Why aren’t we trying harder to make steam heating systems more efficient? It seems to me we can reduce the size of most steam boilers and, at the same time, reduce their carbon footprint.

I’ve had this idea rolling around in my brain for about four months. At first, I thought I was on to some revolutionary idea but I should have known better. Take a look.

I know that we size steam boilers by equivalent direct radiation, and we size hot water boilers by the heat loss of the building. But if we're being energy conscious, why not do the following for a steam system design and steam boiler replacements?

• Perform a heat loss calculation on the building.

• Perform an EDR calculation on the radiation.

• Let’s assume for now that the EDR outweighs the heat loss significantly.

• Remove or resize radiators or, better yet, install orifices on the radiators to match the room-by-room heat loss of each room.

• Size the new steam boiler to reflect the new EDR that we now have.

Turns out, this proposition has been around longer than I have. So why hasn’t it been implemented more?

Getting help from the pros

To me, it makes perfect sense. I know there's a considerable amount of work involved, but for large multifamily buildings, the fuel savings must be significant.

To get a better understanding, I turned to the Steam Pros at www.heatinghelp.com. And with full permission from Erin Holohan, HeatingHelp’s president, I’m sharing some of that conversation (edited for clarity and style).

Steam Pros: Isn't someone running a well-“undersized” system (closely matching the heat loss, not the EDR) with good results?

Steve: For the most part, I'm suggesting the same thing here, but on a much larger scale. Hundreds of large multifamily buildings are heated with steam in Chicago.

Steam Pros: I also have given this some consideration. If there is one-third more connected EDR than needed to heat the building, could we just ignore the last one-third of the radiator?

Isn't that cooler one-third just then part of the return piping, especially in a two-pipe system? In a one-pipe, wouldn't the steam only travel the hot two-thirds and then the thermostat hopefully shuts down the system?

Maybe I am being too simplistic in this thought process. However, this summer I installed orifices in a two-pipe system. Sized to heat only 60 to 80 percent of the EDR of each radiator.

This is in an old school house that has had major window upgrades or removal. This is the coldest winter we have had in many years; if kids/teachers were cold, I would be made aware of it.

Steve: That’s encouraging! Here’s some more information. Again, large buildings with large boilers; 750 MBH to 4,000 MBH. Hundreds of them, so the potential savings would be epic. Industrial combustion modulating burners with tight control, 1.2 psi max.

Also, I have no skin in the game. I’m not a contractor anymore. I work for a well-established, nonprofit company in Chicago whose mission is “Smarter Energy Use for All.” We facilitate and provide oversight for projects such as lighting upgrades, air sealing, insulation, HVAC and smaller measures as well.

I’m just trying to get the gas and electrical meters to spin slower and a way to quantify the savings.

Steam Pros: Ninety percent of the boilers are oversized due to new envelope upgrades and removal of radiation. If not that, then there is usually too much radiation in the original design.

Maybe not if you leave the windows to comply with the 1918 pandemic. But everyone (except me) gets a flu shot, so that is no longer a worry. 

I did ask Dan Holohan at a seminar in the Twin Cities if radiators were orifice-controlled to 80 percent, then could the boiler be sized to 80-percent less than the connected EDR. He replied that, yes, it could be done.

Steve: Yes, this is what I was looking for!

Steam Pros: As the least experienced and least knowledgeable person on this thread, I am going to chime in. I am replacing most of the radiators in my house (for reasons more to do with design aesthetic and less to do with efficiency), and I purchased the replacements based on heat loss calculations because I thought that was how you were supposed to do it. (Is that not how people do it?) So, @Steve Minnich, to put real numbers to your hypothetical: 

• Perform a heat loss calculation: Mine came out to about 44,000 Btu/hour.

• Perform an EDR calculation on the radiation. Current radiators: 75,000 BTU/hour.

• Removed or reside radiators to match the room-by-room heat loss: I built in a fudge factor, so the new radiation totals 50,000 Btu/hour.

• Size the new steam boiler to reflect the new EDR: My current boiler is a Weil-McLain EG/PEG-50 (or 55; I never bothered to count the burners), kicking out 145,000 or 167,000 Btu/hour. It's running just fine, but when it finally kicks the bucket, I was planning on replacing it with a much smaller one. I was actually surprised that the boiler is so oversized relative to the existing radiation. By a factor of two; is that industry standard? 

Steve: You sound very knowledgeable to me. Experience can be overrated at times. You’re running on the same track I am and I like it.

Steam Pros: I know two guys who have written on the possibilities of downsizing steam systems. They focused on using inlet orifices in two-pipe radiators. They pointed out that since the orifices separate the radiators from the boiler and mains, the pickup factor is not needed. Also, by adjusting orifice capacities it is possible to downsize radiators that may have been erroneously oversized, bringing the system into balance. 

Since pickup factor is not a part of the equation, this means that the boiler can, indeed, be sized for the building loss as opposed to the connected radiation. Similar gains by downsizing one-pipe systems may be possible but it is a more delicate balancing act. 

A few years ago, I helped a friend straighten out a Moline system in his house that was a disaster. The guts of the Moline valves had been removed, so it was a two-pipe system with no traps and the oversized boiler made it worse. We installed inlet orifices on all the radiators. The boiler started leaking about the same time, so the Weil McLain EGH-12 with an input of 550,000 Btu was replaced with an EGH-75 with an input of 299,000 Btu. 

The system has connected radiation of 1,197 and the new boiler had an EDR rating of only 750 square feet. But there is more to consider when connecting a boiler to orifice systems such as a Moline or any other similar system. First, since the radiator is separated from the mains and the boiler by an orifice, it never sees pressure. Not even the 6 to 8 ounces that is in the mains. So, instead of 240 Btu/square foot, it’s more like 225 Btu/square foot. 

Then, you leave out the pickup allowance. If you keep the piping loss allowance, it's 15 percent. But, if the mains and risers to the upper floors are insulated, and if the system is oversized, you can leave that off, too. In addition, we made the conservative decision to size the new boiler at 90 percent of the radiator capacity, with zero allowances. That brought down to needing a boiler capable of outputting 242,530 Btu. The EGH-75 has a DOE output of 247,000.

How does it work? Like a charm. The radiators are silent except a rare clank from expansion of one radiator. On a prolonged firing, coming out of a setback, the radiators feel fully heated to the touch. But no steam ever gets into the returns. The sound of steam going through the orifices is barely audible. Overall, the heat is even. All radiators heat uniformly; steam distribution is even throughout the entire system. Operating cost has gone down significantly (I don't have the numbers) and the cost of the replacement boiler was significantly lower than a conventionally matched boiler. 

Steve: I knew I came to the right place. Thanks again! This is beginning to sound like science and empirical evidence.

Steam Pros: The Moline system was also my inspiration for using supply valve orifice plates in two-pipe steam systems. We have a number of homes and multiunit buildings that we've done over the past 10 years and those system work exceptionally well. On two-pipe systems, we always look at the historic fuel usage and, with some educated guesses on the heating plant efficiency, work out the actual heat loss of the structure. For two-pipe systems installed before World War II, the typical radiation capacity calculates out to be about 60-percent higher than needed for design conditions and this is assuming that the piping is making no heat contribution to the structure. 

When we size the orifices, we typically size them to this number. Back in the boiler room, we either down-fire the existing boiler or if replacing, install a much smaller boiler that is sized to this new radiation loss plus a 15-percent pickup factor. This process still has a lot of built-in extra capacity (the contribution of the pipe heating the building, for instance), but results in a heating plant usually about one-third to one-half the capacity of the existing or less.

This process has also allowed us to do many other things that normally are not possible. One big one is to eliminate the need for a vacuum pump on larger systems. Since vacuum system piping is about one-half the capacity of pressure systems and we are now running about one-half the steam to feed the system, the new piping is now the optimum size for the new system capacity. This assumes you have no lifts in the system. 

When possible, we couple an orifice system with an outdoor reset boiler control that modulates the burner based on outdoor temp. 

We currently have these running two ways: one system is resetting the burner output based on the outdoor temperature while the other is modulating the target pressure in the system based on outdoor temperature. Out of the two, the first is probably the best choice due to the nature of flow through an orifice. Orifices flow a very large amount of steam at very low-pressure differences, so as the pressure is increased across the orifices as the outdoor temperature drops, the increase in heating is relatively small. 

The results of these changes naturally mimic those of a hot water system upgraded to a modulating heat source with outdoor reset. Very stable and even heating throughout the structure and extremely quiet operation. Even with systems of questionable piping (some sags in steam mains or radiator runouts), noise is nearly eliminated since the piping almost never cools to allow water hammer to occur. Expansion noises also disappear. 

And, of course, there are fuel savings, despite areas that were previously too cold are now heated. While most of the systems we have worked on have had multiple changes made at the same time as the installation of orifices, the analysis of the first season’s fuel usage was a reduction of about 40 percent. 

Reducing the firing rate below radiation capacity is more of a chance on a one-pipe steam, especially in multiunit buildings since people will tinker with the radiator vents, eventually throwing the system out of balance. In a single-family home, however, it seems to work.

Steve: That’s a big help. Thank-you! How would I know which size orifice to use?

Steam Pros: I have used Tunstall's cup-type orifice plates. I size them from their table, which shows the range of EDR at various pressures. I have always used the 0.5 psi (8 ounces). When doing a Moline system, which were intended to operate on 6 ounces, I choose the next larger orifice if it's at the upper end of the range on the chart. 

My own system controls with a two-stage burner between the pressures of 7 to 11 ounces. Since I have working traps, I don't know if I would be passing any steam into the returns, but I doubt it. On the Moline system, they work just great. Or course, sizing the boiler to maintain the pressure within the parameters of the orifice sizing is key.

Steve: This is sounding better and better. Anyone else?

Steam Pros: It depends on the system and how it will be operated. If we are trying to retain or convert to a gravity return, we will size to one-half psi drop so we will have plenty of stacking height on the radiator return main drops. If we are planning on running outdoor reset of the steam pressure, we design for a 3-psi pressure drop. Tunstall recommends 4 psi, but 3 psi seems to work fine and that's what we have charts for. 

The only complaint we have seen is there is a slight whistling when running at high pressures through the 3-psi orifices. With a properly sized on/off boiler or with outdoor reset, this won't happen very often.

We size the orifices based on the radiation capacity and the heat loss of the structure. We usually find that we can meet the heat loss on the design day with the radiators only 60-percent filled, so that ends up being our design load for the system and we size the boiler accordingly. We only had one system that ran a bit short on capacity during our minus-20 degree days (earlier this year), so we know we are hitting our target pretty well because design day is only minus-4 degrees. 

The new replacement boilers installed with the new orifices are usually about two-fifths to one-half the capacity of the previous boiler. I have been using a 15-percent pickup factor for sizing the boilers, not the 33 percent since the orifices reduce the start-up load dramatically from the studies I've read. I believe one guy, in New York, has also been using a 15-percent pickup factor for quite some time.

Steve: How, specifically, are we saving energy by properly sizing a steam boiler?

Steam Pros: Attached is an efficiency curve that is often referenced if using atmospheric boilers with no stack damper. I have found it quite accurate for predicting fuel savings when reducing boiler capacity to proper levels. Oversized atmospheric boilers on steam systems can be particularly inefficient because the radiators installed in most buildings between 1900 and 1940 are also about 60-percent oversized for the heating load. 

The typical steam boiler sized to radiation with a standard 33-percent pickup factor will have about twice the capacity necessary to heat a building on the coldest day. If your boiler is 2.25 times the size required for the radiators, you may be about 4.5 times oversized for the heat load. This would yield a seasonal efficiency around 52 percent, assuming a steady state efficiency around 78 percent (which is a realistic number for most modern atmospheric steam boilers). A stack damper may help this number some under the right conditions.

Energy-efficient steam

Those are some impressive numbers reflecting the potential savings. I believe, as an industry, we should give more consideration to make steam systems as energy efficient as we can. We owe it to the building owners, ourselves and our environment.

I work for Elevate Energy in Chicago. We work with specific, income-eligible neighborhoods in Chicago, providing rebates and incentives for homeowners, multifamily building owners and nonprofits such as churches. Our goal is to give these neighborhoods the same energy-saving opportunities as others. 

My job is to provide oversight on boiler replacements we award to our pool of contractors, review their designs and proposals for accuracy and detail, and to implement new energy savings measures. 

At some point soon, I’d like to add the use of radiator orifices as a means of reducing boiler size as one of our measures. 

Thanks again to Erin Holohan and the Steam Pros who contributed to this idea. Heatinghelp.com, the place where steam problems go to die and ideas come alive.