I am installing an active open loop solar water heating system. The tank has a side connect to attach the return line from the collectors. But, the installer insists that connecting the return line to the drain at the bottom of the tank is better. My instinct says that he is wrong, because you are sending hot water that just came from the collector back to the collector (the connection to the collectors is only a few inches away) which isn't very efficient. Also, the only way for the hot water to reach the top of the tank is through a slow convective process, unless you use a lot of hot water, pulling the water up from the bottom. His claim is that with a side connect, the hot water doesn't reach the bottom and so you are only using half the tank. But, I think that is wrong, because the pump is forcing the hot water down. Does anyone know of any studies or evidence that a bottom connection is better? It seems that if it were, the manufacturer of the tank would have moved the "From Collector" connection.

Got a manufacturer & model name for this tank?

How about a system schematic?

SunEarthInc SU120-1 The coldwater and hotwater connections are on top. Look at the picture for side connect and you can see how close the drain and to collector connections are. http://sunearthinc.com/wp-content/uploads/155207SunEarthFlyer.pdf

Find a different installer- this guy is CRAZY, and your instincts are fine.

By reversing the flow on the collector loop you inject warmer water into the collector, raising it's operating temp, which lowers it's net efficiency.

You further the insult by mixing the coldest water with the hottest water- the return from the collector.

Plumbed as per the manufacturer's instructions, when hot water is actually being drawn the collector temp at it's input will drop to near that of the cold water stream, which is a big boost in efficiency. When idling, there will be a lot of convection between the middle-port return and the top of the tank, since returning water will be 10s of degrees higher than the top of the tank. There will be some mixing of the hot and cold at the bottom port from the pump flow, but it'll be a lower temp than the hot-half of the tank. Reverse the flow, and the whole tank will convect, with absolutely no chance at all of getting any stratification advantage out of it.

Connecting the collector FEED to the drain might be slightly or theoretically better than at the manufacturers designated feed port but not enough to matter. Perhaps that's what he meant, rather than the return? The designer likely placed the collector feed port optimally for the shortest path between the bottom of the cold-water dip tube, and taking the feed from the drain might end up taking in warmer water than necessary during active hot water draws but would yield similar results when no water is being drawn.

Unless the guy has actually done a lot of field-measuring at different points in the tank under different conditions to prove his thesis, going with the manufacturers' recommendations are probably the right thing to do. But reversing the flow to inject the collector output at the bottom of the tank would be the wrong thing to do from an efficiency point of view.

No, the FEED is connected to the correct spot and the RETURN is connected to the drain. The side connector that has the "From Collector" sticker on it is just stubbed off. Even worse is that the drain is lower than the feed.

I talked with a second installer and he agreed with my installer, that nearly every installer in Hawaii connects the return to the drain port. Many installers use a dip tube to create some separation between the return and the feed, my installer doesn't believe in them. While the installers agree that the system may run less efficiently, the rationale has to do with night time heat loss. The claim is that side connect return lines stayed hot at night and that even with check valves (that they claim to stop functioning in on a year or so), the heat traveled up the return line and radiated out the panels. Having the return at the bottom means that the line is cold at night (our backups are off during the night) and heat doesn't travel up the pipe. What do you guys think? Do you see much heat loss with side connects? Our tanks and panels are sized to heat water to well over 140 degrees on a sunny day and provide 100% of the hot water without the backup ever kicking in.

Posted By kukae on 27 Oct 2010 05:28 PM
No, the FEED is connected to the correct spot and the RETURN is connected to the drain. The side connector that has the "From Collector" sticker on it is just stubbed off. Even worse is that the drain is lower than the feed.

With the feed & return at nearly the same spot, it minimizes the mixing of the upper part of the tank, but still promotes higher whole-tank convection and a higher operating temp at the collector. I'd be very surprised if it operated more efficiently that way.

Hi Kukae, very glad to see you are tapping into the sun, especially given where you live.  I am a homeowner who had a solar thermal system installed last year, but did a lot of research before committing, so take my words as you will.  A few issues.

First off, the reason why the regular suggestion from Sun Earth is to have the return from the collector be on the higher port is an attempt at promoting tank stratification.  You are correct that injecting hot water at the very bottom of the tank (below a colder layer) will lead to inefficiencies in collection.  Basically, the mixing or turbulence as the hot water rises through the colder layer unnecessarily cools the hot water your collector worked so hard to capture.

Second, check valves are indeed possibly prone to failure.  A less failure-prone solution to this problem is to put a raised loop (like an inverted "U") of pipe in line between both hot and cold ports and the rest of the system.  The physical idea behind this arrangement is that hot water will flow to the top of the U-bend and, unless you have a pumped flow or draw of some sort, it will keep the hot water from migrating out from where it's supposed to go.  Way more reliable than check valves, though it seems that people do have success as long as you don't get cheap ones.  The inverted-U has the benefit of no moving parts, but does require some thought about the relative geometry of the tank vs. the piping and the rest of the system, and you do want to ensure that all the piping is insulated.  (but wouldn't you do that anyway to protect your precious solar water?)  One also needs to be careful that you don't put so much of a kink in the U that you get excessive frictional losses in regular flow.  Also, if there is enough flow or potential built up, the U can also be defeated in a way that check valves would not.  So, YMMV.

Here is a link to a site which shows a picture of such a heat trap: http://www.r2000manitoba.com/hot_water_tanks.shtml

Finally, although your installer has a valid point that heat can get sucked out of the tank at night (unless you have a check valve or other U), the real problem is that you have liquid in the collector in the first place.  This is a design limitation of open-loop thermosyphon designs.  The only ways to really mitigate this are to either put some form of check valve (or U), or else to place the tank higher than the top of the collector.  The latter strategy makes it so that when the water cools in the collector, it sinks and stays there, as opposed to the case where the collector is much higher than the tank, where the cold collector water sinks through the piping and pushes hot water back out of the tank to cool some more.  If you look at commercial designs such as SolaHart with an integrated tank, you'll see that the tank is above the collector for precisely this reason.

Arguably a better way to deal with the problem is to go to an active drainback design.  Although this raises the system complexity, you also don't have to deal with the water cooling issue at all if your house geometry makes it so that mounting the solar collection tank above the collector difficult.  The sources I looked into all suggest that drainback is generally most reliable and least maintenance, as one does not have to worry as much about draining and replacing coolant annually (as is the case for propylene glycol active systems; drainback uses  water as the medium), and physically isolating the solar thermal water from the potable water just gives me a lot less concern about contaminants and leakages in the solar thermal system affecting my DHW supply.  Also, drainback systems have the advantage that you don't need to worry about heat dumping (when you get more sun than you use DHW), nor freezing (probably not an issue in Hawaii, but is a concern where I live in Vancouver).  The downside is some inefficiency due to using a heat exchanger coil inside the tank and, of course, the cost issue as there are more bits involved as compared with thermosiphon.

For where I live, I chose the drainback design because of freeze protection and low maintenance (thermosyphon would not work here at all), but you're in a different climate.  For what it's worth, though, the source which really convinced me to go drainback ("Solar Hot Water Systems" by Tom Lane) was by an installer with over 30 years' experience working out of Florida, which has a similar climate to where you live.

A good resource online is: http://www.builditsolar.com/Projects/WaterHeating/water_heating.htm

Hi Everyone,

Thanks For Sharing about solar water heating systems .

Hiiii
Solar is a kind of solar water heating systems that the bellow collector directly connect with the above water tank. This solar water heaters do not need to run the pump to force and do not require pipeline connections between the collector and the tank. This is a simple and secure way to collect solar energy. This working principle of solar water heater is make it natural circulation between the water tank and the collector pipe. After the water in vacuum tube heated, natural convect into the water tank, meanwhile cold water in the tank flows down to the vacuum tubes, that is natural to run continuously to heat the water tank.

Solar material:
1. Outer tank material: Powder coated color steel
2. Inter tank material: SUS304-2B stainless steel
3. Insulation: 55mm thickness polyurethane foam
4. Frame material: Zinc-plated with plastic spray coating
5. Vacuum tube material: AL-N-AL absorb coating
6. Seal material: Stabilized High Temperature Silicon
7. Assistant tank capacity: 2L, 5L, 10L, 30L, 50L, 100L, 150L etc

Thanks,
Regards
-Michelle

Reducing stratification increases the effective size of the tank, ie, it stores more heat. He is right that this has some benefits. But if you run out of hot water (vs it not getting hot enough), then I would just use a larger tank instead of reducing the collector efficiency with his connection method. (Yes, I realize that the install is long past).

Been doing solar thermal systems for over 30 years. Tapping the drain fitting for a return from the collector is by no means correct. You will reduce the tank efficiency and it will have a tendency to fool the storage tank sensor.By the way the tank mentioned is manufactured by Rheem.

The inverted loop design is a great way to reduce thermal siphoning. The technical name would be a Hartford Loop.

Wow, I forgot that I even posted this. So, based on study #3 at
http://oldweb.hawaiirdp.org/hetl/hetlschedule.htm
which showed that connecting to the drain did not show a significant loss in performance, I allowed the installer to proceed with the understanding that he would switch it to a side connect if I wasn't satisfied. Granted, the study was not perfect, but it was good enough that I was willing to give it a go. The study was done in Hawaii climate and so may not be broadly applicable. Also, the study didn't really focus solely on cloudy/rainy days, where efficiency is most critical. Most of the arguments made by the installer made sense. The check valve probably would be a point of failure at some point and eliminating it would create a robust system with almost no maintenance other than replacement of the tank anodes over 25+ years. The inverted loop, while usually preventing thermo siphoning, is not perfect. I heard from several anecdotal sources that at times, the tank would still lose heat without a check valve (you can tell by feeling the temperature of the pipes at night). In fact, study #5 showed that even with a check valve you lose heat. In any case, once the bottom connection was installed, I began monitoring the system performance. Basically, on most normal days I m getting a full tank (120 gal) of 140+ degree water by noon (the backup thermostat is set to 100 degrees, so that is its lowest starting temperature in the morning). On rainy days the system does not perform as well, but I have no way of telling whether the side connect would have performed better. On average, it has been saving me close to $100 per month which computes to less than 2 year payback. The system probably is providing around 70-80% of my total hot water (the house feed has a thermal mixing valve so when the tank is over 120 degrees, it is not being used up as quickly). Basically, what I'm saying is, I can't complain even if it isn't as efficient as it could be.

Depends on whether your strategy is to store the heat from the sunny day for the cloudy/rainy day or try and make the collectors efficient enough to work even on the cloudy day. Different designs for different goals. A two tank system (solar preheat and then a standard water heater of some type) is more efficient.

I find it surprising that a spring loaded check valve would leak.

Spring loaded check valves can and do leak.The cause is usually do to mineral deposits forming on the seat of the valve. The other cause is wrong selection of check valve for temperature application.