DIY Home Improvement Forum banner

Water Heater Recirculation Mystery

1153 Views 11 Replies 4 Participants Last post by  beenthere
I'm just a DIY homeowner with limited plumbing skills. Swap out a water heater, shallow well pump install, change taps, install yard irrigation system, sweat copper joints, glue PVC/ABS, crimp some PEX etc .... plenty of experience. But I'm not sure I have every fully understood the recirculation system in my house. I assume it returns ambient water from the farthest point in the house, via the pump that is about two feet from the HW tank, and returns into a fitting on the bottom drain (check valve between pump and bottom of tank). I may be confused about direction of flow, but that's customary, correct? Pump pressure in the drain, and circulation out the standard hot water port?

Anyhow, here's what's going on. Last night I swapped in water heater #3 (electric) since the house was built in 2000. It's a temporary used unit (Craigslist find) since we're going tankless as soon as I can get the electrical and plumbing in order (relocating HW supply, so it will take some work). I configured plumping and electric as it always has been. It was literally plug-and-play, not even a coupling change, even recycled my brass drain valve in lieu of the plastic crap that was on the replacement tank. Here's how the replacement differs from two water heaters I've previously utilized: it's 50 gallons with two 5500W elements vs. the old 80 gallon with two 4500W elements. It was a "smart" unit that had a failed control board. The smart controller was deleted, the heater was converted to conventional operation with non-simultaneous thermostats. Due to the higher wattage, I noticed a remarkable recovery time compared to the old one. I also increased the thermostat setting up to 135 from our traditional 125, in hopes of not running out of HW given the 30 gallon difference. My impression is it will support just as much shower use as the old one. We used to be almost wide open for a comfortable shower on a single handle valve. Now it's 50% or less to get the same mix temperature.

Problem is, zero recirculation. In our two story house, that means there minutes to feel a bit of warmth in the primary bathroom, and about five minutes for a good shower. SO working record is crucial. I noticed the pipe is warm on the inlet side the pump, but not downstream (well, just a bit of warmth on the outflow side for a few inches; I assume it's just basic thermodynamics at work, not really water flow - it's stone cold close to the water tank). I considered the possibility of an air lock somewhere, and thoroughly ran the system. Every HW tap was exercised, and indeed some air was purged. No change. This morning I decided to crack open the line between the pump and tank. Using a bucket, I demonstrated good flow from the pump to the tank. I opened the drain drain valve, and noticed ultra strong flow, more than I have ever observed. But honestly, in the past I have already turned off city water into the tank when I open the drain, so I have no basis for comparison. But it was really strong flow, seemingly more than a household tap.

So as the wheels in my mind spin, I am assuming the tank pressure is somehow overriding the pressure of the recirc pump, or at least closing the check valve. Due that sound like the most likely scenario? If so, why? The obvious changes are tank volume, element wattage, and temperature. Does the higher operating temperature or smaller tank size create the potential for greater tank pressure? Any issues with running a 5500W element on full power? I assume the original smart controller would dole out power in some sort of metered fashion, but at times would ramp up to full wattage. This set up runs 5500W or zero, nothing in between.

My next try will be turning down the temp. Meanwhile, I'd appreciate any thoughts and suggestions.
See less See more
1 - 5 of 12 Posts
Forgot to mention I rapped the check valve with a hammer as my first move before disassembling the line between tank and pump. Disconnected at the tank, the pump line flows nicely right through the check valve. If the check valve is closing, it's likely due to superior pressure from
the tank. The pump plumbing is unaltered, flow arrow toward the tank. It's on a 120V three prong wall plug (no change there), so rotation change is unlikely. Besides, it's nice and warm on the inlet side, just not pushing the return hot water to the tank. Pump wasn't run dry, and literally performed tip-top earlier the same day with the previous water heater installed.

Thanks for the ideas though, all valid and got me double checking. Pretty much settling on significant tank pressure. I just don't understand why!

For all the good advice I have received (all of which I considered), I was still convinced tank pressure was the record line. That is, at the drain valve where my recirc return dumps back in. I was convinced that the primary changes in the WH system had to be a factor; 5500W elements/faster temp rise, smaller tank capacity, and thermostats set to 135 degrees vs. 125 on the old WH. So I decided to change what I could. That being the temperature. Before dialing down to 125, I was in the process of taking voltage readings to ensure proper element function and upper/lower sequencing. The T&P valve happened to blow as I was in the process. I had actually bumped the upper thermostat up to test it, but it was only heating for perhaps a couple of minutes. Maybe it gained a few degrees from the previous shut-off threshold, but were talking in the ballpark of 135 plus a bit. Not 210! I shut the power down to let it cool. The T&P closed swiftly. I then restored power and finished my checks, confirming proper shift to lower element heating. Both stats set to 125. I reconnected the the record line and ran the pump. A few minutes later I felt balanced line temperatures on both sided of the pump. Sure enough, a check at each household tap produced instant hot water.

So was it temperature or pressure that triggered the T&P? Not sure, but I'd guess pressure. I must emphasize that opening the tank drain produced a violent flow, more than any of the household taps, even unrestricted hose bibs. Anecdotally, it seemed greater than line pressure. There are no restrictions on the HW line out of the tank or any of the shut off valves. In fact, flow at any water fixture is the same as always and very acceptable. All I can think is rapid temperature rise and resultant pressure rise in the tank takes too long to equalize throughout the system, and it blows the T&P and probably slams the recirc check valve closed at the same time - and the check valve never bounces back. Evidently 10 degrees less on the thermostats did the trick. Five hours later and two showers plus dinner dishes after my tinkering, and all is well with the recirc system. No more blown T&P. Why, I have no idea whatsoever! But thanks for all the good ideas. As much as I'd like to understand what happened, the objective was to get the thing running. So I'm happy.
See less See more
Sorry, my spell checker foiled me in the above post. Now I can't edit the post due to views I guess. The first sentence was meant to convey that I still believe tank pressure is overpowering the recirc return line. Oops. Apologies.
If the hot water pipe from the water heater to the faucets and the recirc loop have no closed isolation valve in them. Then the pressure in those pipes must be the same as in the water heater.
Any chance a rapid rising temperature/pressure could outpace the time needed to equalize throughout a large two story house? Recall that I am using 22% more wattage and the tank capacity is 38% less compared to the previous install.

I'll call this my "problem solved" post, but the causes remain unknown. After turning the temperature down to 120 (ballpark indicated on thermostat dial, unverified if actual), the issue no longer exists. But if I turn it up to 130 and demand a significant recovery - as opposed to simple maintenance cycling on/off of the element - the symptoms reoccur. That's okay, because 130 is too hot. The only reason I "went there" was an effort to replicate a similar shower capacity to the 80 gallon, which was set at 115 or so. But so far 50 gallons at 120 is working out. I cannot overstate how ferocious the 5500W element is at recovery compared to the 4500! I guess my first full month power bill will reveal if short cycle, higher wattage compares to lower wattage at a longer duration.

Thanks Beenthere and everybody else who assisted.
See less See more
1 - 5 of 12 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.