[SOLVED] Help: Water leaking back into well shaft

[erkme73]

My home/installation is about 4 years old. I have a 1HP/10GPM gould pump (10GS10) at about 280' with static water level at about 40'. All has been working without incident w/regard to pump and supply of water. I installed the CSV between the poly pipe and the pitless adapter about 3 months ago.

Today I noticed that the pressure gauge in the basement was dropping, pump cycling on, and no water being use din the house. The cycle stop sensor was showing uncharacteristically high current (7.5a vs. the regular 5.5ish) all the way from pump start to shut off. Hearing no water coming down the drain pipes, I assumed I have a leak between the pump and the house.

I shut off the valve coming into the house and the pressure drop on the house-side stopped. I then hiked to the well head and looked down into the shaft. When the pump shuts off, the visible water surface is rippling constantly and appears to be rising (though that is difficult to verify given the narrow field of view and depth). When the pump cycles on, the rippling appears to stop, and the water level drops again. As soon as the pump stops, the rippling and rising begins. There is no audible sound of water spraying or dripping. The poly pipe from the CSV down to the water surface appears dry.

When my son closes the valve at the pressure tank (separating the tank from the pump), the rippling immediately stops. Clearly, water is leaking back from the tank/house. The well driller/pump installer has since retired, but I did verify with him previously that there are no check valves anywhere between the pump and the pressure tank.

Aside from the cycling, the fact that the pump is drawing 7.5A indicates it's probably moving much more water than it normally takes when running up against the CSV - which makes me think maybe there is a leak in the poly. Would water exiting a failed/leaking check valve cause the water 200' up at the surface to ripple?

TL;DR:

Water appears to be returning to the shaft, presumably below the static water level. Unless the poly pipe has burst/developed a leak, the presumption is that the check valve at the pump has failed. With no other check valves in-line, and a CSV installed at the pitless adapter, I need to know what my troubleshooting/problem solving options are at this time. I would prefer to avoid pulling the pump up, but don't know how a check valve between the CSV-pitless, or poly-CSV would affect performance or reliability.

For now, I have shut the valve between the pressure tank and pump, and turned off the pump breaker so it doesn't cycle indefinitely.

Any advice would be appreciated!

 

[erkme73]

I have a home automation system which monitors, among other things, total power use. In retrospect, I can see the cycling of the pump - which started around 6pm on the 10th. At that point, the pump was on for about 1 min, off for about 2 min. Since then its gotten worse - running for about 70 seconds, but off for only minute. So whatever is causing the pressure loss - whether a failed check/foot valve, or a hole in my poly - it has gotten worse over the last three days.

 

[Sponsor]

 

[Reach4]

I cannot explain the rippling, but your symptoms are compatible with an underwater hole in your pipe.

1. 10GS10 pump can produce over 160 psi at the pump. That is not really a lot as pumps go, so it is unlikely to have burst the pipe from just that.
2. What pressure is your drop pipe rated at?

When my son closes the valve at the pressure tank (separating the tank from the pump), the rippling immediately stops. Clearly, water is leaking back from the tank/house. The well driller/pump installer has since retired, but I did verify with him previously that there are no check valves anywhere between the pump and the pressure tank.

It is not normal to have a valve between the pump and pressure switch.

I would prefer to avoid pulling the pump up, but don't know how a check valve between the CSV-pitless, or poly-CSV would affect performance or reliability.

I think you are asking what would happen if you added a check valve to work around the suspected hole in the pipe. You could add a check valve before the pressure switch and pressure tank. That would stop the cycling. It is likely to cause a bang when the pump turns on. Maybe a bang when the pump turns off. That work around could tide you over until you repair the pipe.

Your troubleshooting seems remarkably good.

 

[erkme73]

Thanks for the reply. I do not know what the pressure rating of the poly is - though I will find out tomorrow when I start to pull up the line to find the leak. I know that having a valve between the pump and PS is not typical - but I added it specifically for troubleshooting. It is zip-tied in the open position, so closing it requires effort and premeditation. I have a barn and exterior spigot that are between the pump and PS, so being able to isolate the house from those fixtures lets me isolate where leaks are.

I guess the $64k question is this: would a failed/failing foot valve cause symptoms while the pump is running? IOW, would it cause the pump to lose some of its output through the valve? Or does that valve only come into play when the pump shuts down (to prevent losing the upstream pressure)?

Because the higher-than-normal current makes me think the pump is not running at the higher pressure anymore (where flow rate - and therefor load - is decreased). When all was running as normal, the pump would initially run around 7 to 7.5 amps, but it would quickly settle down to around 5.5a while running against the higher head pressure (and slower flow) from the CSV. Thus, my presumption that the pump is actually running at lower pressure, higher volume. But if that could be caused by a failed foot valve, the I may still have to pull the whole line.

With the rippling occurring all the way at the surface of the water, with the pump 200' below, I'm guessing (hoping) the leak is closer to the surface than the pump.

 

[LLigetfa]

would a failed/failing foot valve cause symptoms while the pump is running?

No, unless the failure was the housing that lets water out. BTW, it is not considered a foot valve but rather a check valve.

Is there only the built-in check valve or is there a check valve installed on top of the pump? A corroded nipple between the pump and check valve, or leaking barb nipple at the poly connection could also be the culprit. Is it a long barb nipple with double or triple gear clamps?

 

[erkme73]

No, unless the failure was the housing that lets water out. BTW, it is not considered a foot valve but rather a check valve.

Is there only the built-in check valve or is there a check valve installed on top of the pump? A corroded nipple between the pump and check valve, or leaking barb nipple at the poly connection could also be the culprit. Is it a long barb nipple with double or triple gear clamps?

According to the installer, he does not install any check valves beyond what comes on the pump. I assumed that would be at the bottom of the pump (on the suction side).

If it is on the pressure side, then a failure could cause blow-by if it's ruptured, I presume. However, that would still be 200+ feet below the static water level, so I doubt (?) That would cause the surface to ripple (and stop immediately) when turning off the pump.

The rain has stopped and I'm gearing up to start pulling this morning. Figures it had to happen on a Sunday, during a global pandemic ... But that's how I roll.

 

[Reach4]

I guess the $64k question is this: would a failed/failing foot valve cause symptoms while the pump is running? IOW, would it cause the pump to lose some of its output through the valve? Or does that valve only come into play when the pump shuts down (to prevent losing the upstream pressure)?

Nothing in a submersible pump system is called a foot valve. The water intake is part of the submersible pump.

The check valve only come into play when the pump shuts down (to prevent losing the upstream pressure).

While pulled, if your well is of sufficient diameter, you could benefit from a flow inducer sleeve.

 

[LLigetfa]

If it is on the pressure side

It is.
You did not address the barb nipple. If it is not brass or stainless steel, It could have rusted through. If the gear clamps were not double/triple, there could be leakage there.
As for the ripples, I think those could be present regardless.

 

[erkme73]

my apologies for using the wrong terminology. Check valve it is. I have about 75 feet of the hose up at this point, it is one inch rated at 160 PSI, I do not know what nipple is at the other end, but the one at the very top where it was connected to the pitless adapter is rusted pretty badly and only has two hose clamps both in the same direction stacked on top of each other.

 

[erkme73]

Pump is out. you guys are awesome...

 

[erkme73]

So, it turns out the hose is fine, and the nipples are made of steel that has rusted through. I'm going to replace both with brass fittings from Lowes. Surprisingly they had them in stock. Now on to another question...

There were no torque arrestors anywhere on my line, and I'm seeing evidence of chafing on the wires. Should I do anything to prevent further damage?

What can I do to avoid the cable damage I'm seeing?

 

[LLigetfa]

What can I do to avoid the cable damage I'm seeing?

I use both a torque arrestor and stand-offs but the general consensus here is not to use anything, not even a rope. Double jacket wire would be more abrasion resistant.
One time when I pulled the pump, the torque arrestor had disintegrated and came out in pieces. I have a 6 inch casing so there is less chance of something jamming and preventing the pump from being pulled.

 

[erkme73]

Well, what concerns me is that the insulation has abraded all the way through to the conductor on one of the hot legs. In fact you can even see the copper has turned green where it's been rubbing up against the sidewalk. So far it hasn't had any effect, but I can imagine given enough time that will be a point of failure. On the upside, pulling this thing really wasn't that hard with my tractor

 

[erkme73]

My apologies for the horrible speech-to-text dictation from my phone above. This is likely my final post, and I wanted to summarize what happened using a real keyboard.

1. The ripples in the water were generated from the pressure returning to the well shaft from the pressure tank after the pump turned off.
2. There was no damage to the 1" poly pipe (rated at 160 PSI) despite my theory that the rippled water was coming from something closer to the surface.
3. I pulled the pump and discovered what others had predicted - the 1" NPT male to 1" barb fitting at both ends of the poly tubing were both made some kind of coated steel (zinc?) - and they were rusted. The one at the output of the check valve had visible holes in the threaded area.
4. Lowe's had brass replacement fittings. They were not located with all the other plumbing fittings, but rather where they keep the well pumps (makes sense). The were direct replacement and fit without issue.
5. There were two hose clamps on each barbed fitting, though both were installed in the same direction.
6. Removing the fitting from the check valve without sheering it along the pin holes in the threaded portion was challenging, but by adding subtle heat to the top half of the check valve (never too hot to touch), it broke free and came out without complication.
7. I had two helpers with the extraction of the pump - though they were just helping guide the poly/rope/wires around trees to avoid kinking. Dropping it back into the well post-repair was done without help (other than gravity).
8. After flushing the lines to an outside spigot (located between the pump and the pressure tank), I pressurized the house and found the pressure held rock solid after the pump turned off.
9. I started at 9A and finished by 2P - with some down time waiting for the fittings from Lowe's.

I did re-tape the electrical wires where they had considerable abrasion. Given that bare hot-leg conductor was exposed to the water/shaft and never tripped the breaker, I'm guessing the consequences are minimal - at least until the cable itself fails. Being pressed for time, having no viable electrical/plumbing wholesale place open on Sunday, I opted to simply reinforce the damaged sections with tape, and returned the assembly to the shaft.

Thank you to everyone who helped talk me off the edge, and for giving me the confidence to do this myself. Wife is happy. I am happy. This could have been much worse. It could have gone wrong in so many ways. And it wasn't. It didn't. And I'm thrilled.

 

[Reach4]

Some people support a big wheel that serves as a pulley above the casing. This support could be built from wood, but there may be some alternative. I wonder if your wire could have gotten skinned up being pulled over the casing edge, despite your nice blue padding.

 

[Valveman]

It is all in the kind of wire used. I have never seen double jacketed THHN wire chaffed to the copper.

 

[erkme73]

Some people support a big wheel that serves as a pulley above the casing. This support could be built from wood, but there may be some alternative. I wonder if your wire could have gotten skinned up being pulled over the casing edge, despite your nice blue padding.

Trust me, I spent quite a number of brain cells trying to come up with a contraption that would allow me to pull the poly up straight without kinking or chafing it. In the end I pulled with the tractor while a helper lifted up to avoid the problem.

I suspect some of the abrasion may well have come during the initial installation. There are a lot of skipped/bite marks that look like it bounced off an edge on the way down. My attempt to soften the edge of the casing (good eye, btw) worked reasonably well - it held the whole time and did keep things from hitting the edge. How it rubbed on the inside of the casing as it made the turn, I don't know. However, the bare conductor is green, indicating some kind of hydrolysis/corrosion which, I'm sure happened over time. Thus, I'm pretty confident that the damage happened either gradually or quite some time ago.

It is all in the kind of wire used. I have never seen double jacketed THHN wire chaffed to the copper.

Ultimately, I may end up replacing the wire. At least I know the failure potential is there, and what to expect in the future. To be honest, removal/installation of the pump was a lot easier than I had imagined/feared. I had about half of it out on my own before I realized the angle of the exiting pipe was getting too extreme. Once more than half of the pipe was up, it required very little manual labor to get the rest out.

Having the exposed conductor for a hot leg in the water - even when the PS is open and power is off to the line, I wonder if there is some kind of transient or induced current that is causing the cycle sensor to show that low (0.1 - 0.3A) draw on the display.

Interestingly enough, the static current is now showing nearly zero - maybe due to taping the exposed conductor or having repositioned the wire in the casing?

Also, now that 100% of the pump output is going through the CSV, the current shown on the cycle sensor is back down to around 5.5A.

I have to admit, when I first discovered the problem last night, I thought to myself, "I wonder if installing that CSV caused this problem". I envisioned finding a blown poly pipe. Clearly, a rusted fitting is no fault of the regulator - though the higher pump pressure may have caused it to fail sooner or more dramatically. One thing is certain... If I didn't have the cycle sensor showing me the 7.5a last night as I was in the basement minding other business, I might not have discovered the problem. It was because of the unusually high current, the lack of sewer sounds, and the cycling of the PS that grabbed my attention. It is safe to say CSV not only didn't cause the failure, but it helped me discover it.

 

[Valveman]

Any copper wire will turn green when exposed. It is not electrolysis. But those steel insert fittings coming apart was from electrolysis, not back pressure. Wrapping them with electric tape would have prevented the problem, but brass fittings are good as well.

Sorry I didn't get to the thread sooner. Cycling on/off and high amperage when you have a CSV is a certain clue you have a leak prior to the CSV.