I installed an infloor heating system in 1995 using Watts Radiant Onix tubing throughout the home. The various zones are pumped and utilize mixing valves to regulate the temperature in these loops. The heating source is a cast iron boiler and does have a primary loop. To the best of my knowledge (I've never been contacted for service or maintenance), and according to the customer, the system has always worked fine.
Recently, I was contacted because the system wasn't heating the entire home. After replacing two failed loop pumps to restore heating, I noticed that the boiler aquastat setting had been reduced to 135 deg F by another contractor. I increased it back to 180 deg F, made necessary adjustments to the mixing valves and explained the design of the system to the customer.
Two days later, they called to say the relief valve was discharging water. Replaced the valve and checked system operation, finding that the pressure would increase as the water heated until reaching 30 psi at which time the valve would lift. I assumed a water logged expansion tank and possibly a weeping reducing valve (having used it for purging on the previous visit). Discussed these items with the homeowner and decided to replace both. As added insurance measure and to provide extra capacity, I chose to replace the Amtrol 30 tank (original) with an Amtrol 60. After making these repairs, I started the system and adjusted mixing valves, noting that there was still a minor increase in system pressure, but it held steady at 18 psi and I assumed everything was back on track. Wrong.
Five days later, the customer again called to inform me the relief valve was lifting again and she had secured the system. On arrival, she showed me a video she had taken with her phone - the valve was releasing steam. Started the system up and watched the pressure indeed rise up to 30 psi once again. I suspected now that the primary pump was not delivering adequate flow across the boiler, and noted that the motor section was extremely hot. After replacing the pump with the same performance pump as originally used, I restarted the system and watched the pressure begin to climb above 20 psi again. I also notice that when the gas valve shuts down after reaching setpoint, the temperature in the primary loop continues to rise – if the setting is 180 deg F, the loop will increase 20 degrees before beginning to drop. The primary pump continues to run during any call for heat. I have reduced the temperature in the primary loop to 160 deg F until I can figure out what has happened to the system (the temperature rise in the primary loop after flame shutdown is currently around five degrees).
Help! I seem close to exhausting my bag of tricks and my customer deserves a system that functions without worry. I do have a couple theories of what may be going on here, but don't want to offer anything that may influence anybody's thoughts.
Again, as far as I know, nothing has been changed in the system. Initially I suspected the possibility of antifreeze having been added, but don't believe that has been done.
First lets figure out the heat load for the building. I suspect your boiler is over-sized, as are most. Given the size of the boiler we will endeavor to size a circulator to drive off the heat produced at the "good old American" 20° Delta T. In residential boilers with P/S piping this is usually a small pump on the order of a 007. Expansion is a matter of temperature and volume. It is very rare that a #30 will not cover the 3% expansion volume likely in a 180°F designed system. Most radiant floors use a lower design temperature.
Here, we are not so interested in the supply as the return (you correctly turned the boiler up to protect it from the fatal results of cold return water temperatures and sustained flue condensate) but it sounds like the high limit is turned up too high or is not functioning properly. If we knew what the design temperature should be (the temperature the water must be in order to satisfy the heat load on the coldest day of the year) we would naturally turn the high limit to that setting, say; 160°F. To let the boiler operate at a higher temperature would only serve to burn more fuel than necessary.
Unfortunately the Onix system was promoted as a staple up and insulate sub-floor system and often specified with a high design temperature. The first couple of sub-floor radiant systems I installed were staple-up PB systems--still working--but not ideal.
Look to the zones, as the boiler is clearly "boiling" and unable to get rid of the heat. If there is no load the water in the boiler will change phase regardless of the circulator since it sounds as if the circulator is circulating around the boiler and back.
Turn the high limit down (poor way to control a boiler) and make sure the zone circulators are taking heat away from the primary loop.
More thinking, less parts changing.