Evac tube solar water system not working

[baku]

Hi everyone,

We purchased an evac tube based solar water heating system 5 years ago. It's a 50 gallon (I think) system with 24 evac tubes. Rather satisfied with it until now. A few months ago, we noticed one of the evac tubes was cracked. Didn't think much of it until last week when we decided to replace it when the water wasn't hot enough anymore. The 'repairman' claimed that a second evac tube was also cracked and that the filter was clogged (we didn't know the filter had to be changed every 6 months or so). But after replacing the evac tubes and filter, the system only heats water during the day. After 5pm or so (we live in Puerto Rico) the water starts to cool down all the way to room temperature. A second 'repairman' was not able to fix it either. Here's what we can tell:

- Water pressure coming out of the serpentine seems fine, so no apparent blockage.
- Evac tubes will have hot water inside if removed during the day. Since we live in the tropics, unsure if our evac tubes have any liquids inside them like they do in cooler locations.
- Evac tubes show only very slight sediment (slight white stains).
- This is a natural circulation system. Unsure if water is properly circulating.
- Water temp becomes fairly hot during the day. Not as much as in the past but quite a bit. Then around 5pm water starts to cool off all the way to room temperature. Honestly, we are not sure if this change in temp during the day was occurring before changing the filter since we weren't really paying attention to water temp during the day.
- When water is hot during the day, it starts to push outward from the water tank (into a buoyed bin on the side). I don't think this is supposed to happen. Water in the buoy bin may potentially leak out if there's too much water. Also hot water would be replaced with cold water later on. But I didn't notice water actually leaking out so I thought that this may not be a major factor.
- Could air bubbles be the issue? Circulation? Faulty serpentine? Loss insulation in tank? Loss of vacuum in the evac tube? Check valve (toward buoy bin) should be replaced?

Any tips appreciated. Neither hiring more 'repairmen' or buying a new system sound appealing at this stage. Thanks,
BB

[LucMan]

Do you have evacuated tubes with heat pipes ? Or U tubes? I'm quessing that you have a thermosyphon system with U tubes. Either way there should be no water in the glass tubes, if there is you will need to find the leaks and repair if possible and find the reason for failure.
Over pressure? Leaking O rings ? Corrosive water?

[J.P.M.]

Without knowing the actual situation/equipment, it's hard to guess, but, one other thing to check for is reverse thermosiphoning at nite due to a check vavlve that has failed, although I'd find of doubt that would cause all the heated H2O to cool down.

[baku]

Hey guys,
Thanks for your responses. I took some photos of the system. Photo 1 shows the type of evac tube system it is. Photo 3 shows the pipes of the system (yes, I know, when they installed it they used mostly pvc). Furthest to the left you see a small bin. This is where the water that fills the tank comes from. Below it there's a copper pipe with a check valve (Photo 2). But I've noticed the buoy bin is filling up from the tank as the water heats up. Thus, the check valve is not working properly. It doesn't seem that the tank loses a lot of water this way (not much water overflows out of the buoy bin, see overflow pvc in photo 1), yet some heat loss occurs because of this check valve. Not only the water in the buoy bin is cooled down, but other water loss from the tank (evaporated perhaps and leaving the tank through third PVC pipe from left) will lead to replacement with new cold water from buoy bin.

Currently I closed the cutoff to avoid new cold water to fill the tank. In principle, if there's a lot of water loss, the water level inside the tank will be below the serpentine, again leading to no hot water.

We have basically been doing trial-error to pinpoint the issue.
Any feedback/similar experience much appreciated.
Many thanks!

[sanramon]

Interesting and obviously perplexing problem. I live in Costa Rica and it appears I have a similar system. As for the glass tubes - I believe they are just like Thermos bottles - no U-tubes and no heat tubes. Water just sinks along the bottom side of the inner glass tube, gets heated, and rises to the top side of the inner tube, returning to the storage tank. Very simple.

I am curious why your system would have a check valve between the filler tank (you all it a buoy bin) and the main storage tank? In my estimation, water has to flow both ways - from the filler tank to the storage tank when the level gets low in the storage tank. And from the storage tank into the filler tank when hot water in the storage tank expands (or boils as I have experienced). My check valve is in the main water line that feeds the filler tank and the serpentine. I did have a check valve stick open, causing heated water to flow back down the main line, warping the PVC pipe, and causing a union to fail in the house. Is this a possibility with yours?

I assume the second pipe out of the overflow tank is just the overflow draining down onto the roof?

Also curious what is the centre pipe into the main storage tank. I reckon the first on the left is the main water into the serpentine, and the far right is the hot water outlet. Right? But from the photos I cannot determine what the centre pipe does.

Any help so far?

[baku]

Do you have evacuated tubes with heat pipes ? Or U tubes? I'm quessing that you have a thermosyphon system with U tubes. Either way there should be no water in the glass tubes, if there is you will need to find the leaks and repair if possible and find the reason for failure.
Over pressure? Leaking O rings ? Corrosive water?

It's a heat pipe I believe (no U pipe inside). Corrosive water does not seem to be the problem. I will check the O rings tomorrow. Thank you

[baku]

Interesting and obviously perplexing problem. I live in Costa Rica and it appears I have a similar system. As for the glass tubes - I believe they are just like Thermos bottles - no U-tubes and no heat tubes. Water just sinks along the bottom side of the inner glass tube, gets heated, and rises to the top side of the inner tube, returning to the storage tank. Very simple.

I am curious why your system would have a check valve between the filler tank (you all it a buoy bin) and the main storage tank? In my estimation, water has to flow both ways - from the filler tank to the storage tank when the level gets low in the storage tank. And from the storage tank into the filler tank when hot water in the storage tank expands (or boils as I have experienced). My check valve is in the main water line that feeds the filler tank and the serpentine. I did have a check valve stick open, causing heated water to flow back down the main line, warping the PVC pipe, and causing a union to fail in the house. Is this a possibility with yours?

I assume the second pipe out of the overflow tank is just the overflow draining down onto the roof?

Also curious what is the centre pipe into the main storage tank. I reckon the first on the left is the main water into the serpentine, and the far right is the hot water outlet. Right? But from the photos I cannot determine what the centre pipe does.

Any help so far?

I'm trying to solve the problem as quickly as possible. Also not being an expert with this systems today I avoided removing all the evac tubes (I'm not sure if the 1st repairman did this, and the second only removed a few of the 24 evac tubes). Instead I closed off all the cutoffs, only opening them at 4:30pm for testing and then again at 7:30pm for bathing. Any water that came up to the buoy bin was removed to avoid it going back to the system. In the past few days the water was already quite cool at 4:30pm and cold by 7:30pm. Today the water was hot at 4:30pm and pleasant at 7:30 when the cutoffs were open. Unfortunately, this still leaves many possible scenarios:

1. leakage in my hot water system, away from solar system. This candidate is not the problem after inspection of our pipes (and no sign of this problem in our water bill)
2. leakage in the serpentine inside water tank. The water pressure coming out of the tank yesterday seemed fine, leading to believe that the serpentine is fine. But it may be better to look at it directly
3. Circulation issue. Since the water was hot while keeping cutoffs closed I don't see circulation being the main issue. But it could be an issue exacerbated by some other issue.
4. Something else or a combination of things

Sanramon, in photo 3, the first pipe attached to the small buoy bin is where the water that fills the tank comes from. The second pipe provides water to the serpentine. The third pipe is just an overflow and the fourth pipe is the one that distributes hot (optimally) water to the house. The cutoff on the first pipe (to avoid refilling the tank with cold water) and just below the filter (to avoid either leakage from serpentine to cool tank or bleeding water from tank in opposite direction) were the ones closed. Water that pushed to the buoy bin from the water tank was scooped out at around 3pm (it had practically filled the bin and some went out the bin overflow but not much). And GOOD point about the main water line check valve. I think the check valve is there, a bit hard to tell with all the sediment from not replacing the filter a while back. Although the cutoff trials today may suggest this is a problem, I'd be surprised if so much hot water bleeds in the other direction down the main line. Also, when I checked the copper pipe it connects to down below it wasn't hot.

TOMORROW, it appears it is time for me to man up and take the system apart. I plan to first open the buoy cutoff to see if tank need filling. If the water tank does not require water then I will interpret this to mean that the serpentine leaks. I expect the tank to take water (which does not give me any information). The next step will be to take the evac tubes out one by one and check for anything unusual (inside evac tubes, water temp, water tank, o rings, etc). The whole system will be emptied out, cleaned and filled with water again.

On the outside currently all evac tubes look fine. I've been told that if the evac tubes are faulty they should look 'smoky'. Any hints on what to look for is appreciated.

[LucMan]

Problem 1
Looking at your piping there is no way to absorb the expansion of the water in the tank. The check valve on the feed tank would keep the expanding water from entering & relieving excess pressure. Check for leaks on your tank seams, gaskets etc from over pressure. Remove the check valve and the trap and install a t between the 2 tanks, the bull of the t should be piped to a vent to relieve pressure. Repair your leaks.
Problem 2
PVC piping , switch to copper or at least cpvc within the first 10 ft of the tank.

[sanramon]

Yes Lucman, I would agree with your assessment. I tried to post a similar response yesterday, but couldn't. I am surprised Baku says his system worked for some years!

Here is a schematic of my system - which looks like a similar type to Baku's. Yes the pipes enter in different locations on the tank, but the schematic should be clear. The overflow/filler tank could be at the highest point on the system The check valve is in the main water entrance to the serpentine and filler tank (lower right in the diagram).

[sanramon]

Here's another schematic of my system that may help you Baku.

[LucMan]

That's the correct way to pipe the system.

[sanramon]

So, Baku please let us know what you have found in your investigation. We're all curious why your system should cease working if it has been fine up to now. And of course curious why it doesn't seem to be plumbed according to diagrams as posted above.

[baku]

Lucman and Sanramon,
Thanks for your responses. A quick update on the situation.

Sunday I took out all evac tubes and emptied the tank. I then rinsed the tank and evac tubes (some sediment but not major). While rinsing the tank it was confirmed that the evac tubes work fine (they started creating vapor and where hot during cleaning). No leaks whatsoever where found in the system. Unfortunately, after a few days the problem persists: water heats up during day, starts to cool down as the the sun goes down.

Lucman, pardon my ignorance. I now understand that the water should go from the water tank to the assisting (buoy) bin naturally to minimize pressure, and it does. Again, leaks were not found. And I stated by mistake our system uses pvc pipes while it uses cpvc pipes.

Sanramon, excellent diagrams. We certainly do not have an optimally designed system. Ours has no safety valve, for example. I never thought about the level of the tank. Went to check it out, and tank is basically on the level which based on your #2 diagram is not optimal. This may have happened with time. Years ago, you could see vapor coming out of the system air vent and this is not the case anymore. I don't see and easy way to adjust this on my own, the water tank is heavy. However, would that really cool down the water so much?

Funny thing is that after closing the cut off several times since the weekend we've started to hear running water in our main shower when reopening the cutoff, even when the shower handle is closed. So perhaps after all, the problem was the most reasonable guess at the beginning of the situation, a leak somewhere in the house. This possibility had immediately been rejected after inspection of our pipes and no sign of this problem in our water bill (see first point in my march 26 post). We will now need to find if this water running sound is officially a leak (either in the shower wall or floor). But whenever I close the cutoffs of the house the running water sound goes away

[sanramon]

So, Baku it sounds like your worries and work were all caused by a pipe leak somewhere outside the solar water heater? But I'll bet you know a lot more about your solar water heater now?

I have had this happen due to a blocked check valve in the main water feed to the serpentine. When workmen repair the street pipes here (all PVC) plastic chips and sand often get into our water system. Before I knew this I did not have a filter in my line. We were away for 3 months during the hot season, and when the check valve jammed open, expansion of the hot water in the serpentine forced super hot water back down the PVC main water line, causing a glued joint to fail above the kitchen. Luckily all pipes are outside the brick cavity wall!

New CPVC fitting, cleaned check valve, filter installed and no more problems. Good luck.

[baku]

Yes Sanramon, it seems that the leak was the issue. Installed a cutoff on the pipe bringing hot water to the house on Friday. When we closed this cutoff the leak sound in our shower stopped and water remained hot. We've been opening the cutoff just to shower and do laundry, the rest of the time it's closed. This weekend water became really hot just like before, so it's pretty official. Tuesday, we'll have some people break into the wall behind the shower. If the leak is right there then they'll fix it. If the leak is before getting to that wall they'll create a new connection through the outside for the rest of the house to avoid having to break into our bedroom floor.

There's a lesson there. If your solar water heater seems to have major heat loss, one should REALLY make sure there isn't a leak somewhere in the house. No increase in water bill or no sound is not enough as we learned. Better to close and open cuttoffs.

I do have another question. The plumber that installed our electric (back up) water heater 3 years ago suggested that evac tubes heaters heat the water so much that they can cause leaks in the copper pipes. Although this is obviously nonsense (melting point of copper is over 10 times higher than boiling point of water) I do wonder if a bad solder job could lead to a leak when combined with very hot water. Any thoughts on this?