help tracking leak in solar loop

[rodvprt]

I have a small leak in my solar thermal loop that I have been unable to track down, and am seeking suggestions.

Here's the deal: 32 Seido 2 series tubes, 20' vertical between storage tank and collectors, 3/4" cooper piping, 50' buried cooling loop, 100 gal storage tank with heat exchanger, a couple of pumps for the heating and cooling loops, expansion tank, check valves, pressure release valve, delta T controller. All pretty standard.

I fill and purge the system, and it'll hold pressure (30 psi) while it's hot. But the next morning it's down to 0 psi, and there's a little air in the system. As long as the sun is shining, everything's fine.

There are no obvious leaks - I never find any water at any of the joints or fittings - I even excavated the cooling loop to check there.

This suggests to me that (one? some? all? of) the compression fittings on the tubes themselves are leaking when they get cold and contract, but at such a low rate that I can never find water beading out. I've tightened them to what should be an appropriate amount, but I'm afraid to really crank down as that'll break them.

Any clever suggestions as how to track this down?

cheers, Roy.

[Naptown]

First place to start is check the expansion tank on the air side. It should have 15-20 pounds of pressure on it when the system pressure is at 0
If not then what could be happening is when things cool down the pressure drops to almost nothing.

Next is make the compression fittings go away and solder all the joints.

Lastly and probably the worst case is to start digging.

[rodvprt]

The tank is already pressurized at 20psi when empty. In any case, I don't believe this to be the problem, as I find air in the system after it's lost pressure, and it had been previously purged, which implies there's a leak somewhere.

I can't solder the compression fittings - this is how the seido evacuated tubes attach to their manifold.

Don't think it's a problem with the cooling loop, as there are only 2 solder joints there, and I've excavated those points and not found any moisture.

I'm 99.99% sure it's the compression fittings between the seido tubes and the manifold. I just can't figure out _which_ ones they are, since they don't leak when it's warm, and the rate of leakage is slow enough that when it's cold I don't see any water coming out. There are no obvious oxidation marks on the pressure fittings to show where a leak might exist.

[MikeSolar]

The tank is already pressurized at 20psi when empty. In any case, I don't believe this to be the problem, as I find air in the system after it's lost pressure, and it had been previously purged, which implies there's a leak somewhere.

I can't solder the compression fittings - this is how the seido evacuated tubes attach to their manifold.

Don't think it's a problem with the cooling loop, as there are only 2 solder joints there, and I've excavated those points and not found any moisture.

I'm 99.99% sure it's the compression fittings between the seido tubes and the manifold. I just can't figure out _which_ ones they are, since they don't leak when it's warm, and the rate of leakage is slow enough that when it's cold I don't see any water coming out. There are no obvious oxidation marks on the pressure fittings to show where a leak might exist.

My guess is that you didn't get all the air out. Put an auto air vent on the hot side coming down with a ball valve and fill up to 40-45psi. Run it for a while and see what happens.

[rodvprt]

My guess is that you didn't get all the air out. Put an auto air vent on the hot side coming down with a ball valve and fill up to 40-45psi. Run it for a while and see what happens.

I don't think it's that either - Seido 2 tubes are notoriously hard to purge, so what I do is circulate water through the system with an external pump and include an open bucket in the loop (just upstream of the pump). Any air in the system bubbles to the surface as water is sent into the bucket, and the pump feeds from the bottom. I run like this for about 20 minutes, long after no more bubble are seen. There's also an air trap/extractor in the system, which has no air in it at the end of the first day when it's still hot. (I open the extractor a number of times when I purge the system to make sure no air is trapped in it.) By the next morning though, it has air.

[MikeSolar]

I don't think it's that either - Seido 2 tubes are notoriously hard to purge, so what I do is circulate water through the system with an external pump and include an open bucket in the loop (just upstream of the pump). Any air in the system bubbles to the surface as water is sent into the bucket, and the pump feeds from the bottom. I run like this for about 20 minutes, long after no more bubble are seen. There's also an air trap/extractor in the system, which has no air in it at the end of the first day when it's still hot. (I open the extractor a number of times when I purge the system to make sure no air is trapped in it.) By the next morning though, it has air.

Yes, this is the standard way to fill a system, I have a jet pump and pressure bypass fitted to the pump for different fill pressures.

Are you sure the exp tank is set for 20psi and not 12psi which is more normal. What size exp tank is it and how far from the pump to the top of the panels?

[rodvprt]

Yes, this is the standard way to fill a system, I have a jet pump and pressure bypass fitted to the pump for different fill pressures.

Are you sure the exp tank is set for 20psi and not 12psi which is more normal. What size exp tank is it and how far from the pump to the top of the panels?

Yes - I pressurized it to 20. It's a 2.1 gal tank, and it's there's about 25' of 3/4" pipe between the pump and the panels. After I purge the system, I close the exhaust valve, and pressurize the system to 15 psi (it's cold at this point). When hot, it'll climb to about 35 psi.

[MikeSolar]

Yes - I pressurized it to 20. It's a 2.1 gal tank, and it's there's about 25' of 3/4" pipe between the pump and the panels. After I purge the system, I close the exhaust valve, and pressurize the system to 15 psi (it's cold at this point). When hot, it'll climb to about 35 psi.

Now we are getting somewhere. First the tank is too small. You should have no more than 5psi difference between cold and hot pressures, otherwise the diaphram is moving a lot inside the tank. I will bet that there are times at night when the the panels are at a partial vacuum and air can come in to the system through the compression fittings. Remember that when it says 0psi at the gauge, it will be less at the panels.

[rodvprt]

Now we are getting somewhere. First the tank is too small. You should have no more than 5psi difference between cold and hot pressures, otherwise the diaphram is moving a lot inside the tank. I will bet that there are times at night when the the panels are at a partial vacuum and air can come in to the system through the compression fittings. Remember that when it says 0psi at the gauge, it will be less at the panels.

Thanks. I'll try upsizing to a 5 gal, and see how that works. Any suggestions for makes? I originally had an amtrol, which had a steel fitting that corroded. Replaced it with a watts DET-5, which has a brass connection. It's hard to find a model that has a high temp rating, and a brass or stainless fitting.

[MikeSolar]

Thanks. I'll try upsizing to a 5 gal, and see how that works. Any suggestions for makes? I originally had an amtrol, which had a steel fitting that corroded. Replaced it with a watts DET-5, which has a brass connection. It's hard to find a model that has a high temp rating, and a brass or stainless fitting.

Try caleffi.com for parts or any Viessmann boiler dealer. Both buy tanks made for solar from Cimm or Zilmet in Italy. They have the best products. 110C is about the highest you will find and remember that if the tank can hang down a couple of feet below the fitting tee, the temps the tank will see will be much less.

[rodvprt]

Try caleffi.com for parts or any Viessmann boiler dealer. Both buy tanks made for solar from Cimm or Zilmet in Italy. They have the best products. 110C is about the highest you will find and remember that if the tank can hang down a couple of feet below the fitting tee, the temps the tank will see will be much less.

Though I like the Caleffi stuff (I have a flow meter from the, and really wish I had bought their solar pump module when I build my system), the problem is that the connection is 3/4" straight thread, and it's a real pain trying to find converters to NPT.

[MikeSolar]

Though I like the Caleffi stuff (I have a flow meter from the, and really wish I had bought their solar pump module when I build my system), the problem is that the connection is 3/4" straight thread, and it's a real pain trying to find converters to NPT.

It is a BSPP thread and it took me quite a while to find them but i now have a couple of sources in Europe. You don't have to have the high temp one if your tee is on the cool side of the HX and you have a long vertical tube to the Exp tank which could sit close to the floor. Chances are, if you could find a Watts #60 at the least or #90 tank, it won't get near its max temp and all will be good and a normal 3/4" FIP will fit (use pipe dope, no teflon tape)