Designing airflow in large low load homes


I am looking for some feedback: I am working on a 3850 sq ft house in So. Cal., Castaic to be exact. It will be well insulated and air sealed (blown in blanket [fiberglass] in 2x8 with an R-4 continuous on the exterior. It has a 44' long x 22' high x 23' wide Kitchen-Great room with a cathedral ceiling. Manual J calls for only 340 CFM total, but how to throw that all the way across the 44 ft long room to achieve adequate mixing in such a cavernous space? Even a tiny 12x4 (Truaire) only has a throw of about 30 ft and that is with a NC of 40 which is starting to make audible noise. Thoughts suggestions? 

You need to be a member of Building Performance Community to add comments!

Join Building Performance Community

Email me when people reply –


  • Thats very low CFM?  If cathedral ceiling and no bodys above the 8" mark just put returns there and let the peak get hot.    If you have a walk way near the peak I tend to put return very high.    With very dry areas I use higher CFM/ton 500/600CFM ton  In wet not super hot I will go as low as 350 CFM  Mostly 400 CFM.   Its a cold coil that takes water out of air. If temp can get below 40' put a 29' cut off in "A"coil weave in so coil will not freese.   Is South Cal dry? Just bump up air flow for more throw.    Most of time .4 CFM/SF is as low as I go = 1540 CFM, part of each hour.   

  • Air distribution 'by the book' is always challenge in load load homes, especially in milder climates as is the case here. You do what you can (see Wes Davis' post re: Manual LLH, diffuser selection, etc.) but in the end, don't sweat the shortfall. The more efficient the enclosure, the less important the throws become. In a normal size room, you just need to get the air into the room. In a long space such as you describe, it's best to locate difussers on both ends.

    As Eric noted, ceiling fans can help mix the air in the upper part of a volume ceiling. But I would argue that it's more efficient to allow the air to stratify, thus reducing the ceiling-roof delta-T.

    You wrote: "Manual J calls for only 340 CFM total"...
    Not sure if you meant total for Kitchen-Great Room, or total for the house?  In any case, MJ doesn't determine the design airflow. The software needs to be told the design airflow (based on a proper Manual S calculation). If the designer doesn't enter the design airflow, MJ software will make its own calculation, which I would argue is incorrect. Here's why:

    Total design airflow depends on the system's capacity and the desired sensible heat ratio, not the load. For example, if the cooling load is, say, 1.1 tons and you install a 2-ton condenser, then the design airflow must be based on 2 tons, not 1.1 tons.  In a dry climate, the designer may decide to operate at 475 cfm per ton to achieve a high sensible heat ratio, so the design airflow in this case would be 950 CFM. Once the design CFM is entered into the MJ software, it will determine the design airflow for each room based on room loads relative to total load.

    • Thank you David for your reply. The specific 340 CFM came from Wrightsoft based on my entering actual equipment airflow capacity (Mitsubishi's PEAD/PEFY units on high) and the software's calculation as to how it parcels the air to the various rooms in the zone based on each room's needs. I have realized/learned the default airflows are typically an overstatement.  Your reponse though begs a related question: if Manual J specifies a 2 ton, for example, but a machine only has a max air flow capapcity at 0.5 of 800 CFM, how do you specify if you want 470 or 500 CFM/ton? You have to mix-match the forced air unit? But many manufactuerers don't allow/advise doing so, especially Minisplit manufactuerers. I can't pair a 2.5-t air handler to a 2-ton condenser (I keep asking).    

      • Dav wrote: "Your reponse... begs a question: If Manual J specifies a 2 ton, for example, but a machine only has a max air flow capapcity at 0.5 of 800 CFM, how do you specify if you want 470 or 500 CFM/ton?"

        Don't let the equipment dictate your design! It's supposed to be the other way around.

        With conventional splits (US brands), finding 'unbalanced' matchups (e.g., 2-1/2 ton AHU with 2 ton outdoor unit) is rarely a problem. As you noted, mini-splits are much more limited in terms of match-up flexibility. BTW, MJ doesn't dictate tonnage. One of the Wrightsoft MJ reports shows 'tons' but that's not intended for equipment sizing. It's based on nominal tons and whatever sensible ratio you may have entered. Manual S deals with equipment selection. You must select a system that can satisfies your your sensible (and latent) loads at your design conditions and airflow.

        But I'm still not clear if 340 is for the entire house (you specifically referred to '340 CFM total'), or that's just for the Kitchen-Great Room? If you meant the latter, I don't see that as a problem (although I hesitate to say that without seeing the plans).  BTW, the projects I work on routinely have an airflow-to-area ratio on the order of 0.25 cfm/ft2 or lower. The ratio for my own home (SE Arizona) worked out to 0.15 cfm/ft2.

        • Can you post a sketch of that area that shows where you intend to locate the supplies? Also, it sounds like the supplies are in the (cathedral) ceiling. Is it a scissors truss with encapsulated attic above?

        • Sorry: 340 CFM is for that area/room only. It comes out to 0.3 CFM/sqft., less than Eric's but higher than you. What I considered such a low flow made me nervous that the room will will not maintain a uniform temperature from the glass door (wall) to the interior wall.    

  • For cooling, we would recommend more air for that square footage. The Manual J value is good for the heat loss or cooling load of the space, but the actual distribution of the air is another issue.  This will make your heating system appear to be "oversized" assuming your whole house needs 1700 CFM of conditioning air. We would design a cooling system for 1.5 cfm per squrefoot. There are 2 options if the actual load doesn't need that much conditioned air. First would be a low delta T. This means if we would design to heat with 115F, you could reduce the temp to 95F by nominally doubling the air flow. This would increase your room mixing, furnace run times and can benefit more from variable speed equipment during partial load conditions. 

    Other options would be a seperate mixing fan. Could be as simple as a ceiling fan. You could also mimic an induction unit or a fan powered VAV. This would be a separate fan that would respond to the furnace or a local thermostat to create a flow of about 1000 to 2000 CFM to mix the room generally when the space becomes stratified or when the furnace is heating the space. 

    Little things I would do would be make sure you have supply and return registers are at very different elevations. Also, multiple registers are probably in order. If it isn't too crazy I would have seperate systems for heating and cooling, but I am crazy.

  • Dav,

    First, Congratulations on starting the job right! Know the load, the equipment capacity, the design airflow… and room air distribution (selecting grilles and registers)!

    Second, ACCA wrote the book on the solution(s) in Manual LLH. There are six different strategies described that address air mixing. 

    Keep up the good work.

    PS. Depending on the architectural details of the vaulted ceiling and the register’s location the Coanda (surface) effect may be of assistance, see Manual T Section 2-8.

  • You need to look at better grills.   You only need to throw 80% of the way across a room at 450 fpm to 600 fpm to get mixing of the room.  Look at the Commercial Shoemaker grills if you are doing a ceiling grill look at the CB10A if you are doing high sidewall look at the 901 or 903 without any integrated dampers.  Consider throwing the short distance - the 23' width.  A 12x12 CB10A will get you almost 25' of throw at about 500 fpm, with a noise criteria <20.   If the house is really going to be a good house it is much easier to mix a room so some of your worries should be alleviated, but go the short distance if you are still worried.  The beauty of a square grill with one direction of throw is that you can rotate it 90 degrees with no problem and see if it works better with in that direction.  The 901 high sidewall is 20' of throw at 500 fpm with a <20 NC.  If you place the grill within 1' of the ceiling you should manage to engage the conanda effect which gives you a boost of about 40% of the rated throw.  There are also linear grills which may have even better throw characteristics.  Look at better grills - Shoemaker commercial.

    • I actually was using those grilles for my calculations and they came up about 10' short b/c I wanted to get 100% across the room to address the heat gain on the south facing glass wall (door). I think I will just split the load and have one unit on one side of the room for the glass and other vent from another unit at the oposiste end to "sandwich".

      Thank you for your input.  

This reply was deleted.