Farm Collector

Changes in Safety Valves for Hand-Fired Boilers

At almost every show I attend, I ask a few engine owners and operators whether or not they are satisfied with their new-style safety valves. It would be only a small exaggeration to say that I get just two responses. “I have been using a new-style valve for 15 years and I haven’t had any trouble with it” or “They are junk!” I have cleaned up the second response to spare the reader the unpleasant expletives.

As the result of these conversations, I have attempted to understand why there is such a discrepancy between the reactions to modern safety valves. It now appears that there are a few simple steps we can take when purchasing and installing these valves that might improve our satisfaction with the new-style valves, which are the only ones currently available.


To understand the issues involved in the selection of a safety valve, it is necessary to review the history of safety valves used on hand-fired boilers. I am referring to hand-fired boilers rather than historical boilers because the issues are determined by how the boilers are fired, not how they are constructed or how old they are. The requirements for a modern welded boiler made to the American Society of Mechanical Engineers code are the same as for a 100-year-old riveted boiler, if both boilers are hand-fired. The requirements for a safety valve for a boiler that is automatically fired are dramatically different.

With automatic firing, the safety valve’s function as defined by Anderson Greenwood Crosby, a manufacturer of modern safety valves, is to protect life and property if all other safety measures fail. A safety valve on a hand-fired boiler, as defined by ASME almost a hundred years ago, is to give notice of the highest pressure permissible and to give alarm that more water or less fuel is needed. The evolution of the purpose of the safety valve is summarized in Figure 1.

When hand-fired boilers, such as found on traction engines, steam cranes and locomotives disappeared, most of the boilers that remained were automatically fired. The safety valve manufacturers adapted their designs accordingly. The old-style valves with bottom guided, beveled seats were capable of withstanding vibration and operating near their setpoint, and were replaced by smaller top-guided valves with flat seats.

At the same time, steam system designs were adapted so there was no need to operate within 10 percent of the setpoint of the safety valve. Not all old-style valves had beveled seats, but the ones that didn’t were designed much differently from the modern flat-seated valves.

Evolution of the Safety Valve

When I speak of a modern-style safety valve, I am referring to a valve shown in Figure 2 (see the Image Gallery). An example of an old-style safety valve can be seen in Figure 3. Changes in the design of safety valves had a dramatic effect on their capacity in pounds of steam per hour. If a boiler built in 1920 required a valve capable of releasing 1,000 pounds of steam per hour at a pressure of 100 pounds per square inch, it would have been equipped with a 2-inch safety valve. Today, 3/4-inch valves are available to release that much steam at that pressure.

The evolution of the safety valve did not end with the development of the modern, flat-seated valve. In the last 20 years, the design of safety valves has continued to evolve. In 1985, a new-style 3/4-inch valve set at 150 pounds per square inch could have a capacity of 1,497 pounds of steam per hour. In 2002 this same valve could have a capacity of 1,651 pounds of steam per hour.

From 1914 until 1998, the blowdown allowed by the ASME boiler code was 2 to 4 percent of the set-point. In 1998, this was changed to allow the blowdown to be as high as 6 percent. Beginning with the 2004 ASME code, there is no limit on blowdown. The code has not required that the amount of blowdown be stamped on a safety valve since 1986.

When I asked owners and operators how well they liked their new-style safety valves, I was not aware of the need to ask about the age of the valve. Instead, I would ask about the amount of blowdown. In almost every case, the people who were satisfied had valves that would blowdown 4 percent or less. It appears that this is almost the same as if I had asked the age of the valve. If the valve were purchased prior to 1998, it would have been set for 4 percent blowdown. If purchased after that date, unless specified otherwise, it would have been set for 6 percent. The change from 4 to 6 percent causes a 50 percent increase in the amount the pressure changes in a boiler each time the valve pops. The resulting increase in the flexing of the components of the boiler may be associated with a corresponding increase in seepage at stay bolts and tubes.

Ordering a New Safety Valve

When you order a new safety valve, you will need to provide four pieces of information: the setpoint, the capacity in pounds of steam per hour, the blowdown and the requirement that the valve be stamped with the ASME “V” stamp. If you specify the pipe size, you may get a valve with far too much capacity, as I have already explained. To determine the capacity you need, do not use the capacity stamped on the old valve. First, if the valve has been replaced, it might not be the right capacity. Secondly, the capacity stamped on the valve is probably the capacity of the smallest valve available and might be significantly larger than the required capacity.

Calculating the heating surface of the boiler, in square feet, and multiplying the number by a factor can determine the required capacity. The ASME code requires a factor of 5 pounds of steam per hour per square foot of heating surface. Ohio requires a factor of 7. The heating surface is the area above the grates that is exposed to the fire. This includes the firebox, the tubes and the front tube sheet.

As I talked to many owners, they would offer other comments regarding their valves. One comment I heard from several owners who were satisfied with the new-style valves was that their valves were larger in pounds per hour than the minimum required by the ASME code. I am not certain as to exactly how the capacity affects the operation of the valve. What I do know is the larger the valve the more force it takes to raise it off its seat. The force of the steam on the boiler-side of the valve cannot overcome the force of the spring in the safety valve until the pressure in the boiler rises to the setpoint of the valve.

When this happens, the valve pops open. It seems the larger diameter, and thus the greater forces, may result in more stable operation of the valve near its setpoint. There is a concern in the boiler code that safety valves should not be so large that water is drawn out of the boiler. It would seem, because of the relatively small size of portable and traction boilers, the pressure would drop so quickly that little water, if any, would be lost. I have seen boilers where the owners have installed modern safety valves of the same pipe size as the old style valves installed by the factory. The capacity of these valves was far greater than I believe anyone would recommend, but I am not aware that they caused water to be discharged from the boiler. It is important to be careful when sizing a safety valve. I suggest owners talk to each other and share their experiences before making such a decision.

If you would prefer to have a top-discharge safety valve, which looks more authentic, shop around. They are available in a few sizes. You might also want to consider specifying that the valve have a non-metallic seat.

New Valves

When installing a safety valve, do not install any fitting smaller than the inlet to the valve and do not install any kind of valve between the safety valve and the boiler. Examples of what not to do can be seen in Figure 4. Do not use a pipe wrench on a safety valve, it can damage or destroy the valve.

Once you have carefully selected your safety valve and have installed it on your boiler, it is important to verify the setpoint and the blowdown have been set according to your specifications. The first step in this process is to have the accuracy of your steam gauge checked with a dead-weight gauge tester. If your gauge does not agree with the setpoint of your new safety valve do not assume that the gauge is wrong.

Do not use the lifting lever to lift the valve from its seat until the boiler pressure is up to 75 percent of the setpoint of the valve. If the valve is lifted from its seat at a lower pressure, any dirt or foreign material in the valve might not be blown clear of the seat and could damage the seats when the valve closes.

Because new-style safety valves are not designed to be operated within 10 percent of their setpoint, many owners have elected to install the new valve along with an old-style valve. In doing so, the old style valve operates in the range of 5 to 10 percent below the setpoint of the new valve. With this arrangement, the new valve satisfies the code requirements while the old-style valve performs the function for which it was designed. Two possible arrangements can be seen in Figure 5.

Concern About Quality

In response to complaints from owners of historical boilers who had recently purchased new safety valves, Dean Jagger, Ohio’s chief boiler inspector, requested that the National Board test valves from the manufacturer to determine if the valves complied with the requirements of the ASME boiler code. As a result of these tests the Ohio Department of Commerce issued a safety notice:

“The State of Ohio Boiler Division has been made aware of the fact that some recently purchased Kunkle safety valves, which were assembled by Allied Industries, have been tested by the National Board Testing Laboratory and found not to be in compliance. The tests indicated that the valves blowdown and set-point pressure settings were out of tolerance as established by Section I of the ASME Boiler Code.”

This may be an indication that all of the problems with modern safety valves are not entirely the result of design issues, but insufficient oversight of manufacturing and quality control processes may also be a factor. A new valve may be “junk” as has been so often alleged.

The errors found by the National Board Laboratories were significant. One of the valves was stamped 165 psi but popped low at 148.8 psi. Another was stamped 150 psi and popped high at 164.5 psi. On the other three valves the pop was consistent with the setting stamped on the valve. The 2001 edition of the ASME Boiler Code specifies that for pressures from 70 to 300 psi the tolerance, plus or minus from the set pressure, shall not exceed 3 percent of the set pressure. The 165 psi valve popped 9.6 percent below the set pressure stamped on the valve, and the 150 psi valve popped 9.7 percent above the set pressure stamped on the valve.

The blowdowns on all of the valves that were tested were out of tolerance. The 2001 edition of the ASME Boiler Code specifies that for pressures from 67 to 250 psi the blowdown shall not be greater than 6 percent of the set pressure.

With such a wide range of variations in both set-point and blowdown, in a sample of just five valves, it seems reasonable to suspect that even greater variations may exist. The results of the tests are shown in Figure 6.

Complaints about quality problems are not limited to the five valves recently tested by the National Board; for example, an engine owner told me of purchasing a new 1-1/4-inch valve stamped 175 psi. When installed on a traction engine, the valve consistently popped at 185 psi and blew down 15 psi. (A pressure of 11.1 psi equals the allowed 6 percent.) The manufacturer told the owner that the valve had been tested properly prior to shipment but accepted it back. The owner had verified the accuracy of the pressure gauge prior to contacting the manufacturer.

The ASME and National Board procedures for safety valves merely confirm the adequacy of the design of the valve and do not assure the adequacy of production and quality control practices. Each boiler owner and operator must carefully confirm the accuracy of the setpoint and blowdown on every safety valve and not rely on the ASME and National Board stamps as assurances of quality. At this time, I have no reason to believe the monitoring of the ASME and National Board requirements at other valve manufacturers and assemblers is any different than what existed at Kunkle and Allied.

One scenario that concerns me is the owner who installs a new safety valve on his boiler and, seeing that the pop does not coincide with the reading on his 80- or 100-year-old gauge, decides that obviously his gauge must be wrong. This is a conclusion I am sure I would have considered when I first began my study of safety valves.

Incorrect settings of safety valves are more likely to be detected when the valves are used on hand-fired boilers than when used on modern boilers. If the controls on a modern boiler are to limit the pressure to 10 percent or more below the setpoint of the safety valve, the valve can be set as much as 10 percent below its rating and the error might not be apparent. Errors above the setpoint also wouldn’t be obvious even if the boiler were operated up to the setpoint stamped on the vale. Also, incorrect setting of the blowdown would not be apparent until the valve had operated.

Contact steam enthusiast Bruce Babcock at: (740) 969-2096; e-mail:

  • Published on Sep 1, 2006
© Copyright 2022. All Rights Reserved - Ogden Publications, Inc.