What is the key to an efficient piping layout for domestic hot water? The answer is to keep the volume of hot water between the water heater and the tap as small as possible. The difficulty is that most buildings have only one source of hot water and the many uses are spread throughout the floor plan.

The least expensive answer for new buildings would be to group all of the bathrooms near the kitchen, so that the home has one or more plumbing cores. That way, that hot and cold water uses would all be located close together, near the water heater or boiler that served them. In houses, this would typically mean one or two such cores.

In the absence of such cores, we need to plan the plumbing to supply hot water as efficiently as possible. Those of you who know my work have heard about Structured Plumbing. (For more information on Structured Plumbing, see Saving Water and Energy in Residential Hot Water Distribution Systems and Guidelines for Specifying Structured Plumbing Systems.)

Structured Plumbing is a method of running the trunk line of the hot water distribution system from the water heater past the hot water uses so that the volume from the trunk to each use is as small as practical, ideally less than 1 cup. In addition to routing the hot water supply close to each use, the Structured Plumbing approach includes a demand-initiated pump that allows the occupants to prime the trunk line shortly before they want hot water.

Green Building Advisor has had several articles on topic of efficient hot water supply over the years; see, for example, Hot-Water Circulation and Waiting for Hot Water.

A proposed amendment to the plumbing code

Current building codes permit builders to install piping systems that waste too much hot water. I’m now drumming up support for a proposed amendment to the International Plumbing Code.

The proposal would set requirements resulting in small volumes of hot water in the pipes between the hot water source and the use. It is intended to apply to all occupancies, although in the article I am focusing on residential applications.

The ICC is holding Final Action Hearings on October 22-28, 2012 in Portland, Oregon. The proposed amendments to the International Plumbing Code (IPC) will be heard on Wednesday, October 24th, starting at 1:00 pm.

Among the key provisions in the proposed change are:

• a requirement that 3/8-inch pipe should only be used for fixture fittings that have a flow rate that does not exceed 1.5 gallons per minute;
• a requirement (for some residential applications) that the maximum volume of water in the pipe between the water heater and a tap shall be 64 ounces;
• a requirement (for some residential applications) that the maximum length of piping between a water heater and a tap be restricted to 50 feet for 3/8-tubing, 43 feet for 1/2-inch tubing, and 21 feet for 3/4-inch tubing.

To understand how these provisions would be implemented, it is essential to read the code proposal in its entirely. (See the link provided below.)

Assembling the required votes

To get this proposal adopted, we need to have enough code officials who support the measure to attend the hearing and vote for it. We need a two-thirds majority to prevail. So please talk with your local code officials and ask them to support it. If they are not able to travel to Portland, ask them to get you in touch with those from your state who are going to attend. If that is not possible, please ask them to send a short letter of support that can be read into the record.

If you are able to attend the meeting in Portland, short statements providing reasons why this proposal is worthy of a “yes” vote are in order. We need perspectives from the point of view of architects, plumbing engineers, builders and plumbers. I am willing to help you craft your testimony so that is most effective for this audience.

Proposed code changes

Now to the proposal itself: P130-12 (AMPC)-Klein is the most important of the proposals I have submitted. The proposal can be viewed on the ICC website. (To navigate through the document, click on the “P130” link in the sidebar that appears on the left side of the screen.)

Three other proposals — P89, P92 and P129 — concern recirculation pumps and pump controls for domestic hot water systems and are also worthy of your consideration. They can be found in the International Plumbing Code section on the ICC website linked to above.

The “reasons statement” accompanying the proposed code change

As an I-code, the IPC specifies minimally acceptable requirements for plumbing. Delivery of hot water to a user in a timely manner is one measure of plumbing performance. The American Society of Plumbing Engineers (ASPE) specifies the delivery of hot water to the user in 10 seconds or less as “acceptable performance,” delivery of hot water to the user in 30 seconds or less as “marginal performance,” and delivery of hot water in more than 30 seconds as “unacceptable performance”. Implementing this proposal will improve health and safety by correlating the IPC with local health codes and with good plumbing engineering and plumbing practice. It will also result in satisfied users, including those in areas with low water pressure.

The core of this proposal is to make sure the volume of water in the pipes, which must be cleared out before hot water can be delivered, does not prevent the delivery of hot water in a timely manner. If you agree that delivery of hot water should be at least “marginally acceptable” in terms of time-to-tap, then you need to support this proposal.

Supporting Information

The following documents the values in this proposal and demonstrates the response to the committee’s comments. The committee [previously] disapproved P130 because the “volume limitations were too restrictive and unrealistic to apply to all buildings.” In response, this comment increases the lengths for smaller diameter branches from circulation loops or heat-traced lines. It also improves the readability of the code text.

Why implementing the 2012 IPC often results in “Unacceptable Performance”

The 2012 IPC allows for 50 feet of developed length [of piping] – of any diameter – from the source of hot or tempered water to the fixtures. However, the delivery of hot water is a question of volume (length and diameter) between the source and the uses and flow rate of the use. At current legal flow rates for faucets, showers and many appliances, 50 feet of piping contains more water than can be cleared out in the Marginally Acceptable time of 30 seconds or less, let alone the Acceptable Performance time of 10 seconds or less.

We are all familiar with the problem of waiting for hot water to arrive. When it takes too long at hand-washing sinks, many of us just give up and use whatever temperature comes out. When it takes too long at a shower, we watch the water run down the drain until the water is hot enough to use. When it takes too long in public restrooms or at hand washing sinks in food service establishments, it becomes a concern for our public health code colleagues.

Providing hot or tempered water to public lavatory faucets is a special case, and the reason we have called it out in this proposal. The time-to-tap is particularly important for hand washing events, which tend to be of very short duration, generally 5-10 seconds long. Large volume in the fixture supply piping, low flow rates and short events result in it taking a very long time for the water to get warm. Correcting this requires keeping the volume small enough so that hot water arrives in a timely fashion when only one faucet with a maximum fixture fitting flow rate of 0.5 gpm or a maximum volume per cycle of 0.25 gallons is being used. Having even Marginally Acceptable performance requires piping lengths much less than 50 feet long.

Can a volume limit be applied to all buildings?

Yes. The specifics have to do with the configuration of the hot water system within the building. There are three typical configurations for a hot water system: a central water heater (or boiler) with trunks and branches serving each use or group of uses; a central water heater (or boiler) with a circulation loop or heat traced trunk line and branches to each use; distributed water heaters (or boilers) located close to the uses they serve. Buildings can have one or a combination of these systems as long as the 2012 IPC requirement of no more than 50 feet of developed length on any path from the source to the use is met.

The volume limitations in this proposal work in any building. Buildings with vertical risers will be able to comply by locating the fixture fittings and appliances close to a circulated riser; this should not be a problem as they are relatively close already. Buildings with a central corridor circulation loop will be able to comply by locating the hot water fixture fittings and appliances closer to the corridor or by moving the loop closer to the fixture fittings and appliances. Buildings with public lavatories can meet the volume and length limits in this proposal in several ways including bringing circulation loops closer to the faucets, priming the branch lines that run behind the wall when people enter the lavatory, heat tracing the branch lines or installing water heaters in the bathroom or under the sinks.

In some buildings, no changes to architectural design will be needed – it will only be necessary to design and install the plumbing to meet the new code. In other buildings changes in the architectural design will be needed so that the hot water system will meet the new code. It is likely that we will see more buildings with combinations of hot water delivery systems. Based on my experience with improving the performance of hot water systems throughout the US, costs for additional water heaters or for somewhat longer circulation loops and heat traced trunk lines will be more than offset by the savings in smaller diameter trunk lines and in shorter branches that are often of smaller diameter because their length is smaller too.

What should be the maximum allowable volume?

Implementing the IPC should result in at least marginally acceptable performance at legal flow rates, in all occupancies, even in areas with low water pressure. The American Society of Plumbing Engineers (ASPE) has established performance criteria for the timely delivery of hot or tempered water (Domestic Water Heating Design Manual – 2nd Edition, ASPE, 2003, page 234). Table 1 [reproduced below as Image #2], taken from text in ASPE’s manual, presents the time-to-tap performance criteria. According to this table, 30 seconds is the maximum amount of time to have Marginal Performance. Anything longer is unacceptable.

So how much water is contained the IPC allowable limit of 50 feet of developed pipe length? Will clearing out this volume of water result in at least marginally acceptable performance? Table 2 [reproduced below as Image #3] shows the volume contained in 50 feet of pipe for nominal diameters up to 4 inches. (I realize that 50 feet of developed length is almost always shorter than 50 feet of pipe, but for simplicity, I have used 50 feet in the table.)

Let’s look at a few examples: 50 feet of ¾ inch tubing contains 1.2 gallons, 50 feet of 1 inch contains just under 2 gallons, 50 feet of 2 inch contains 7 gallons and 50 feet of 4 inch contains more than 28 gallons. This is the minimum volume that must be cleared out of the pipe before hot water will get from the source to the use. (Based on research conducted by the California Energy Commission, the actual volume that will come out before hot water arrives is more than volume contained in the pipe. In ¾ inch nominal pipe, approximately 25 percent more water will come out at 2 gpm; 50 percent more will come out at 1 gpm and 100 percent more will come out at 0.5 gpm. The amount of additional water that comes out gets larger as the pipe diameter increases.)

Table 2 [Image #3 below] also shows the consequences of the volume in terms of the time-to-tap for flow rates of 2, 1 and 0.5 gpm. This range of flow rates is typical of showers, sinks and public lavatory faucets. Near the top of the table, the minimum time to clear out the cold water in the pipe is shown in seconds, further down it is shown in minutes. (NA is shown when we considered the flow rate to be excessive for the pipe diameter – either too much pressure drop or excessive velocity, or both – based on an analysis using the Hazen-Williams formula.)

None of the times shown in Table 2 are within the Acceptable Performance range. This means that if plumbers or plumbing engineers design a hot water system to meet the minimum 2012 IPC, without also paying attention to the volume in the piping it will most often result in Unacceptable Performance. The best they can get is Marginal Performance in a limited number of cases. 

Table 3 [reproduced below as Image #4] compares the time-to-tap performance different volumes that are being discussed at this Final Action Hearing. The flow rates in the table are typical of faucets and showerheads, and many appliances such as dishwashers and washing machines.

Using ASPE’s criteria, only 3 data points in Table 3 have Acceptable Performance; 9 have Marginal Performance; all the rest have Unacceptable Performance. None of the volumes have Acceptable Performance for the low flow rates (0.5 gpm and smaller) found in public lavatory faucets. In addition, the Performance times shown in the 0.25 and 0.5 gpm columns are longer than the actual event itself, which is often only 5-10 seconds long. To make any sense at all, hot water must reach the faucet before the event is over, which is why there is a separate volume requirement in this proposal for the fixture fittings with these flow rates that are found in public lavatories.

We need to assess the performance when flow rates are between 1 and 1.5 gpm, not the maximum values of 2.2 and 2.5 gpm allowed by code for faucets and showers respectively. Why? One reason is that the flow rates of faucets and showers are rated at pressures of 60 and 80 psi respectively. In practice, operating pressures are often less than the rated pressure and the actual flow rate is less than the rated flow rate. In addition, hot water is only a portion of the total flow rate. The reduction in flow rate is most noticeable in areas with low water pressure to begin with. Another reason is that studies done in Indiana, California and Minnesota have found that even when full flow rate faucets and showers had been installed, the hot water portion of the flow was most often between 1 and 1.5 gpm. In this range of flow rates, the 300-ounce volume has Unacceptable Performance. The 75-ounce volume has both Unacceptable and Marginal Performance. The 64-ounce volume has Marginal Performance. The 24-ounce volume has both Marginal and Acceptable Performance. I believe the IPC should provide at least marginally acceptable performance at typical flow rates for all areas in the jurisdiction, including those with low pressure.

This section only applies to Public Lavatory Faucets

The time-to-tap is particularly important for hand washing events in public lavatories, which tend to be of very short duration. It becomes essential to keep the volume from the source to the use very small when the fixture fitting flow rate is only 0.5 gpm. Looking at the row for ½ inch nominal tubing in Table 2, the minimum time to clear out the cold water would be 1.2 minutes. Assuming that each hot water draw lasts 5 seconds, and that all draws happen right after each other, the 15th user would get hot water. If the branch line were larger, say ¾ inch, the minimum time increases to 2.3 minutes and the 28th user would get hot water. If the branch line was 1 inch, the minimum time increases to 3.9 minutes and the 47th user would get hot water.

The delivery of hot water to public lavatory faucets needs to be considered separately because of potential health issues. The events are short and the flow rates are low. Table 4 [reproduced below as Image #5] shows the time-to-tap performance based on the requirements in the proposal. The 0.25 and 0.5 gpm columns show that even at very low flow rates this volume will result in Acceptable Performance according to ASPE criteria. Given the short amount of time people spend washing their hands in public restrooms, it does not make sense to Marginal Performance category for determining the volume from the source to the use for public lavatory faucets. The volume was chosen so that hot water would arrive in the first part of the hot water event so that every person who uses the public lavatory will have the benefits of hot water.

Now to the decision

The provisions in the 2012 IPC (and previous versions), which only limit the feet, do not give guidance on the volume and as we have shown, often as not result in Unacceptable Performance. Unfortunately, many of us have experienced this! In contrast, this proposal contains the provisions necessary to support the correlation of the plumbing and health codes with good plumbing engineering design and plumbing installation practice.

There are 3 key questions that we are asking you to answer:

1. Do you want the IPC to support the provisions in local health codes to supply hot or tempered water for hand washing for every user of public lavatory faucets?

2. Do you want the IPC to support the ability of plumbers and plumbing engineers to provide hot water within 30 seconds after opening the tap; this is the Acceptable and Marginal Performance ranges as defined the American Society of Plumbing Engineers. (See the arrows next to Tables 3 and 4.)

3. Do you want the IPC to provide these levels of performance in all parts of your jurisdiction, including those with low water pressure?

If so, please support this comment.

Gary Klein is the managing partner of Affiliated International Management. His firm provides consulting on sustainability, primarily on the water-energy-carbon connection.