Heat Exchanger Relief Valve sizing

[Cool_Controls]

I have come across a shell and tube heat exchanger in the basement of my office building that has no pressure relief valve on it and I would like to know how to select the right size relief valve for it. The heat exchanger provides hot water for the buildings HVAC system, 15psi steam on the shell side, 40psig water in the tubes, centrifugal pumps. Its always been my understanding that the relief valve in this context protects against thermal expansion either from a blocked in condition or failed expansion tank. A tube failure on these units just results in heating water being lost into the condensate line and making the boiler plant operators annoyed at the hard water coming back to the plant.

 

[EnergyProfessional]

You mean you have a steam to water HX and need a relief valve on the water side?

That depends on a lot: height of system, location of the relief valve in relation to high point, max pressure of all equipment. the pressure in the basement needs to be high enough to be above atmosphere at the highest point. but not so high, that at some point it is above the allowed pressure. Every 2.31 feet of height you lose or gin 1 psi if this is pure water. So if the basement has 40 psi, at 23 foot height, you only have 30 psi. but 23ft below the basement, you have 50 psi.

so thsi is a bit more complex than someone on the internet just telling you a number.

 

[Cool_Controls]

Yes I need a relief valve on the water side, cold fill pressure of the system is 40psi (I understand the relationship between system height and static head). I'm trying to understand what equations or ratings are to used to size the relief. ASME Section IV HG-400.3 would seem to indicate a 1" relief valve with a set pressure equal to or less than the heat exchanger MAWP is all you need, but that seems too simple.

 

[EnergyProfessional]

You need to evaluate ALL system components, not only the HX. And yes, the relief valve has to ensure ALL components are below their rated pressure.

Cold fill pressure of 40 psi sounds high assuming this is a low rise building. But again, depends on at what height.

Yes, a relief valve located at the XHX and a setting lower than the HX pressure will protect the HX. But that may not be correct for the rest of the system. You cold over-pressurize a different component, or the relief valves open all the time....

 

[Cool_Controls]

I can confidently say the HX has the lowest MAWP in the system, its not the pressure that concerns me but the BTU/Hr rating of the relief valve I'm trying to figure out. I would also point out the proper placement of relief valves in this context does not seem to be well understood in the HVAC design community, I have come across a number of designs where the engineer placed the relief valve in a location with intervening valves between the valve and the pressure vessel or attempted to protect multiple HXs with one relief valve even though there were multiple isolation valves between the HXs and the relief valve. If it was a boiler the relief valves BTU/Hr capacity would only need to be greater than or equal to the BTU/Hr rating of the boiler.

 

[EnergyProfessional]

The relief valve could be at any location that CANNOT be isolated off from the heat-generating device, like a boiler or an HX in your case. So if you have an isolation valve for the HX, the relief valve must be on the HX side of the valve.

I can't speak for the rest of the HVAC designers, but I don't see relief valves at the wrong location and the inspector would see that. They look for that. I see many HVAC design and installation problems, but relief valve locations isn't one of them.

btu-rating is based on the capacity of the HX. For an HX this is a value that depends on the fluid temperatures and flowrates, so that isn't a number just written on it (like for a boiler that has a gas line limit rating)

 

[Cool_Controls]

In Delaware these type of heat exchangers do not get inspected like boilers do so I think a lot of things get overlooked (or modified by maintenance over the years). I was planning to use the BTUs based on the design flow rates and temperature rise, but than I came across HG-400.3 "Safety and Safety Relief Valves for Tanks and Heat Exchangers

(a) Steam to Hot Water Supply. When a hot water supply is heated indirectly by steam in a coil or pipe within the service limitations set forth in HG-101, the pressure of the steam used shall not exceed the safe working pressure of the hot water tank, and a safety relief valve at least NPS 1 (DN 25), set to relieve at or below the maximum allowable working pressure of the tank, shall be applied on the tank."

 

[TBP]

The steam-side safety valve is SET for 15 PSIG, or the steam service to the building OPERATES at 15 PSIG? If the steam system OPERATES at 15, that means that the safety valve setting is some higher pressure.

 

[Cool_Controls]

The steam system operates at 15psi. I'm concerned with the water side pressure relief valve.

 

[TBP]

There are 2 sides to your HX - I'd look at both of them.

 

[Cool_Controls]

The steam side is good, there is a relief valve after the pressure reducing station.

 

[BronYrAur]

Given that this system has been apparently operating successfully for years, the relief valve pressure setting can be determined by knowing your system water volume, whether or not glycol is present, your cold fill temp, your max water temp, and the size (and acceptance) of your expansion tank. Normally, the relief valve setting is a factor in determining the size of your expansion tank, but you will have to reverse-engineer this one.

 

[EnergyProfessional]

system volume has exactly zero to do with relief valve setting. it is all about pressures and height. the system could be 10 gallons, or 1000000000 gallons... as long as the heights are pressures are the same, volume does NOT play a role.

 

[LittleInch]

Your only issue is thermal expansion of cold water trapped in the HX between two isolation valves.

This only needs a very small thermal relief valve set at 40psig.

The steam is unable to provide enough temperature to boil the water at 40psig.

If you have no trapped volume then you don't need a relief valve.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[BronYrAur]

What planet am I on here? I'm the first to admit that I don't know everything, but I could not disagree more with the last 2 responses.

EnergyProfessional, your response is too surgical. The OP is trying to reverse-engineer this system and make a relief valve selection. Although you are correct from the perspective of a new design that volume has nothing to do with relief valve setting, how can this be true with an existing system? You have a given volume and a given expansion tank size. There will be a give thermal expansion which will result in a pressure increase. The relief valve setting will have to be higher than this or it will pop during normal operation.

LittleInch, how is it that a relief valve is not needed regardless of isolation valves? There is a heat-producing piece of equipment that is heating up a closed loop of water. What if the expansion tank floods and the system fills with cold water and then starts heating up? There will be no where for the expansion to go except a rupture. Similar situation if the steam valve gets stuck open.

Correct me if I'm wrong, but that's how I see it.

 

[EnergyProfessional]

If the expansion tank is insufficiently sized, you can't just make up for that by installing a larger relief valve. Once you are out of expansion volume (for the given temperature limits) pressure will go up and the relief valve will constantly open up to release pressure. Or you install a relief valve way too large and you break equipment and piping. So if the expansion tank is too small (or faulty!), it needs to be replaced with a large enough one. No relief valve will resolve am expansion tank sizing or performance issue. Your desired system pressures will impact expansion volume. Based on all the equipment (and its elevation) you need to determine min and maximum pressures. Based on that and volume (and elevation of the tank), you size the expansion tank.

Once expansion tank functions properly and is sized correctly (which it should), volume is not relevant. Pressure is pressure, regardless of if you have a gallon, or the entire Pacific Ocean in the system. A well designed system will never experience the relief valve opening except for testing. It is a safety valve that will open to prevent catastrophic failure in case your expansion tank fails (broken bladder etc.), your heat source runs amok etc. it is not supposed to open up under normal operation. Not ever. it is like an airbag in your car, if it deploys, something went wrong....

I recommend starting with this book

 

[LittleInch]

BroYrAur,

Out problem is that we can't see the whole system and the OP hasn't posted any sort of diagram as to where the trapped volume could be.

Expansion tanks normally come with relief valves or there may be one somewhere else. We don't know.

Even if there is a locked in volume, the system a described only needs a small thermal relief valve as the steam temperature cannot over pressure the water side due to steam production.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[BronYrAur]

So for an existing system with a given volume and existing expansion tank, there should already be a relief valve, but there is not one in this case according to the OP. All that is known is the volume (estimated), the cold fill pressure, and the temps. I would of course first check the pressure limits on the heat exchanger, coils, etc. Then I would use the expansion tank selection software to "select" an expansion tank. I would do this by inputting the volume, fill pressure, temps, and making a guess at the relief pressure. It would be an iterative process until the software selected the tank that is actually installed. The water volume must be known in order to do this. That would tell me the relief valve pressure that would be appropriate so that the valve would not lift under normal circumstances.

Otherwise, you might as well just select the relief valve pressure based on the pressure rating of the system components.

 

[EnergyProfessional]

Relief valve is (or should be) on the HX side of any isolation valves. Not by the expansion tank. There likely is an isolation valve somewhere between HX and expansion tank (there is at least a lockable one for expansion tank replacement). so a relief valve by the expansion tank will not do anything for code compliance or safety since it could be isolated from the heat source.
Relief valves have a Btu/h rating that corresponds to the volumetric flowrate they could discharge. This depends on the heat source maximum capacity to produce heat. And a pressure rating... that, as the name implies, is based on pressure rating. To size the valve, you need to determine both values. Not more, not less.

The relief valve also should be piped to a drain so you don't spill boiling water over all the people around it and not to make a mess when you test it. They should be tested monthly, and that also gets the dirt out.

I didn't think a relief valve can cause that much uproar. They have been successfully used since the time insurances got fed up with frequent boiler explosions. It is also one item inspectors actually look for.

 

[TBP]

I would look at it like this: If this system had it's own stand-alone hot water boiler, that boiler would have shown-up with whatever relief valve(s) were required as determined by the manufacturer, in order to protect the boiler. The boiler manufacturer has no idea what the system volume may be, and I doubt that he cares all that much. For the installation under discussion, the boiler has been replaced by a HX. What's the maximum heat input to that HX? That's why I asked the question I did regarding the setting of the steam safety valve - people very often confuse their operating steam pressure with the safety valve setting. I've seen LOTS of steam systems full of 125# valves & fittings, because that's their operating pressure. But the boilers that supplied many of these systems had their safety valves set for 150#. And those 150# safety valves were the ONLY ones on those systems. Lots of people confuse OPERATING pressure with DESIGN pressure.