Solar hot water system pump running too long at end of day. Wasting heat.

[CJV]

My direct (no heat exchanger) solar hot water system pump runs an hour or two longer than I think it should in the late afternoon. Later in the day it continues to run when there is very little sunshine and for one or two hours it seems to be pulling hotter water up to the collector than it is returning to the tank. It doesn’t seem to shut down as early as I think it should and so theoretically seems to be wasting hot water at the end of the day.
Hoping someone might be able to give me some suggestions as to why this happens. All components otherwise seem to work as they should. The pump starts up when it should in the morning and seems to produce lots of hot water thru the day when the sun is shining. I have double checked all components and both sensors resistance tests out as specified and the controller (Steca TR301U) starts and stops when the T1 and T2 temps displayed match the default settings specified in the manual. Water temperature returning from the collector to the tank as measured by feeling the pipe from the collector as it enters the top of the tank seems to roughly match the T1 temperature displayed on the controller at all times of the day so I think it’s giving me good measurements.
The bottom of the tank T2 temperature does warm up but very slowly over the course of the day. In the morning temps are mid 80’sF and will increase up to 5F over the day in the Hawaii winter . Much more in summer. The tank is a top connected older 80 gal solar tank with upper element (rarely needed even in Hawaii winter and hasn’t been powered on for this discussion). It seems the tank remains stratified most days with cooler water staying at the bottom. As mentioned we have lots of hot water for two people so the top of the tank is quite warm without the heating element needing to be powered. With higher usage by more people we will sometimes power it up. All dip tubes have been checked, recently replaced, and deliver cold water to near the bottom of the tank, pull water from near the bottom of the tank up to the pump and on to the collector. Water from the collector is delivered by the dip tube to the middle of the tank 6 inches below the single upper element. (Mid tank)
Depending on the day, most days roughly between 4 -5 pm, comparing the temperature (by hand) of the pipe between the top of the tank and the pump (heading up to the collector) and the temperature of the pipe returning from the collector to the tank, the water moving up to the collector is significantly warmer than the water returning to the tank. The temperature of the returning water seems to match the T1 reading quite well as mentioned so I believe this component is operating correctly. The temp of the water moving up to the pump and then up to the collector is much hotter than the T2 temp displayed.
As mentioned, I’m confident the dip tube delivering water to the pump is intact and pulling cooler water from near the bottom of the tank, so my only explanation of what’s happening is that as water from the lower part of the tank moves up through the dip tube it is being heated by the hotter water in the upper part of the tank. Enough that later in the day when the collector has cooled off the system is sending warm water up and it is being cooled by the collector. T1-T2 differential temps don’t drop enough to shut the system down as soon as it should as the T2 measurement remains low with only slight increase in temp as the day goes on.
Apologies for the long description. So my question is, is this normal behavior for a top connected tank system?
My three thoughts are:
1. This is normal behavior and this is the reason why side connect tanks are available and preferable. In a side connect setup water going to the pump and the connector is pulled directly from the lower part of the side of the tank and does not pass through the hotter water in the upper tank.
2. Perhaps the pump is not moving water up from the bottom of the tank thru the dip tube fast enough to keep it from heating excessively as it passes up thru the upper part of the tank? It does seem to move water fast enough to produce good hot water passing thru the tank on sunny days. Could it be partial blockage, trapped air, or an underpowered pump? (Pump is brand new, replaced as identical to the previous one. This behavior happened with both pumps and didn’t change with the new pump). (Grundfos UP 15-29, bottom of tank to top of collector is ~ 13 ft).
3. The tank is remaining too stratified and I’m not getting enough mixing in the bottom of the tank to bring T2 temps up as the day goes on to make T1-T2 differential low enough to stop the pump at the right time. Should my dip tube from the connector be longer and move heated water lower in the tank to mix things up a bit.
Thanks in advance for your thoughts. Conrad

[Rade]

Good morning! You do not indicate how old your water tank is. I am not an expert in solar hot water, but have had to replace a lot of hot water tanks over the past several decades due to the local water supplies. I would look at option #3 - it sounds like you have a lot of deposits built up in the bottom and sides of the tank and it is interfering with the sensor and causing it to run longer than you expect it to. Depending on the heat source, a tank will last from 5-8 years (oil burning water tank) to 18-20 for gas, upwards of 25 for electric. Don't know if that same consideration applies to a solar hot water tank, but it's the water actually being used in the house where the deposits come from. Well water is notorious for calcium, iron, etc. We use locally treated pond water (BIG pond) for our part of town here in RI and get a lot of rust buildup that has knocked out two oil burning tanks in 17 years. I had well water in Florida and the calcium took out our electric HW tank (well, that and a lightning strike...).

[J.P.M.]

To your thoughts:
1.) This is not expected behavior.
2.) Dip tubes are usually plastic and have low thermal conductivity relative to metal. That, and the relatively short amount of time spent in the tube means that little heat transfer takes place there.
3.) If the flowrate is low, the tank water will most likely be more stratified than if the flowrate is much higheer

Steca makes the (rebranded) controller for SunEarth products. That cut sheet (from SunEarth) for your (same) controller mentions 2 different temp. sensor configurations. One, 1 a probe type, usually gets inserted into a thermowell.
A second type shown is a lug type that gets strapped to the pipe carrying the fluid whose temp. is being measured.

The Steca manual, on the other hand, shows 2 identical probe types but isn't extremely detailed about the temp. sensors or their placement.

Which one do you have, or do you have one of each?
If one of each, which one is at the top of the collector?

2.) What is the flowrate of the water?

3.) Do you have, or can you get any electrical resistance measurements on the sensors using an ohmmeter?

4.) One thing often not known by lots of people (including some installers) is that if sensors for differential controllers are not thermally well insulated (and mostly equally important to the same high level of insulation) and to the equal degrees against thermal losses to the environment, the one with the lesser insulation and/or the one in the colder environment will report a lower temperature than it would if the insulation level were the same as the other sensor.

Assuming the sensors are in the correct locations, the upshot of all of this is that if, for example, the insulation on the tank sensor was a lot less effective than the insulation on the collector sensor, the tank sensor would report a lower electrical resistance to the controller than it would if the insulation were at the same level as the collector sensor insulation and so report a lower temp. This condition would cause the controller to think the tank water to be colder than it is and so keep the pump running longer.

5.) So, do you have probe type, or lug type, and for whichever type you have, are they well insulated, or at least insulated to the same level. I'd also check the electrical line integrities for breaks, loose connections and wire insulation cracks or breaks.

NOMB, but if I understand what you write correctly, the way you describe your setup is different than the way I did mine which is also a direct system.
My direct system takes water from the bottom of the 80 gal. tank and returns it to the top of the tank. The 2 sensors are located as close to the tank outlet and the collector outlet as possible (within ~ 5 cm each) and heavily insulated.

[CJV]

Thanks Rade and JPM for the quick response. I'll respond in order.
The tank in fact was manufactured in 1985, so yes, it's shockingly old (38 years old). All these questions came up while I contemplated replacing it, started paying a little more attention to how it was working, deciding on a new tank and set up, and talking to my plumber for a quote to replace it. He’s a good plumber but not a solar specialist so had some suggestions but not an immediate explanation of what was happening with the old setup. In the process of writing my note above I started to think about a few things to try while I waited for some feedback from the forum, which I’ll mention below. (There are obvious signs of corrosion and possible reduced flow).
To JPM:
My Steca user manual does suggest two slightly different sensors for T1 and T2. Mine are the same for both and to be honest I am not sure if they came with the controller, or whether I ordered them separately a couple years ago. I did remove both, tested for resistance, and at close to room temperature (~68F) got very close to the 1078 ohms suggested on the chart in the manual. Didn’t test at higher temps but I also tried switching the original T2 (lower tank) sensor up to the collector outlet and vice versa, without any change in performance so I assumed they were both pretty good.
The insulation around the T1 sensor at the top of the collector was very poorly insulated, I put a layer of foil backed thin foam insulation topped by a layer of foil backed fiberglass over top. That perhaps helped a bit but didn’t solve the problem. The T2 sensor in the bottom of the tank is clamped onto an area of the tank where there is a thin layer of rust, as mentioned it’s not the flat sensor mentioned in the manual. I expect these two things make a difference in terms of how quickly the sensor responds, but it is fairly well insulated with fiberglass and the cover that is stock with the tank. I was fairly confident I was getting reasonable T2 temps as I drained a bunch of water out of the bottom of the tank and measured that water with a good quality fast read cooking thermometer and the water drained measured very close to the temps shown on the T2 readout.
Flow rate: I’m not sure, but….One of the things I tried yesterday was to make sure the dip tube coming from the collector and entering the tank mid level, was intact, so I opened it up and pulled it out. In the process I noticed the elbow just before the top of the tank lug was halfway blocked by crud. This was cleaned out, I had a dip tube kicking around so I replaced it, cut it a bit longer than the original to see if delivering hot collector water a little lower in the tank might give me a little more mixing. The result was a definite improvement in performance! Over the course of the day my T2 temp came up higher than previous days, even though it was a partly overcast day so I did get a bit more mixing. And monitoring the temp by hand below the pump and above the top of the tank of the to collector pipe, it seemed not to be pushing quite as hot water up to the collector. This could be because the top of the tank never reached as high a temp with the cloud cover but I will repeat this over the next few full sun days. So one explanation is that getting rid of a bit of crud has improved flow and transit time thru the top of the tank. So maybe moving through quicker and not picking up a additional heat passing through. (Theory only!) I’m sure there is crud in other places in the system. The lower dip tube seemed to give me some extra mixing as the T2 temp reading came up higher which would help shut the pump down sooner. The bottom line though, is that the pump shut off at a reasonable time yesterday at the end of the day and as tested by touch, at about the same time that the flow coming down from the collector had cooled to about the temp of the water heading up. So this is an improvement but still not quite correct I think as I have the T1-T2 off differential set at 12F instead of the default 8F set in the ‘hidden’ menu settings. So with the default settings in place for T1-T2 off differential it would have taken longer to shut down the pump and I suspect we’d still be wasting some hot water, but not as much as before.
Intake, return, and sensor positions: If I understand your setup we are both pulling water from near but not right at the bottom of the tank. I am delivering new hot water from the collector to the middle of the tank whereas you are delivering it to the top of the tank? Am I correct in assuming that your tank is top connected with both collector intake and return thru the top of the tank? Not the side of the tank?
Looking at the manufacturers sheets (Ruud and Rheem), they both show for top connected tanks, dip tube locations to deliver hot water from the collector to roughly the middle of the tank, and pull water from the lower part of the tank near the bottom. For side connected tanks the lugs are positioned roughly the same as that. Interestingly my plumber suggested pulling water from the middle of the tank and delivering to the bottom. One of my hopes was to clarify things to make that decision. My tendency would be to listen to the manufacturer, but that said, your sensor position for T2 as close to the actual tank outlet heading to the pump and connector makes extremely good sense. So you know exactly what temps are heading up to the collector. I could easily set that up with the current sensor, a hose clamp, and lots of insulation wrap.
Most stock tanks would have the T2 sensor mount at the bottom of the tank on the outside of the inner layer so you know what’s down there but not what temp is actually going up to the collector. Which is part of my dilemma. A new tank with a clean contact for the lower sensor probably would help but maybe not solve my problem. A side connection set up would probably help a bit as well. Maybe enough to work as expected.
I am interested in the logic for your setup. Thanks a bunch for both sets of suggestions so far. Conrad

[CJV]

So my lingering question/problem is why does the water being pulled from the bottom of tank seem so hot as it exits the tank?
Just occurred to me to wonder, is it possible that there is so much blockage in the system that the pump is struggling and overheating, and actually heating up the water and piping exiting the top of the tank on it's way to the collector? The pump is about 4 inches above the top of the tank.
Conrad

[Rade]

So my lingering question/problem is why does the water being pulled from the bottom of tank seem so hot as it exits the tank?
Just occurred to me to wonder, is it possible that there is so much blockage in the system that the pump is struggling and overheating, and actually heating up the water and piping exiting the top of the tank on it's way to the collector? The pump is about 4 inches above the top of the tank.
Conrad

I would concur with that assessment; the volume of corrosion in the tank is restricting the flow of the water an accurate reading on your sensors..

If you have a plumber that you trust (it sounds like you do), they should be able to step through the proper installation of a new tank with all of the sensors. We recently migrated from a failing oil-burning HW tank (our 3rd in 19 years) to an electric heat-pump that required a bit of re-engineering of our plumbing and electrical. Our plumbing team professed they had only done a handful of these installations, but got it installed to manufactures spec. Our old tank was only 6 years old, one year out of warranty, but when they hauled it out, all you could hear was the sound of the corroded bits and debris sloshing around inside the chamber.

Yes, the new tank cost about double what an eye-for-eye replacement would have cost, but we wanted to migrate off oil heat. This new tank has a 20 year warranty and should have a 50+ year lifespan (according to the manufacturer). Certainly longer than we expect to be here. Where I have seen the tangible ROI was in the reduction in kW usage of our home from this time last year using the old system (and we went from a 30-gal tank to 55). So the investment in the new hardware is going to pay off. If you can see getting 30+ years in a new tank, might just be worth the investment rather than trying to cobble a repair on what you have in front of you.

[J.P.M.]

So my lingering question/problem is why does the water being pulled from the bottom of tank seem so hot as it exits the tank?
Just occurred to me to wonder, is it possible that there is so much blockage in the system that the pump is struggling and overheating, and actually heating up the water and piping exiting the top of the tank on it's way to the collector? The pump is about 4 inches above the top of the tank.
Conrad

It's highly improbable that heat generated by the pump motor, or (related) energy loss (as a result of friction against the water from the pump impeller and perhaps against the tubing walls is responsible for any noticeable heating of the water except on a very local level and only then if there is zero flow or something very close to it.

Reason: Your pump probably has an electrical input of something like 20W or 30W or less. even if all of that energy goes into heating the water it'll amount to less than 100 BTU/hr. or so of heat equivalent going into the water. That amount of thermal energy will heat 80 gal. of water by maybe 0.1 to 0.2 deg. F.

Looks to me like you need some information on the basics of solar thermal water heating systems to understand what you'll read in discussions around here.

Check out a website called "builditsolar.com". Lots of info on the basics. Check out the section on solar heating.
That's one place to start.

[CJV]

Truly, thanks for your time so far. Ok, so cooler water from the bottom of my tank shouldn’t pick up much heat on the way up through the top of my tank and it seems even a struggling pump shouldn’t put much additional heat into it either, but I am still left with relatively hot feeling pipes heading up to my collector. My pump is mounted 3-4 inches above the top of the tank, that area is generally pretty hot, even the tank fittings where the cold supply comes into the tank become fairly warm when there is no water usage and no new cold water entering the tank. Maybe all these things and a very old tank and poor flow are contributing in minute ways to a later than desired shutdown of the pump.
I did spend most of the afternoon looking through builditsolar.com, following as many links as seemed to point to basics, primers, and trouble shooting. Followed 3 or 4 links to azpower.com and Home Power Mag which I hadn't seen before.
I do appreciate the value of a bit of research. And to date hadn’t found any specific explanation on the web of what my system was doing. I did learn a few things today: “ Maintain Temperature Stratification in Your Tank. Hot water returning from the collector should enter the storage tank about a third of the way down from the top. Always supply your collector with the coolest water you have available”, AZSolar
But didn’t find anything else that came as a revelation. My intention from the beginning was to replace the tank and all of the fittings lower down. I'll do what I can to eliminate all of the possibilities discussed. I will make sure the upper piping and collector get a very good flush, and test flow through the collecor to make sure there is no blockage. That and I plan to get my plumber to use the side connections on the tank, pulling colder water directly out the side from the bottom of the tank. This should eliminate the possibility that I’m getting any undesired heating along the way. I may just use the manufacture’s placement for the T2 sensor at the bottom of the tank, but be prepared to move it over to the outlet as it leaves the tank and place the pump higher up and away from that outlet.
Thanks again. Conrad

[LucMan]

How old is your controller?
Electronic components lose their values over time.
Here's the bottom line, if the collector temp drops below the differential setpoint the pump should turn off. You have checked the thermistors and they check out. Water temps check out. Crank up the differential until the pump switches off, your controller may be out of calibration.
If your tank is loaded with mud or minerals your panels are also so you may have to do an acid flush of the system.

[CJV]

How old is your controller?
Electronic components lose their values over time.
Here's the bottom line, if the collector temp drops below the differential setpoint the pump should turn off. You have checked the thermistors and they check out. Water temps check out. Crank up the differential until the pump switches off, your controller may be out of calibration.
If your tank is loaded with mud or minerals your panels are also so you may have to do an acid flush of the system.

Thanks LucMan:
Both the collector and the controller were replaced about two years ago. Just the tank is a real old timer. Water here in Maui is moderately hard, but over two years I'm worried less about minerals than mud from the old tank. Definitely a good flush of the collector needed.
The pump does stop once the collector temp drops down to within 8F by default, the controller starts and stops the pump when the displayed temperatures hit the set points. I did adjust the differential up to 12F which seems to be the maximum it will go. That definitely got the pump to shut down a little earlier in the day but that was still an hour after the water heading for the collector seemed warmer than the water coming back to the tank,
I expect a new tank and setup will eliminate my problem for all the proposed reasons mentioned in the thread so far. Still not 100% sure why the original system behaved like it did if none of my suggested faults were likely. By default I'll assume it's a combination of all of them and try to eliminate all the possibilities in the new setup.
Thanks. Conrad

[LucMan]

You can try installing a T in the bottom of the tank and then screw in a thermostat well, insert your sensor into the well to sense the actual water temperature being circulated instead of the outside of the pipe.

[CJV]

Update on the new solar tank: Just replaced the 38 year old solar tank and the system has been up and running for a couple of days now. All is well. The new system is heating very nicely and the pump is shutting down at the end of the day, later afternoon, as it should. Running the original pump, controller, and sensors with the same settings as the previous system so just the new tank that has changed.
Turns out the problem must have been corrosion and debris in the bottom of the tank that was insulating the lower T2 sensor from the actual water temperature in the bottom of the old tank. As a result the controller was underestimating the water temperature lower in the tank and the T1-T2 differential to shut off the pump was happening way too late, long after the actual water temp in the bottom of the tank was actually hotter than up at the collector.
Thanks to Rade for the discussion of old tanks, LucMan and J.P.M for the suggestions on the placement of the lower sensor. I'm using the manufacturer placement for the lower sensor since I'm getting reasonable temps now, but a placement on the actual outlet pipe makes a lot of sense as well for future consideration.
Conrad

[Andrew Smith]

It sounds like your solar hot water system's pump may be experiencing inefficiencies, possibly due to the design of your top-connected tank system. Considering your observations, it's plausible that a side-connected tank system could offer a more efficient solution by bypassing the hotter water in the upper tank. Alternatively, investigating the pump's flow rate and potential blockages or air traps could provide insight into optimizing its performance. Additionally, extending the dip tube from the connector to facilitate better mixing within the tank might help regulate temperatures more effectively throughout the day.

[peakbagger]

MY SHW uses a DC pump with PV panel. Suns out pump runs. That didnt work so well in the mornign adn afternoon so I eventually went with a simple DC controller with two RTDs. One gets installed at the top of the solar collector on the roof and one on my tank in the basement. It switches the pump on and off based on running the pump if the top of the panel is hotter than a tank. It made a big difference. I got if from a guy in Maine who had a home business building them. They were featured in Home Power magazine and its been running for 20 years. I dont think he is still in business but a slick device.

I found his website but no mention of his controllers Art-Tech dot net