Primary-Secondary Systems Sometimes Overheat

The guy had told his architect to have a corner of his basement dug out so that he could play half-court basketball with his two kids.

The ceiling was 14 feet above the parquet floor and there was an electronic scoreboard hanging on the wall that would have made Michael Jordon smile. Down the hall there was a theater, a mahogany bar with lots of stools, a workout room, and a wine cellar. And that was just what was in the basement! This guy had all the money.

The contractor who installed the heating system in this yuppie wonderland had pulled out all the stops. The owner had asked for the best that was available and that’s pretty much what he got. The boiler room looked like a collaborative effort between NASA, Bill Gates, and Leonardo da Vinci. It took my breath away when I first saw it. Four boilers stood side by side along the wall. The pipes lined up with military precision in a primary-secondary fashion. The controls were in charge and knew exactly what had to be done. It was everything I would have assembled if I had all the money.

"Feel this pipe," the contractor whispered, glancing nervously toward the boiler room door. We were standing in front of a long primary manifold. There were 20 pairs of secondary lines coming off the bottom of the manifold. Each dipped down a foot before turning up toward the ceiling and then heading out toward the far reaches of the palace.

I grabbed hold of the secondary line and winced. "Ouch! Is the circulator on?"

"No," he said nervously. "Now feel this return line over here. It’s even hotter, and this circulator’s off too. I’m pretty sure the primary pump is pushing water up into the secondary circuits. Here, feel this one over here." He led me to a pipe on the far side of the boiler room. I reached up and grabbed it. It was hot. "That’s a return line," he said. "How could it get that hot unless the primary pump is doing it?"

This was one of those times in life when the reality of the job supersedes what you read in textbooks, but I just couldn’t see it in my mind’s eye. The supply and return tees to each secondary circuit were so close together. They were practically touching each other!

"I’m not so sure about the primary pump," I said. He rolled his eyes in frustration. I pressed on. "I mean, if you were water, and you were flowing through this primary main and you got up to this first tee, wouldn’t you go straight through into this second tee?" I pointed at it. "It’s only an inch away. Why would you go through the bull of that first tee, all the way up through that long circuit, and then back through the bull of the second tee? It just doesn’t make any sense. Water’s always going to take the path of least resistance, and in this case, that path of least resistance is such a clear choice. It’s only an inch away, for Pete's sake."

"Then how come the pipe’s so hot all the way over there?" he asked, hitting me in the puss with the reality of the situation. I hate when that happens.

"I don’t know," I admitted.

"It’s gotta be the primary pump," he insisted.

"I don’t know. I think it’s probably gravity circulation."

"How can it be gravity circulation?" he asked. "I’ve dropped every secondary circuit a foot below the primary main. Hot water rises. How can it go down?"

"I don’t think a foot is enough of a drop," I said, feeling the hot pipes again.

"Well, what would be enough of a drop then?" he asked. "Should I go all the way to the floor? The reason I did the heat traps was to avoid having to use forty flow-control valves. Am I going to have to take all of this apart and install flow valves?"

"Maybe."

"I still think it’s the primary pump," he insisted. "It’s gotta be pushing up into the secondary circuits. Look at how fast it’s moving."

"Then why are some of the returns getting hot?" I asked. "I mean, feel this one over here." I grabbed hold of a return line, about ten feet away from the primary main. It was red hot. "The supply line for this same circuit is lukewarm, but feel this side."

"It’s gotta be the pump," he insisted again.

And we went around and around like that for a while. You’ve been there, haven’t you? You know how two guys can go around and around in a troubled boiler room? That was us.

Anyway, we got to the point where we were just staring at the equipment. Jobs like this one require a great deal of staring.

Finally, I suggested that we drill and tap the supply and return side of one of the secondary circuits and put some pressure gauges on it. My thinking was that we could get a pressure reading on the two gauges with the primary pump off. Then we could start the primary pump to see if the gauges changed. If the primary pump moved water through the secondary circuit, the gauges would show a difference in pressure from supply to return. If the gauges didn’t move, that meant that the movement of heat that we were feeling was happening by gravity circulation alone.

We spent a half-hour getting the thing set up and then we started the pump. The gauges held rock-steady. We looked at each other and knew for sure that what we were seeing was gravity circulation. "It can move that fast?" he said. "I guess so," I said, hardly able to believe what I was seeing. If you were there with us, I’ll bet you would have thought it was a pumped flow as well. It was moving that fast! But it didn’t seem to care whether it was doing it on the supply or return side, and that’s what was so strange.

So we learn. He wound up putting in 40 flow-control valves and the problem went away. Neither of us knew how low a heat trap would have to be to effectively stop this radical a gravity circulation, so we took what looked like the safest path. Yeah, I know that flow-control valves add considerably to a circuit’s pressure drop and that they require a secondary circulator with a higher head, but we wanted this one to go away once and for all.

When I was driving home from the job I started to think back a bunch of years to a time when I worked for the Bell & Gossett rep in New York. We used to service everything we sold for the first year. This included bypassing flow-control valves. If a contractor had a problem with gravity circulation, we’d send a serviceman out to fix it. Usually, the problem was a bit of dirt or a lump of solder that got stuck under the flow-control valve’s weighted check. The serviceman would grumble, curse the heating contractor, wipe off the weighted check and put the flow-control valve back together.

But then there were those jobs where things didn’t go as smoothly. The trouble started in the Seventies when boiler manufacturers began to make smaller boilers. This was right after the 1973 OPEC oil embargo. Those little boilers would come up to temperature so quickly that they’d pop the weight inside the flow-control valve like bread in a toaster. The calls came in from the contractors and we’d have to send our serviceman to take a look. It got to be pretty chronic so the engineers at the factory looked into it and came up with an expression that I will never forget. They said the little boilers were having "thermal burps." Isn’t that picturesque?

It seems the rapidly expanding water below the flow-control valve was just too powerful for the brass weight in the flow-control valve. Once our serviceman figured out what was causing the trouble he solved it by removing the cap of the flow-control valve, wrapping solder around the brass check and putting it all back together. That usually did the trick. On jobs where it didn’t, the serviceman installed a special heavy insert we had made up by a local machine shop. These inserts each weighted about 95 pounds (well, almost!). It probably required three-quarters of the circulators total head capacity to open the thing – but it made the gravity circulation go away once and for all.

And so we learn.

You know, if I had it to do all over again, I would rewrite every book and every article I’ve ever penned and this time around I’d strongly suggest that use flow-control valves on both the supply and return sides of every stinking secondary circuit you install.