F-bomb was on the line. He was all nervous and excited, dropping the f-bomb in every sentence. I’m no prude and I’m not from Jersey. I selectively use the term to get someone’s attention. But f-bomb was on a roll, which I was enjoying because I knew I could help him.
“I open out the f-ing air vent and f-ing air gets sucked in. Not f-ing out!” (Sometimes it’s an adjective, sometimes it’s a verb. You get the idea.) He was having air control problems at a church and school complex where they just took over the maintenance contract. I was going to explain the importance of the relationship of the expansion tank in regards to the system pump but knew he was currently too worked up to grasp that concept. F-bomb needed a site visit.
With air control, just like real estate, it is all about location. In the real estate biz, location has a huge impact on price. In the heating biz, location has a huge impact on air control, and sometimes creates unusual results. Obviously to me, F-bomb found one of those jobs.
When I met him on the job, the situation was exactly as I had imagined from the phone conversation. The oversized system pump was on the return side of the boiler, while the expansion tank was piped from a tapping located behind the supply outlet on the boiler. In simple terms, the expansion tank connection was on the discharge side of the pump.
A circulating pump in a closed loop heating system, the type of system I work with on a regular basis, creates flow through the piping system by creating a differential pressure. That differential pressure causes the water to move from the higher pressure at the pump’s discharge to the lower pressure at the pump’s suction. It is as simple as that.
But if it is so simple, why is air being sucked into the system through the open air vent? That is not normal. Well, the key is the location of the pipe that connects the expansion tank to the system. When it is on the discharge side of the system pump, the pressure at the discharge of the pump stays the same, while the pressure at the suction side goes down. When it is on the suction side of the system pump, the pressure at the discharge of the pump goes up, while the pressure at the suction side stays the same.
In either scenario, the pressure is greater at the discharge than the suction, so water will circulate from high pressure to low pressure. Boiler manufacturers used to ship their packaged residential boilers with the pump on the return side of the boiler and a tapping on the boiler for the expansion tank. The same relationship as F-bomb’s dilemma.
Lots of installations are like this and work fine. Have for years. Will work for many more years. Maybe your dad used to do it this way. Someone through the years probably told you to do it this way. It isn’t right from the standpoint of air control, but it can work.
The right way to pipe the expansion tank is to the suction side piping of the system pump, so the pressure at the suction stays the same while the pressure on the discharge goes up. A law of physics says something like this: when water is under more pressure, the air bubbles get smaller. So the opposite is also true. When water is under less pressure, the air bubbles get bigger. It is science, and despite our best efforts, we cannot modify or get around that.
So when we turn on that pump that has the location of the expansion tank connection on the discharge side, we allow the air bubbles to get bigger. Get enough big air bubbles together in a small enough pipe or baseboard and you have an air lock and a big headache. How big they get is a function of the size pump you are using. More specifically, the amount of head pressure the pump develops when it turns on. The head pressure it develops is determined by the resistance to flow, which is caused by the friction of the water as it moves through the pipes. When a fixed speed turns on, it tries to move as much water as it possibly can, as fast as it can. The pipes, boiler and radiation fight back with friction.
The pump finds the differential pressure it can create and subtracts that from the suction side of the pump and the return side of the system. Again, the pressure lowers on the radiation, since the pressure has to remain the same on the discharge side. What is that pressure on the discharge side? It is the system static pressure, usually 12 lb. for a typical two-story residential system.
At F-bomb’s job, the static fill was set for 15 lb., because of the third floor of the school. For air to be sucked into the system, the existing pump’s differential pressure would have to be greater than 15 lb. Pump curves are typically marked in feet of head. The conversion from feet to pounds has always been one of those things I have to think about after 35 years. It is one pound of pressure equals 2.31 feet of head. Therefore, 15 pounds times 2.31 gives us 34.65 feet of head.
The existing pump was rated for 75 gallons per minute at 50 feet of head. The head seemed high for an older system. That pump has more than enough head to overcome the resistance to flow in the system and therefore be able to suck air into the system when an air vent is opened up. Historically, pumps were designed for low head applications since pipes were generally sized larger in the good old days of gravity circulation.
Through the years, pipes have gotten smaller and pumps have gotten higher differential pressures. This is causing air control problems in jobs with expansion tank connections on the discharge side. This is also helping to cure air control problems when the expansion tank connection is located on the suction side of the pump. Water under more pressure makes air bubbles smaller, preventing air locked radiation, or air being sucked into the system.
Less callbacks and happy customers equals less headaches. Look for the expansion tank location in relation to the pump on your next air control opportunity. Best practice is locating the pump on the supply side of the boiler with the expansion tank connection between the boiler and the pump. Got it Kid?
Patrick Linhardt is a thirty-five-year veteran of the wholesale side of the hydronic industry who has been designing and troubleshooting steam and hot water heating systems, pumps and controls on an almost daily basis. An educator and author, he is currently Hydronic Manager at the Corken Steel Products Co.