boilerman said:
Goose,
How would one plumb in a separate supply/return line to each radiator from a single pump? Would one use diverter T's? I ask because I am thinking about how to plumb a similar problem. I was hoping not to have the circulator run all the time, but switching on whenever one of the zones called for it. More details, please!
Will try, but remember I don't do this stuff for a living - some of our pro-members can probably give better details, but this is the overview as I understand it - If I get something wrong I hope they'll tell me!
What is fairly common these days is to do a "home run" system where you have a pair of manifolds, one each for supply and return, and a pair of fairly small bore (3/8" or 1/2") PEX lines going to each rad. If you had a sprawling sort of setup, you could have a hybrid setup with "sub-manifolds" fed from the main manifolds, which would cut down on the number of trunks needing to get all the way back to the boiler room. In an older setup you might have a supply and return pipe running around the house, with the rads pumped across them. In all cases, each rad would have it's own TRV, which would be separately set and control that rad regardless of what the other rads in the system are doing (If you had multiple rads in the same space, you could possibly plumb them in series and control them with one TRV)
While there are definite differences that need to be looked at when doing the system design, these are all essentially the same from a functional standpoint - think of it as being like a ladder, with one side rail being the supply, the other being the return, and the rungs each being a rad with a TRV.
I'm not sure if you could do TRV's on a one pipe diverter-T system or not - in theory it might work, but I don't know how much pressure a TRV needs across it to work, and if you can get that much from a diverter-T setup. You would also have problems in that the flow volume and temperature reaching the downstream emitters would be constantly changing as the valves on the upstream units opened and closed - I don't know what sort of issues this might cause.
What WON'T work is to have all the rads in one string, like you will often see with baseboard systems, where there is one pipe going from one emitter to the next... Think of the difference between the old fashioned series Christmas tree lights where one bad bulb would take out the entire string, vs. the modern parallel style. If you have rads in a series, any valve closing would block flow through all of them, obviously no good...
Given the ladder model that I mentioned above, you would have the single pump pushing into the supply rail. The system "head pressure" will obviously be constantly changing as the TRV's open and close, which would cause problems for a traditional constant speed pump, as it would be flowing way to much volume if most of the valves were closed, causing flow noise and other problems in the valves that stayed open, or possibly not enough if all the valves were open. Since the valves are mechanical, and don't "talk" to the pump, the pump has to be on 24/7 to feed whatever valves are open at a particular moment... The excess flow problem can be solved by putting in a pressure bypass valve, but this and the always on pump both waste a good bit of energy...
The modern approach, and what was being suggested above, is one of the new ECM pumps that are variable speed, controlled by a pressure sensor. The pressure sensor tries to adjust the pump to always keep the same pressure on the supply line. As valves close, the head resistance goes up, raising the pressure, so the sensor slows the pump down to a lower volume, which drops the pressure down... As valves open, head resistance goes down, the pressure drops, and the sensor speeds the pump up to restore the pressure by increasing the flow volume... If all the valves are closed the pump is barely turning over and draws something like 10W, which isn't much more than a nightlight. At full blast, they supposedly draw 60-80W, about the same as most standard single speed circs...
The argument is that even though the pump is running 24/7, it is still saving energy, because most of the time the pump is going to be ramped down to some intermediate power due to some or all of the valves being closed, so that it might only draw about 40W on average - In that case it would draw the same juice per day as a standard 80W pump that was running 12 hours / day feeding a zone valve system...
Gooserider
