"It's making that sound like when you blow into a bottle of pop. Only it's a lot louder. Whenever the circulator starts, the noise starts too. It's driving the customer nuts and I don't know what to do."
"What have you done so far?"
"I've changed the circulator a couple of times. I've tried different types of circulators. Nothing seems to help much. The high-speed circulators make more noise than the low-speed circulators. I've got a feeling that the return lines may be too small. I'm thinking about repiping the returns but I don't know what size to use. That's why I'm calling. What size should I use?"
It sure sounded like velocity noise to me, and it might be the returns but first I asked him if he had removed the guts from the thermostatic radiator traps. He told me that he hadn't and I asked him why not. "Because there aren't any radiator traps," he said. "There are just these convectors. There are no traps at all," he said, and that's when I knew what was wrong.
But before I tell you about it, let me back up a bit and fill you in on some of the things you have to watch out for when you're converting a steam system to a hot water system.
Before the 1920s rolled around, most contractors used cast-iron column radiators. Those are the freestanding ones with the real wide sections. Column radiators were perfect for one-pipe steam heating because they had a lot of internal space. They allowed the steam to rise up and displace the heavier air. The air then left the radiator through an air vent that was mounted low on the radiator, on the side opposite the inlet valve.
Two-pipe systems were easier to control (and nowadays, to convert) than one-pipe systems. You can throttle the steam as it enters a two-pipe radiator (you couldn't do this with a one-pipe radiator). With two-pipe, the air usually escapes through a steam trap or some Vapor gizmo, and leaves the system through a big air vent near the end of dry return, down in the basement.
During the 1920s, contractors began to use cast-iron, tube radiators (the ones with the thin sections). These were actually hot water radiators, but they also worked well on steam systems. Building owners preferred them because they were nicer to look at. As time went by, the Dead Men also began to use convectors, and this is where things can get interesting.
Converting from two-pipe steam to hot water is usually pretty straightforward. You have to look things over and make sure the piping is sound enough to hold both the static pressure and the pump pressure that's normal in a hot water system. A steam system that didn't leak with 1-psi steam pressure in the pipes may change its mind when you hit it with 12-psi static pressure, plus the pump's differential pressure (assuming you're pumping away from the compression tank).
You also have to make sure that the radiators or convectors can overcome the building's heat loss on the design day with the relatively cooler water inside them (compared to the steam, that is). To be sure, do a heat-loss calculation on the building and then survey the radiators. One-psi steam will bring the surface temperature of a radiator up to 215 degrees and put out 240 BTUH per square foot EDR. Hot water, averaging 170 degrees, will put out just 150 BTUH per square foot EDR. For instance, if you have a five-tube, 26" high, 10-section, cast-iron radiator running on steam, it will emit 8,400 BTUH. Convert the system to 180- degree hot water and you'll get only 5,250 BTUH out of that radiator. Is that enough? Only a heat-loss calculation will tell you for sure.
If everything looks good, you'll move on. Remove the innards from all the thermostatic radiator traps and the float & thermostatic traps near the ends of the main (if you have them) so the water will be able to flow freely and not whistle. And keep in mind that a given quantity of steam takes up about 1,700 times the space as the water that generated it, so there's always going to be a big difference in size between the supply and return lines in a steam system. You can leave the old radiator supply valves in place as long as they're working. You'll be able to use them for balancing the water flow once you have the system up and running. Most two-pipe steam systems are similar to two-pipe, direct-return hot water systems. The big difference is that, unlike a hot water system, the return line doesn't increase much in size as it returns to the boiler. With steam, the lines are picking up condensate, not the much greater flow of water you expect to see in a hot water system. Because of this, you may have to redo the return lines to accommodate the greater flow you'll be seeing once you've converted the system, but in a small building, this usually isn't necessary.
Pipe the new hot water boiler the way you normally would. Size your circulator based on your heat loss calculation, not on the existing radiation. Use a good air separator, and pay close attention when you're sizing the compression tank. There's going to be a lot more water in those old steam pipes than you'd expect to find in a modern hot water system of the same BTUH rating. This is important because you have to base the size of the compression tank on the system's total water volume. Calculate how much water you have in the system, and then talk with the folks who sell the compression tanks. Don't depend on their quick-sizing charts in this case. They base those charts on standard hot water systems, not steam conversions.
"Is there a name on the radiators?" I asked him.
"Yes, it says Trane," he said.
Trane was once at the forefront of Vapor heating in America. If you call them today and ask them about this they won't know what you're talking about, but I have their old literature and I know that they had a special way of doing things in the old days.
"Those radiators have orifices in them," I said.
"No way!" he said.
"Way," I said. "If there are no steam traps, there are orifices. There has never been a two-pipe steam system that didn't have some way of stopping the steam from moving through the radiator. If there's no trap then there's an orifice or some other sort of Vapor device in there. And Trane was a big user of orifices."
"But I took one of those convectors apart," he said. "There was nothing in the supply or return elbow."
"The orifice is inside the convector element," I said. "It's a part of the unit. It's cast right into it."
"Ooooo," he said. I could feel his pain.
I faxed him Trane's old literature that explained in glorious detail how they buried this tiny orifice deep inside the convector to keep the steam from getting through to the return. "That's what's making the whistling noise," I said.
"You mean I gotta replace all these convectors?" he said. "I didn't figure that into my price!"
"Well, you either have to replace the convectors or you have to drill out the orifices - if you can get at them," I said.
"How big a hole do I need to drill?" he asked.
"Big enough to make the whistling stop," I said.
"How big is that?"
"Beats me," I said. "Trane never meant for you to run those convectors on hot water."
And we went around and around with this for a while. I didn't hear from him again so I figured he got lucky with the drill and got paid for the job. I just hope his story makes you cautious the next time you're putting together a price on a steam-conversion job. If you don't see traps, know for certain that there are orifices in there. And an orifice will whistle.
And please don't get mad at the Dead Men. They had no idea you were going to make these changes to their fine work.