Gravity hot water system conversion

Gil Carpenter

The gravity hot water system consists of an American Standard boiler with two main loops. One loop has radiators on both the first and second floors while the other loop has radiators on only the first floor. The house recently had the kitchen and upstairs bath remodeled with the installation of kickspace heaters under the counters, which unfortunately do not work well with the gravity system. The heating system was then modified with the installation of a circulator pump in each of the main loops. To simplify this installation, the pumps with integral check valves were located in the return lines to the boiler.

Question #1 The pumps were installed with “reverse flow direction”: water flowing out the boiler return. Are there any serious consequences associated with this configuration? Should the pumps be turned around so that the pumps are pumping into the returns to create a “normal” flow direction??

Question #2 Should the integral check valves be removed in order to reduce the operating cost of the system even though it will likely cause temperature overshoots?? The overshoots have been a systematic characteristic of this gravity system.


Gil

Al Letellier_9

gravity system changeover

I would reverse those pumps ASAP. They should always be on the supply side and pump away, but thats tough to do with gravity piping. Reverse flow will bring cold system water back into the crown of the boiler, which is usually the hottest and thermal shock can become a big issue. As to the check valves, leave them in. I assume that both circs run together? the check valves will help reduce overshoot by preventing gravity flow in the mains.

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Mike T., Swampeast MO

Will Try...

Most gravity systems have two "loops" or main pairs. This both reduced the length (restriction) in the horizontal mains and reduced the pipe size.

ZERO wonder that a kickspace heater did not work via gravity circulation. They typically have quite high restriction to flow even by modern standards.

When converting a gravity system to forced circulation it is very unusual to use two circulators for the radiation piping. Generally, the main pairs are brought together using a simple rule of thumb: The connecting pipes should be 1/2 the main size minus one additional pipe size. For example: If the supply main pairs are 3" you would use 1¼". For 2½" you would use 1".

A SINGLE circulator for the radiation is then installed in the new reduced size common piping. B&G 100 or Taco 007 are most commonly used. The circulator is usually installed on the supply side, sometimes on the return side. Regardless of position, the flow direction MUST be from the return through the supply of the boiler!

-------------------------------------------------------

Your historic overshoot problem could be from different sources.

1) The thermostat could be inappropriate to this sort of system. Many? most? programmable thermostats are not well suited.

2) The thermostat anticipator could be improperly adjusted. True gravity systems tend to give off significant amounts of heat WAY after the burner stops and the anticipator must stop the flame before setpoint is reached.

3) The thermostat could be defective.

The other likely cause of overshoot is reduced flow in the system due to increased restriction. As the pipes age and acquire both scale and pits, their restriction to flow increases. As the restriction increases, a higher temperature is required to produce a given amount of flow. Eventually you're left with a boiler that cycles on high limit during a heat call in less than frigid weather. You're left with a LOT of heat in the boiler and the main pipes in the basement--some of which works its way upstairs after the thermostat is satisfied.

----------------------------------------------------------

Your kickspace heaters may still be a problem.

Without the circulator the problem was insufficient flow.

With the circulator you'll [probably] get decent flow, but supply temperature is likely to be too low. Why? Because the circulator is now moving heat VERY, VERY rapidly compared to gravity. The supply temperature in the mains will be GREATLY lower than it was under gravity.

If the kickspace heaters were VERY generously oversized AND the fan will operate at quite low supply temperatures (often this is called the "low temperature kit"), you might be OK.

Otherwise, you'll need a dedicated system for the kickspace heaters with its own properly sized piping and circulator AND a fairly high temp source of water.

-------------------------------------------------------

In all honesty it sounds to me like you or someone has created a BIG mess with little chance of efficient success. Dan's book "How Come?" would have helped whoever modified this system GREATLY!

Greg Swob

One of the biggest problems with flowing out the return is due to the location of the boiler's high limit aquastat. Located typically at or near the supply outlet, if piped in reverse, the limit device will be sensing return water temps instead of true supply temps. This could lead to interesting moments ie- boiler trying to 'jump' and it's pressure relief letting go during certain moments. I witnessed this on one of the first boiler projects I got to forensically diagnose. It took me a while to decipher the situation, but when a very learned tech looked the system over, he concurred. We reversed the situation and the boiler quit trying to dance. I almost made a back door through the basement wall of the boiler room first time I experienced that Burnham jolting- we laugh about that today.

Frank Eppert

I have a 100 year old house with a gravity heating system using the original coal fired boiler with a gas conversion. It has functioned perfectly for the past 35 years we have lived there. I estimate it is somewhat less than 50% efficient. My plan is to install a new high efficiency boiler, and use the old boiler as a tank. The new boiler will have a small circulator to supply hot water to the "tank". Hot water distribution to the radiators will still be by gravity from the hot water in the "tank". Have I missed anything in my plan?