Hank Rutkowski on sizing HVAC equipment

[The following is based on an interview with Hank Rutkowski, the mechanical engineer who wrote Manual J-the document, which describes how to size heating equipment. The interview gets down into details aimed normally at the HVAC contractor. However, based on a question from the SIPTalk section, the interview is being included in its entirety for those who have to deal with equipment over-sizing problems on a regular basis in energy-efficient SIP homes. Note: Manual D is the document used to size ductwork-another item of some contention. And the items in bold are those of the interviewee, not the interviewer.]

"We could cut residential heating and cooling equipment size in this country by 30% and as much as 50% if we just did the following," says engineer Hank Rutkowski. "First, the contractor should perform an aggressive Manual J calculation-taking credit for every load-reducing feature or detail he can, including drapes and blinds. Second, he should use manufacturer's application data to select equipment that has the appropriate capacity for the conditions that will exist at the site; the ARI rating is useless for this work. Third, he should use the room load estimates, the equipment manufacturer's blower data and the Manual D procedures to find room the CFM values and then use published performance data to select the appropriate supply air outlet (type and size) for each room; one size outlet cannot fit all rooms. (There also must be a low resistance return path for every room that has a supply outlet; door undercuts are not acceptible.) Forth, he should use the Manual D procedures to size all the duct runs and he should install a system that complies with ASHRAE standards; completely seal all runs located in an unconditioned space and insulate these runs to at least R-6 and preferably R-8. Fifth; he should certify the performance of the system after it has been installed - measure all air flows, balance the air distribution system and use certified protocols to check the refrigerant charge.

If you did all that, and then you installed equipment sized according Manual J loads (with no safety factor) and Manual S selection procedures, comfort would go up, humidity control would improve, air movement in the comfort zone would be better, installation costs could be $500 to $1,500 lower, utility demand loads would be cut by at least a third or even a half, energy loads would be significantly reduced and the return on the time and effort invested on this work would be enormous - for individual families, the community and the nation."

Rutkowski authored "Manual J, Load Calculation for Residential Winter and Summer Air Conditioning." The 7th edition and the other design manuals are published by the Air Conditioning Contractors of America. Unfortunately, most HVAC contractors don't use these procedures to size equipment and design duct systems. Rutkowski estimates that only 10% of heating and cooling equipment sizing decisions are based on some type of Manual J calculation and that less than 1% of the jobs are based on an aggressive implementation of the recommended design procedures. Rules of thumb like "install 1 ton of AC per 500 ft2 of floor area, figure ariflow at 400 CFM per Ton and use 0.10 on the duct slide rule to size duct runs" are ridiculous; but they are commonly used to design systems that feature oversized equipment, oversized supply outlets and duct runs that are too small, too leaky and inadequately insulated.

Proctor Engineering and others have criticized Manual J as being a calculation tool that results in significantly oversized equipment. Rutkowski isn't phased.

"We did a study of four different tools for estimating loads, and Manual J was one of them. We performed over 800 different computer runs for eight different climates. Using the same details as inputs, we found that there was generally less than a 3% difference in the heating and cooling load predicted by the various procedures. The truth is that there has been no significant effort to validate or improve the accuracy of the residential load estimating procedures for the last 45 years. "

Never the less, Rutkowski claims that Manual J gives appropriate answers if you use an "aggressive" set of assumptions. But, he also knows that HVAC contractors tend to fudge input data to make the calculations match their favorite rules of thumb. In his words, "Manual J is a good tool if you use it aggressively. Follow the rules and it will give you a reasonable margin of safety. But the average contractor says, 'Better safe than sorry,' so he fudges here and there. He adds a 5 to 10 degrees to the summer and winter design temperatures, calls the building 'average' instead of 'tight' and doesn't take credit for shading by interior blinds and drapes. Then when he finally comes up with a load number for sizing, he throws in an extra half-ton of AC just for the heck of it." No wonder equipment is so oversized!

"I'm always telling contractors what I did in my home here in Ohio. It's 2,400 ft2 with 600 ft2 of windows, so they figure it takes 4 tons or so to cool it. But I have a 2-ton unit that moves 1,000 cfm of air; in fact, the 40,000 Btuh furnace is a little oversized because I needed a littler larger blower to handle the air flow through the cooling coil. Even when it's 95 degrees and sunny outside with very high humidity, my cooling equipment still cycles on and off. It's very comfortable for me at 75 F and less than 50% RH."

Rutkowski admits the Manual J could be improved. But he kept using the word "aggressive" in describing how contractors should use the tool to assure they get good answers. He reports a gradual increase in the number of code jurisdictions requiring Manual J calculations, yet he says that is a half-step. "If contractors don't want to size according to an honest J-calculation, they simply change some of the inputs to make the procedure spit out answer they're comfortable with and no one questions their answers."

In its own way, Rutkowski's description of equipment sizing decisions is a "systems" way of thinking. First, you need a tight, well-insulated building shell with good windows. Then you need to make sure the person performing the load calculations has accurate information so they won't "fudge" the numbers. Then they need to be aggressive with their assumptions. Finally, once they have their equipment sizing answer, they need to be sure the ductwork is properly sized, sealed and insulated. Sounds simple enough.