CGa Boiler Keeps Rusting (Long, long story)
I think I figured out why the boiler keeps rusting, and have corrected it. I'm posting what I did below, and would love to hear any feedback on anything else I should do.
If you don't feel like reading my story below, I believe I fixed the rust problem by moving the hot-water-loop temperature sensor from the side of the loop right AFTER the boiler to the side of the loop right BEFORE the boiler.
I think I fixed the odd impossible to get the temperature right problem by putting the outside air sensor OUTSIDE instead of in the air intake.
The rest of the story, below -
I'm not an HVAC person, but the company's network administrator. Two HVAC companies and a plumbing outfit (who installed the boiler TWICE) have looked at this and couldn't fix it. One of them insisted on a mixing valve being installed that did absolutely nothing (Which makes sense, the Tekmar just sped up the variable speed pump to overcome the mixing valve).
Anyway some background -
- This is the second boiler for this location. The first one rusted out in 2002, at 7 years old.
- Is probably the third chimney - they keep rusting off
- The system is a radiant floor heater, with the floor hot water loop and a boiler hot water loop. A Tekmar mixing controller and variable speed pump seperate the two loops.
- Because the boiler rusts so badly, one of our maintenance staff cleans all the rust off at the beginning of the heating season.
- Adjusting the thermostat up or down takes up to three full days to see any changes in actual room temperature, and once the floor has heated up, a warm winter day can see temperatures above 85 in the room
- There were complaints that the floor got so hot on a warm winter day that nobody could stand on it
So, knowing absolutely nothing about this, I started reviewing the settings in the Tekmar - First problem I found, the system was set up as a radiator system instead of a high mass radiant floor system. It was trying to heat the water flowing through the concrete floor to 150 degrees! I changed this setting to High-Mass Radiant, and the Target went to 90. I *thought* was done - it was a pretty obvious screw up in the configuration. Don't worry, the boiler loop is set to a minimum of 150, and it never gets below 145 when the system is running.
The next part gets really screwy - The system still didn't really work. The floor no longer got hot enough to turn the room into a sauna, but it didn't get hot enough to heat the room, either. At the same time, the boiler was almost always running. So, I checked the controls again. At the time, it was 32 outside. According to the Tekmar, it was 65 F. What the heck? So, I started tracing wires until I found the "outdoor" temperature sensor -- Inside the air intake duct for the boiler's combustion air. It was reading 65 degrees because the boiler had shut off (after the room thermostat was satisfied) and then it shut the outside air damper, and then went into WWSD mode.
I took the outside air sensor and put it OUTSIDE in a shadowed area. Then I watched the Tekmar to see what it was doing - it wanted the water at 88 degrees and was running the boiler. The boiler came on, the injection pump started, and the water temperature was heated to 94 degrees in about 2 minutes. The boiler shut off, and the water temprature almost instantly cools back off to 87 degrees. Boiler starts back up, water is back at 94 in under two minutes. It just didn't seem right, so I went back to tracing wires to see where the sensors were.
Boiler loop sensor - in the middle of the boiler loop, opposite the side with the in and out connections.
Room loop sensor - Attached to the loop right after the hot water pipe from the boiler. Soooo....as soon as the boiler would start injecting hot water in to the loop, it would think the loop was satisfied and the boiler would turn off.
I moved the sensor to the "cool" side of the loop (where it returns from the room), and watched the whole system for a full cycle of on/off/on
This time, the boiler stayed on for a good 20-30 minutes, the loop temperature gradually rose to 94 degrees, then stayed off for 45 minutes as the loop temperature gradually dropped back to 87 degrees, when it started back up again.
And for the first time ever - the room temperature went to 68 degrees, and varied between 67 and 68 degrees the whole day. No more wild 10 degree swings or the heat shutting off because the sun came out...
I'm not sure if I fixed it in time to save the boiler from rusting out, but at least when we replace the boiler again, the new one will have a chance at lasting for 20 years.
What temperature are you maintaining the boiler at . If it is 94F this is too low of a boiler temperature for the boiler and you will find it will start to condense the flue gasses and you will be back to where you started with rotting out the boiler.
The simple answer is its set in a range from 150 to 180.
A more complicated answer - a variable speed pump is in between the boiler and supply loops, if the boiler temp goes below 150 the pump is slowed or stopped until the boiler loop temp goes back over 150.
The boiler is allowed to shut completely down when there is no heating demand, and can cool completely to room temp, but when it lights again it has to reach 150 before the injection pump starts.
Just imagine how many times the boiler was cycling on and off in an hour when the temp sensor was on the supply instead of return pipe! I bet at least 10 if not more.
is the flue correct size and sloped properly?also is the intake air far enough away from the exhaust outside? can rain get into either one?
The intake vent is on the side of the building and comes in through a duct shaped like a U - so no rain should come in that way. The intake is also just a room air intake - its not directly attached to the boiler. The exhaust is on top and has a large rain cap on it --- as for slope and size - I'm a network admin, not a furnace installer So - I took a picture of it! It has two ninety degrees in the exhaust pipe and goes straight up through the roof. The horizontal section is about 2 feet long.
** The exhaust duct diameter is the same all the way through to the rain cap
Rust is formed in a boiler system thru galvantic corrosion which happens two ways , Water Temp & hardness and thru an improper grounded curcuit. Start off with an edy-test and then check your tds - I can garranty one is off. After that make your chiminey stack and fire adjustments.. If all is correct & you have a good balanced system it will last over 50 years.
"Rock-n-Roll " Ain't noise pollution..
I assume that vent is drafting properly,check with a match under mushroom above boiler. looks like water is coming down on cap above boiler,also with high humidity in room can cause rust also. have a tech check burner air intake for proper combustion. this boiler should last u 20+ years if maintained properly. any other fossil fuel burning appliances in room with this? is it in a wet basement,or mechanical room?
The variable speed injection pump failed yesterday and I got to call one of the local plumbing & heating outfits to come replace it. This is the red pump you can see in the top left of the picture, just started squealing.
He had a few (dozen) choice words for whoever did the original install, no isolation valves, anywhere. He put in two valves for the next time the same pump dies. Said if the original install had been done RIGHT (with valves) then it would have been a 30 minute job instead of a 3.5 hour job. He also replaced the air vent on top of the boiler, since as you can see in the picture it was extremely rusted.
I asked for his opinion on moving the sensor, he agreed that it was probably what caused the rust in the first place - since the boiler was short cycling, the flue was never getting hot enough to burn off the condensation gases.
He also said that he has seen many worse boilers - though this is the worst boiler he's ever seen that is actually still in operation. Suggested that when it fails we should consider a CGb to replace it, for higher efficiencies.
its hacked jobs like that which cause the industry to get a bad name. you have a bad set up. most likely the short cycling of the boiler caused the condensation in the chimney, which is acidic and eats up your casting. since this is radiant, there are much better options today thasn what you have in terms of a boiler.
Tekmars are great controls when used properly, they can make your system run efficiently.
However, Weil McLain says their boilers are not to see return water temps of under 130 degrees. If you are rusting sections out, you are condensing flue gasses in the sections as another observant poster pointed out.
You can correct this issue by setting the minimum temperature on the Tekmar to approx 140 degrees, but risk overheating building during milder temperatures if you are not using thermostats or room sensors.
Or, more appropriately, you can install a bypass from the supply to return lines and use a circuit setter valve to maintain a hotter return. This will work without hurting your efficiencies too much.
I'd like to take a little different approach to your situation than offering advice or opinion on what your problems were or what causes a hot water generator to rust - those things have been pretty well covered.
If you're a network administrator and not a mechanic, and you figured all this out with minimal to no help, think what you could do with a little bit of guidance. I'm thoroughly impressed. If you ever get tired of the snow and want to move to a warmer climate, come on down. I may not have a place to put you, but I'll sure try to find one. Folks who'll use their head for something besides a hatrack are few and far between.
Gudtek - This CGa has the Tekmar Mixing Controller on it - It has two temperature settings, one for the boiler loop and one for the supply loop to the zones. It always fires the boiler to 140 no matter what the outside weather is, then it uses its variable speed injection pump to control the supply loop temperature (which it changes based on the outside temp). It slows down the pump if the boiler temp drops below 140, and speeds it up if it goes over 140. Of course, when the supply loop temperature is satisfied, the boiler turns off - so if supply sensor is on the side of the supply line that comes OUT of the boiler instead of the side that goes IN - it pretty much gets satisfied instantly and shuts the boiler back off, leading to the short cycling and flue rusting and all the other fun stuff.
We have another style of Tekmar on our CGb boiler (Part of our Trane system) - it has no seperate boiler loop and the Tekmar has no variable speed output. That Tekmar just controls the boiler temperature based on outside temp, I've seen it running as low as 105 on an extremely cold day, or an extremely hot day (It was trying for 150 on the very cold day, but the boiler just didn't have enough BTU's in it to get that warm).
I should have read the 'long story' more thoroughly before my original reply! Sounds like you did the right thing to correct the issue in part, but without knowing more about the piping I would still be concerned, so please entertain my thoughts here.
Important: where does the return from the floor radiant come back into the boiler loop? If your mixing / injection control 'boiler' sensor is installed between the radiant supply tee and the radiant return tee, then you could still have some issues with rusting this or any other iron boiler.
It would be my belief that the boiler sensor, (in this situation) should be located closest to the boiler inlet in order to sense the actual mixed water temperature entering the boiler. *If* the sensor is located on the boiler loop prior to the radiant floor return, your pump could ramp up because the boiler loop is warm enough, sending slugs of 80ish degree water into the boiler block. Injector pump will not slow down until water has passed through the boiler block back into the hot water loop. Combined with the lag from having a thermistor sensor strapped to a pipe, you will still have issues with condensate despite maintaining a 'hot' loop.
If you were to relocate the sensor closer to the boiler inlet, you could re-program your injector control to slow pump down when boiler sensor temp reached 130 (min return temp). My only remaining concern with this would be that you may not be carrying enough flow through the primary boiler loop in comparison to the radiant loop. When the injector pump ramped up, return temps would fall causing injector to slow back down. It would be much safer for the boiler, but would effectively limit the amount of heat you could put into the radiant loop.
You could circumnavigate this problem by installing a bypass pipe immediately off the boiler supply to the radiant floor return. This would keep the boiler hot and your sensor warm enough to keep the injector pump on.
The best way to fix these issues would be to scrap the injector pump and install an indirect storage tank. Boiler indirectly heats the water in the tank, which would be used for the radiant loop. Cold water never comes back to the boiler, and you have more stored for demand as needed.
Hope I interpretted your problem better this time around! And if you replace the system at any point, get away from a cast iron boiler. Go to a condensing high efficiency unit like the WM Ultra. Could save you another boiler installation down the road!