone down to one bedroom.

3.) With all zones open you're going to have to be careful to maintain at least 600 FPS velocity.

A design alternative would be to seal and insulate the home so well that you can drop down to a 3 ton unit. Now your minimum CFM drops to 750.

Hope this helps.

Best regards,

Bill

I don't quite understand all that, Bill, but I like the way you concluded -- "...drop down to a 3 ton unit." I would like to achieve a 1,000 SF per ton ratio.

I did a precursor whole house Manual J using HVAC-Calc from http://www.hvaccomputer.com/. I'm quite impressed with how easy it is to use. I came up with Heat Gain: 24,865 BTU, and Heat Loss: 39,866 BTU. This weekend I want to do a careful room by room calculation using take-offs from the drawing files.

I am curious, and a little concerned about the difference between the numbers, in that I am sensitive to cold. I understand that part of the difference is made up by sensible loads (people, refrigerator, freezer, computers, etc.) in the building. I would love radiant floor heating but don't want to deal with the added cost and complexity.

I suggested slab edge insulation, and my architect says we don't do that in this climate. Too little benefit and too hard to make look good. Do I have any need for concern if the system is sized for the cooling load?
--David

P.S. Myself, I don't really care about zoning. I'd just as soon have the simplicity of a single zone. But I wanted to ask this here to get a feeling for whether I would be missing important energy savings by not zoning.

East West zoning makes sense as it accommodates heavy PM western sun while not over cooling rooms on east side. Its a comfort issue. What energy savings you achieved might never pay off the increased cost of a zoned system.

I used HVAC-Calc and then Elite Soft for my house. Elite is spiffier, much more expensive, but agreed well with HVAC-Calc

Heat Loss is likely higher than heat gain since the differential is so much more in winter (45 or so degrees across envelope in winter vs 25 in summer)

If you are confident of the load calc numbers I think an Envision 038 would be a good fit. You'll need the smallest option in heat strips to make up a few thousand extra Btuh on a design day, but they likely won't run often.

An 049 would provide the heat on geo alone but be substantially oversized for cooling. You don't want to be oversized for cooling in a humid climate. I'm somewhat surprised your cooling load is as low as that given Austin's 98 degree design temperature. What did you use for indoor design temps for summer and winter?

Need to be certain of the load numbers as well as have loop water design temperatures to precisely select equipment model.

Engineer, thanks much for that advice. I have heard that zoning should be thought of for system balancing and not so much for energy savings, and that seems to be what you're saying (and to go against a lot of conventional wisdom).

I used 77F for cooling indoor (instead of HVAC-Calc's default of 72), but I bumped up the outside design temperature to 102F (instead of 98). I'm going to do a careful room by room calc using take-offs from the DWG files, so that should help prove out the numbers.