Drain Back Pump air locking?

[Tom Sullivan]

I've had a nagging problem with my solar collector circulation pump sporadically air locking.  I thought it was from the system restarting after a shutdown (partly cloudy days), before drainback was complete.  I could "cycle" the power to the pump and usually get it going fine.  I also was concerned the pump was too high and the suction line was too small (3/4" in and out of pump).  My suction does go through the side wall of the tank, so the pump should clearly be in prime at all times.

Recently I drained the tank to install heating elements for my newly restored Jacobs, and decided to address a couple concerns with the pump.  I installed a 1" suction line to the pump, and lowered the pump as much as possible at this time too.  That seemed to help a bit, but the occasional air lock (at start up) continued.  Now, figuring air in the line during drainback with my controls reactivating the pump to quickly, as my problem, I took a delay timer out of the control panel of my old Breezy turbine and installed it in the wiring to the pump (after the GL-30 controller).  I set the timer at 8 minutes and my array only takes about 2 minutes to complete the drain back.  This seemed to help for a bit, but the same problem was noticed again yesterday.  I'm using the Taco 0013-F3 pump, have about 23' of head, on a 336 sq' array.  There's more info on the system on my website
 www.aluminum-solar-absorbers.com , under the "large solar collector" tab.  Hope someone can give me some guidance??

[jarrod9155]

   Tom , the guy I purchased my flat solar panel from had a similar problem during drain back or after he was getting air locked also . He added a 1 inch check valve that flowed in the discharge direction and drilled a 3/8 hole in it so it would drain back still . He also used a u pipe not thru hole like you have but might be worth a try .
                                                                                                                                                                   Jarrod

[GaryGary]

Hi Tom,
I don't really have any good suggestions -- sounds like you've done all the reasonable things -- but here are a couple out there thoughts -- the kind of things to try when the reasonable stuff does not work

Just as a diagnostic idea, one thought would be to slow down the drain back and see if the air lock is due to the system going through the drain back process to rapidly and somehow entraining some air that gets into your pump.

Maybe just temporarily add a restriction to the return where it ends in the tank.  This would slow down the drain back by not letting air up the return pipe as fast, which will make the system drain back more slowly.

If this solves the problem, then I guess you would have to figure out a way to slow down the drain back while still making sure its complete.

Another thought would be that there is an air leak in the system somewhere that is pulling in some outside air during the drain back.  I know this sounds unlikely, but remember how strangely the air leak in Thom's system made it behave.

Gary

[Tom Sullivan]

Gary,

I think you may be onto something with he possibility of a leak, introducing air to the water in the drainback process.  I remember when first going live last year, when I had a leak, I discovered the leak by hearing air noise before I actually saw a water leak.  I think I'm going to remove my glazing and do a good inspection of the plumbing.

Thanks,

Tom

[MattM]

When water is pumped uphill and comes back downhill on the return, what prevents the syphon effect?

[tomtank]

In this type of installation, there should be no air in the impeller causing an air block since there is the standing head of water in the tank pushing water into the pump to help keep it primed.

I would get into the pump and make sure everything is okay. I would also double check the flow from the tank to the pump.
The galvanized fittings and iron pump make me uncomfortable, even though this system has not been in service too long.

It might help to orient the pump at the bottom of the tank, running horizontal instead of vertical. This will increase the head pressure on the pump.

[Tom Sullivan]

I pulled the glazing off all 8 panels yesterday and hope I've found my problem.  I had 3 lower "T" connections leaking.  I used a brazing process instead of the 3/4" x 1/2" x 3/4" "T"s when I built the panels, and had regretted that process afterwards.  I was able to sandblast and solder them.  I also noticed one of my outside "top return" double elbows bent out and almost completely disconnected.  The 3/4" tubing extending as part of the manifold at the lower elbow was actually bent.  I heated the lower elbow and removed the double elbow, finding the top one had never been soldered!  I suspect at one point when the collector had been stagnated from the pump in air lock, and then I cycled the pump to get it primed, the water flooding the panel was boiling at initial entry and created some steam pressure.  I can't think of anything else that would create enough force to push the double elbow fitting outward and bend copper.  I was able to straighten the manifold tubing, clean and re-solder it all.  I opened my tank and installed a regulator with an air gauge and pressurized the system for an hour with 40 PSI; no leaks.  We have good sun forecast for the next 3 days.  I'll leave the glazing off for at least a day to check it in operation.

One last item I think could have bearing in any air not clearing the entry piping to the pump.  On the suction side I have a 10' length of tubing inside the tank.  This was to provide for some circulation of the tank water during pumping operations.  I would imagine it's possible if any air is introduced during the drain back, it could stay trapped in the tubing if the tubing is not pitched.  I will see if I can add some pitch to that tube to provide for venting any air.

Tom