Monoflows use fittings that essentially scoop the water out and redirect its flow. Here is a section of Holohans article that explains the system.
http://www.heatinghelp.com/article/17/Hot-Water/72/Gravity-Hot-Water-HeatingThe old one pipe gravity uses gravity and the fact that hot water rises to effect circulation. [Q: Were there other types of gravity systems?
A: Yes. If the original owner of the home went first-class, he would have installed an overhead gravity system such as this one.
Overhead_Gravity
Q: How does the overhead system differ from the upfeed system?
A: In the overhead system, water goes first to the attic (or to a main suspended from the top floor ceiling) and then feeds down to the radiators. Because this "express riser" is very large, it offers less frictional resistance to the water. As a result, the hot water moves more quickly from the boiler to the radiators than it would in the upfeed system.
Another plus is the way the cooler water pulls the hot water through the radiators as it falls down the return risers. This force counteracts the effects of friction and makes the radiators heat faster. As a result, an overhead system generally costs less to operate.
Q: Is this type of system easier to vent?
A: Yes, much easier. In fact, because of the way the radiators are connected to the mains, you don't need radiator air vents with this system. All the system air vents automatically through the attic tank. It doesn't take long to fill this system either, and you don't have to worry about spilling water all over the floor while venting, as you do with the upfeed system.
Q: How did they pipe the radiators into the mains on this system?
A: They always used top and bottom connections. They could enter the top on one side of the radiator and leave through the bottom on the opposite side, or they could enter and leave through the same side. This second method saved a riser which made for a less-expensive installation.
Q: Didn't they need special fittings to make this work?
A: Yes. They had to divert water through the radiator. To do this, they used a special type of tee. Here's a picture of one.
Q: What did they call this tee?
A: They called it an "O-S" fitting after its inventor, Oliver Slemmer of Cincinnati, Ohio. It was a beautifully simple device.
Q: Is this similar to a "Monoflo" tee?
A: It is, but the O-S preceded the Monoflo by many years. During the 1930s, the Bell & Gossett Company introduced their "Monoflo" tee (the name is a trademark). It went on to play a big part in American house heating during the years before World War II.
Q: Do these special tees "tell" the water where to go?
A: In a sense, they do. They create a path of least resistance for the water and direct it toward the radiator.
Q: Is there any other way of directing the water in this type of system?
A: There are a number of ways, and all of them are critical to the system's operation.
Q: Why is this?
A: Because the pipes in a gravity system are very large and contain a lot of cold water on start-up. Not all of that water is going to get hot at the same time. And since hot water is lighter than cold water, it has a tendency to shoot directly up to the top-floor radiators - just like a hot-air balloon. That's its path of least resistance.
Q: So the top floors tend to heat more quickly than the bottom floors in a gravity system?
A: Yes, and that leads to system imbalance.
Q: How did the old-timers get around this problem?
A: They sometimes added orifice plates to the top-floor radiator hand valves. Here's what one looks like.
Plates
Q: What exactly is an orifice plate?
A: It's a round piece of metal with a small hole drilled through its center. You could make one yourself out of sheet metal; most of the old-timers made their own.
Q: How did the orifice plate direct the water?
A: By increasing the resistance through the radiator it was assigned to. If water found it difficult to enter, say, a top-floor radiator because of the orifice plate, it would go to a radiator on a lower floor instead. In this sense, the orifice plate was similar to the "O-S" and "Monoflo" fitting. The big difference, however, was that instead of directing the water into the radiator it was assigned to, an orifice plate directed the water away from that radiator.
Q: What other methods did the old-timers use to make the water go where it was supposed to go?
A: More often than not, they'd pipe the job in such a way as to avoid the problem in the first place. Here, take another look at this upfeed system.
Upfeed_System
We have three radiators - two on the second floor, one on the first. The hot water's tendency is to race up to the second floor. But look closely at the way the fitter makes his lateral take-offs from the supply main. Notice how the hot water supply to radiator #1 comes off the side of the main. The fitter did it this way because on start-up, the hottest water will be at the top of the supply main.
That hottest water wants to go to radiator #1 but it can't get there right away because the water near the bottom of the horizontal main is colder than the water near the top of the horizontal main. That colder (and heavier) water is crowding the hotter water out of the way and driving it toward radiator #3, which just happens to be on the first floor.
Q: So you can tell from the basement where the risers are going?
A: Yes! They usually fed upper-floor radiators from the side of the main and first floor radiators from the top. That way, the system went into more of a natural balance.
Q: Did they do similar things with their vertical risers?
A: Yes, they did. Frequently, they'd supply a second-floor radiator from the top of the riser and a third-floor radiator from the side of that same riser.
In this case, the second-floor radiator is the lower of the two. That's why it gets the water from the top of the riser.
Q: How about the horizontal mains? Did the old-timers use the same size all the way around the building?
A: Not usually. It was customary to reduce the size of the supply main as it worked its way around the building, but if the fitter reduced the pipe too quickly, flow would stop because there would be too much overall resistance.
Q: What rules did they follow?
A: Generally, they wanted the internal traverse area of the main to meet or exceed the internal traverse area of all the attached radiator hand valves. If the main was too small (or if someone added radiators to an existing main) some radiators wouldn't heat well. The competent fitters sat and calculated every job they worked on. They knew no two were quite the same.
Q: What's internal traverse area?
A: Look down the round end of the pipe. The interior circle at the open end represents the internal traverse area. By using mathematics, you can figure out how many square inches of space there is inside that circle.
Q: Can you give me some examples?
A: Sure! Here's a list of common size pipe used in gravity systems.
Pipe Size Internal Traverse Area
(in Square Inches) Pipe Size Internal Traverse Area
(in Square Inches)
1" 0.86 3-1/2" 9.89
1-1/4" 1.5 4" 12.73
1-1/2" 2.04 5" 19.99
2" 3.36 6" 28.89
2-1/2" 4.78 8" 51.15
3" 7.39
Q: How about the supply and return mains. Do they have to be kept close together?
A: Yes, Ideally, the return main should parallel the supply main within a distance of no more than 8-1/2 inches. It should drop only when it reaches the boiler room.
Q: How did the old-timers bring their returns from their radiators back into their mains?
A: They followed this rule: Returns from first-floor radiators have to enter on the side of the return main because they leave from the top of the supply. This is important because the return from one radiator could block the return from another if the temperatures coming back from the two radiators are slightly different, which they almost always will be.
Q: Were there any special fittings for the mains?
A: They used a number of them. Here are two examples of the more common ones. This is called a Eureka Fitting.
This one was known as the Phelps Single Main Tee.
Notice how the hot water leaves from the top of the fitting while the cold water flows back into the side. Those old-timers were clever, weren't they?
Then we went in and fixed the air controls to the outlying zones - possibly the original problem that was to be corrected by over sizing the pumps - but --- well 
