*
I live in a building with some baffling hot water heating problems. 1) Overnight, the cold water in the building heats up. After long periods of low HW use, if you feel the cold-water inlet pipes to the hot water heaters (2×100 gal) they are scalding hot, two and three feet back from the HW heaters. Cold water from the taps in the building is hot. 2) There is a hot-water re-circulation pump, and the return HW is only luke-warm. 3) The unit furthest away from the HW heaters, complains that their HW in never hot enough.
So I took these clues and came up with the following hypothesis: The reason cold water is hot, is because hot water is migrating back into the cold-water system. The hot-water re-circulation system is responsible. There is an obstruction somewhere in the hot water line before the re-circulation line take-off. Thus, when the re-circulation pump runs, it creates a pressure drop in the hot water line it is tied to. This pressure drop is balanced by a cross feed from the cold-water line (through a shower mixing valve or some other unknown mechanism). The result of all of this is that cold water is being mixed with hot water in the return line, which explains the luke-warm return temperature. This would also explain the backwards flow of hot water into the cold water system: it is being driven by the re-circulation pump. The cold water mixing with the hot water in the hot water feed branch that the return is tied to also explains the low temperature of the hot water reported by the furthest unit.
Is this a half-baked theory, or could it be correct? I’d feel a lot better about it if I knew that shower or sink mixing valves had a failure mode that allowed cross links between the HW and the CW. Would any plumbing experts care to comment on this mystery?
Replies
*
Put a one-way valve in the cold water supply line to the hot water heater. Just before the heater.
*A check valve will keep the cold water from getting hot, but it won't do anything for the unit with warm hot water. I'd like to know if there could be cross-flow between the cold and hot water systems, and if there is, how to find and correct it.
*If there is no check valve, the cold water is getting hot. Where is that hot water coming from ? Now, why do you think there is not enough hot water left to get the warm unit actualy hot ?Yes, there is cross-flow. It's called no check valve.The forest is there, just quit looking for more trees.
*PaulWhen I read your problem, I came to the same conclusion as you. Been in facilities maintenance for a long time and I've seen this before.I doubt that the circulating pump is actually causing the pressure imbalance, anyone who is using the cold water will drop the pressure in the cold water system and let hot water migrate in.Here's a couple of things I've seen before that contributed to this:A hose shut off valve was attached to the faucet of a slop sink, instead of turning off the sink at the faucet handles, both the hot and cold handles were left on all the time and the water was shut off at the hose. Water could migrate through the faucet.Same problem with shower heads that can be turned off at the shower head and are always left on at the shower control knob or knobs.The circulating pump only contributes because it forces the entire plumbing system (hot and cold if there's a cross connection) to homogenize to whatever temperature the hot water heater or mixing valve is shooting for.
*Luka: Unless you're proposing something other than what I think, "Put a one-way valve in the cold water supply line to the hot water heater. Just before the heater."Is a bad idea. Potentially really bad. After hot water use, the HWH fills with cold water. Fine. But then the water in the HWH gets heated and EXPANDS. In typical layouts, it just expands several feet back out the cold feed. If it can't do that, you discover what item has the lowest pressure rating in your house. Hopefully the P&T blows off like it is supposed too, and vents a quart or so after each big hot water use. But I don't want my P&T dumping water on the floor many times a day and would strongly advise against using it as a primary pressure control device, only a back-up safety. If the P&T fails or is blocked than a ruptured HWH tank is possibility.A check valve on the cold water line as it goes to rest of the building (after the branch to the HWH) could solve the problem without as much risk. But those lines are often buried in the walls.Ryan: I assume the next step in the diagnosis would be to (first thing in the morning) feel all the fixtures, looking for one that is warm or hot and therefore the culprit. And/or shut off sections or piping or individual fixtures at night and see if the problem resolves.Can a mixing valve leak a little and cause the same symptoms? I've seen a bit of this (cold water is warm in the morning) after adding a tempering valve in order to get stable temps out of the HWH and to allow cranking the HWH. Both of which worked nicely, but created this other minor issue. -David
*David -You make a good point. Maybe an expansion tank could solve this problem, though.
*Current practice, as I understand it, requires a pressure tank if there's a pressure reducing valve on the main, which has the same effect as the check valve suggested to cure the back flow problem.Or, you could hook up a video camera pointed at the TP valve on the water heater and make a training tape of what it's there for!
*The natural expansion of the cold water when heated is something I never considered. It certainly may explain part of the problem, as the HW heaters are on a short (3 ft.) branch off of the cold water inlet to the building about 4 ft. from the main water shut-off valve.I'd be tempted, in fact, to ascribe most of the problem to expansion except for two experiments. First experiment) each tank has a valve where the re-circulation line goes in. Turning off the return to one tank causes that cold water inlet to stay cold, while the cold water inlet on the remaining tank gets hot. Reversing the valves, reverses the behavior. The Second experiment) shutting off the re-circulation pump for a day. Everyone reported that the heated cold-water had gone away. Of course the unit complaining about luke-warm water went ballistic, claiming it took 10 minutes of running the HW to get luke warm water.
*Paul: I'd look for the types of valves that Ryan indicated can cause the problem. My thoughts about expanding hot water was only as it related to safety and to not install an inlet check valve without an expansion tank.I think your warmed cold water is definitely from the circulation pump plus crossover at some point. Possibly a fixture or possibly the tempering valve. Get up before anyone else and feel the pipe temperatures in the utility room. It may become evident where the hot water is crossing over. -David
*
I live in a building with some baffling hot water heating problems. 1) Overnight, the cold water in the building heats up. After long periods of low HW use, if you feel the cold-water inlet pipes to the hot water heaters (2x100 gal) they are scalding hot, two and three feet back from the HW heaters. Cold water from the taps in the building is hot. 2) There is a hot-water re-circulation pump, and the return HW is only luke-warm. 3) The unit furthest away from the HW heaters, complains that their HW in never hot enough.
So I took these clues and came up with the following hypothesis: The reason cold water is hot, is because hot water is migrating back into the cold-water system. The hot-water re-circulation system is responsible. There is an obstruction somewhere in the hot water line before the re-circulation line take-off. Thus, when the re-circulation pump runs, it creates a pressure drop in the hot water line it is tied to. This pressure drop is balanced by a cross feed from the cold-water line (through a shower mixing valve or some other unknown mechanism). The result of all of this is that cold water is being mixed with hot water in the return line, which explains the luke-warm return temperature. This would also explain the backwards flow of hot water into the cold water system: it is being driven by the re-circulation pump. The cold water mixing with the hot water in the hot water feed branch that the return is tied to also explains the low temperature of the hot water reported by the furthest unit.
Is this a half-baked theory, or could it be correct? I'd feel a lot better about it if I knew that shower or sink mixing valves had a failure mode that allowed cross links between the HW and the CW. Would any plumbing experts care to comment on this mystery?