Water pumping problem at off-grid cabin - Video included

Hairfarm

Greetings,

This post is sort of a followup to one of my earlier posts here:
http://forum.solar-electric.com/discussion/8314/small-cabin-dc-water-pump-set-up#latest

I'm having a new problem that recently surfaced after installing a DC pump in-line with an existing AC pump. After adding the DC pump to the system the existing AC pump is now rapid-cycling at it's cut-off point.

It's all laid out in this video I made today. If anyone can please shed some light on this it would be very helpful to us.
Sorry for the length of the video. I was trying to add as many details as I could.

This problem only surfaced after I installed the DC pump. Everything was working fine before I made that modification. Doh!...

Video: https://www.youtube.com/watch?v=gUvxbKFkBmc

thanks,

Hairfarm

westbranch

I am guessing that the new pump is restricting the flow , old pump senses reaching 60 psi and shuts off, the new pump bleeds  enough pressure ( and volume)  for old pump to cycle... what happens when tank is full, can you get to that point?  Our RV pump bleeds a bit of pressure and that came to mind..

hth

Aguarancher

I watched your video twice and have a couple of questions. Did you bleed off the air in the pipes on both the hot and cold side? .Did you ever run the DC pump to bleed the air out of its side of the system? Last of all maybe your pressure switch on the AC pump is failing. On a side note, what is your pre-charge pressure on your holding tank? It should be set at your cut in pressure.

Ralph Day

Hi
Definitely isolate the dc pump with ball valves.  If you're going to use it, cut the ac pump power and open valves accordingly.  Reverse the process to return to ac pump use.

The ac pump (looks to be a shallow well jet pump) pressurizes it's intake line when it's shut off.  Jet pumps work on the venturi effect...shooting a stream of water past the fresh water inlet and drawing a bit of water along on the return journey.  The dc pump looks to be a diaphram pump, which, in this case, acts as a tiny pressure tank.  The ac pump reaches shut off, shuts off, but the diaphram deflects and the switch on the ac pump actuates to "on" again, and so on.

Once you isolate the dc pump with valves, and only use it when you need/want it, the cycling will dissapear.  Doesn't appear to be any way to automate the supply/delivery system unfortunately.  The rapid cycling with a jet pump happens when your pressure tank is waterlogged too...not enough air to compress in the tank.  The pump shuts off and on repeatedly until you stop it or it burns out.

Ralph

Ralph Day

Blah Blah Blah!  Ralph talk too much.

ball valve the 1/2 inch dc pump inlet pipe.

 Turn Ac pump power off, open dc pump inlet valve, use dc pump.  

Close dc pump inlet valve when operation returns to Ac pump.  No cycling happen.

Ralph

MarkC

Have you tried operating the new DC pump solo?  I'm assuming that in your video the DC pump is not operating.  In "spare" pump installations, isolation valves are always used for maintenance as check valves cannot be relied upon for complete isolation (can leak).  In parallel pump systems (both operating by design), you would not use dissimilar pumps because they will invariably "fight" each other due to the characteristic operating curves (pressure vrs flow).  Aguarancher has a good point of insuring that the DC pump circuit has been properly bled so that it does not form a gas pocket between the suction "tee" and the DC pump check valve that would allow water hammering.  Operating the DC pump "solo" should clear any gas out.  If your pressure gauge has an internal "snubber" (keeps it from vibrating itself to destruction!), you might not see the actual pressure spiking. 

Aguarancher

While not ideal, I see no inherent design problem with his system. I’m still leaning towards an air lock as the problem. Once the system is properly purged of air I bet all will return to normal.

Aguarancher

When I design above ground pump systems, I like to use this type of true union ball valve. This one should be fine for most homeowners. Makes system isolation and service easy. http://www.amazon.com/Red-Flag-Products-Socket-Connector/dp/B00482SRBK/ref=sr_1_4?ie=UTF8&qid=1446995381&sr=8-4&keywords=union+ball+valve

Hairfarm

Lots of great suggestions and I appreciate those that took the time to chime in about my pump problem. This is one of the reasons I like visiting this forum for off-grid stuff.

westbranch said:

I am guessing that the new pump is restricting the flow , old pump senses reaching 60 psi and shuts off, the new pump bleeds  enough pressure ( and volume)  for old pump to cycle... what happens when tank is full, can you get to that point?  Our RV pump bleeds a bit of pressure and that came to mind..

hth

Correct. Cut-off is 60lbs, cut-in is 40lbs. The pressure tank is set to 38lbs, 2lbs below cut-in per manufacturer guideline. The pump will fill the tank and hold pressure as evidenced in the video. The only problem is the rapid-cycling when it hits the 60lb cutoff.

Aguarancher said:

I watched your video twice and have a couple of questions. Did you bleed off the air in the pipes on both the hot and cold side? .Did you ever run the DC pump to bleed the air out of its side of the system? Last of all maybe your pressure switch on the AC pump is failing. On a side note, what is your pre-charge pressure on your holding tank? It should be set at your cut in pressure.

No, I didn't bleed the smaller DC pump yet or the hot and cold lines. I probably should've though. The DC pump hasn't been wired to the battery bank yet either. The pressure tank is set to 38lbs, 2lbs below cut in per manufacturer guideline.

Ralph Day said:

Blah Blah Blah!  Ralph talk too much.

ball valve the 1/2 inch dc pump inlet pipe.

 Turn Ac pump power off, open dc pump inlet valve, use dc pump.  

Close dc pump inlet valve when operation returns to Ac pump.  No cycling happen.

Ralph

I tend to agree and this will be my next step if bleeding the DC pump first doesn't solve the problem. I have ball-valves/unions that I'll install on the suction and discharge side of the DC pump. The pressure tank and pressure switch (on the larger AC pump) were both working perfectly before my modification so I have hard time believing that they one or both suddenly went bad after my mod. But anything is possible though...

MarkC said:

Have you tried operating the new DC pump solo?  I'm assuming that in your video the DC pump is not operating.  In "spare" pump installations, isolation valves are always used for maintenance as check valves cannot be relied upon for complete isolation (can leak).  In parallel pump systems (both operating by design), you would not use dissimilar pumps because they will invariably "fight" each other due to the characteristic operating curves (pressure vrs flow).  Aguarancher has a good point of insuring that the DC pump circuit has been properly bled so that it does not form a gas pocket between the suction "tee" and the DC pump check valve that would allow water hammering.  Operating the DC pump "solo" should clear any gas out.  If your pressure gauge has an internal "snubber" (keeps it from vibrating itself to destruction!), you might not see the actual pressure spiking. 

Thanks for the info! I haven't yet tested the new pump yet. But based on the comments here I'm leaning toward this being the possible cause of the rapid-cycling. Isolation ball-valves with built in unions will be next step.

Aguarancher said:

While not ideal, I see no inherent design problem with his system. I’m still leaning towards an air lock as the problem. Once the system is properly purged of air I bet all will return to normal.

Aguarancher said:

When I design above ground pump systems, I like to use this type of true union ball valve. This one should be fine for most homeowners. Makes system isolation and service easy. http://www.amazon.com/Red-Flag-Products-Socket-Connector/dp/B00482SRBK/ref=sr_1_4?ie=UTF8&qid=1446995381&sr=8-4&keywords=union+ball+valve

I'm going to take your advice and purge it tomorrow and retest. Hopefully it works mainly because it means that I don't have to modify anything. If it still happens I'm going to add ball valves. I already have the ones you describe in your Amazon link. Like you said, great for servicing.


Also, I read on Flotecs website that too many 90degree turns in the system can confuse the pressure switch. But if you look at my video it's really a straight shot from the discharge to the pressure tank inlet. But it a different story on the AC pumps inlet. I think there's three 90 degree elbows on that side.

Hope the purging and/or ball valves will do the trick. Thanks all! I'll report back with results.

Hairfarm

waynefromnscanada

Hey Hairfarm! I see the cause of your problem!
The check valve on the OUTPUT of the AC pump! As soon as the pump stops, that check valve slams shut, the pressure switch then no longer "sees" the pressure in the tank, and as the pump slows down it's output pressure drops, the pressure switch sees this pressure drop and restarts the pump, whereupon the pressure at the switch hits the stop point again instantly. If you had a gauge on the pump / pressure switch side of that check valve, you'd see the pressure going crazy as the pump starts and stops. Remove that check valve on the OUTPUT side of the AC pump, replace it with a straight pipe and problem solved!
You COULD run into similar problems with the DC pump, also requiring removal of the check valve on it's output.
You MIGHT wish to install ball valves in place of those two "output" check valves, but only for the purpose of isolating one pump or the other in case of having to remove one of them for service, or if one of them springs a leak. Other than that, you shouldn't have check valves on the outputs - - UNLESS you locate the pressure switches on the TANK side of those check valves so they always "see" the actual pressure in the tank.
Simple once you understand what's happening.  

Hairfarm

waynefromnscanada said:

Hey Hairfarm! I see the cause of your problem!
The check valve on the OUTPUT of the AC pump! As soon as the pump stops, that check valve slams shut, the pressure switch then no longer "sees" the pressure in the tank, and as the pump slows down it's output pressure drops, the pressure switch sees this pressure drop and restarts the pump, whereupon the pressure at the switch hits the stop point again instantly. If you had a gauge on the pump / pressure switch side of that check valve, you'd see the pressure going crazy as the pump starts and stops. Remove that check valve on the OUTPUT side of the AC pump, replace it with a straight pipe and problem solved!
You COULD run into similar problems with the DC pump, also requiring removal of the check valve on it's output.
You MIGHT wish to install ball valves in place of those two "output" check valves, but only for the purpose of isolating one pump or the other in case of having to remove one of them for service, or if one of them springs a leak. Other than that, you shouldn't have check valves on the outputs - - UNLESS you locate the pressure switches on the TANK side of those check valves so they always "see" the actual pressure in the tank.
Simple once you understand what's happening.  

OMFG!  Right under my nose.  That check valve was only added when I added the DC pump. That would explain the behavior exactly. The pressure stays at 60lbs and the pump continues to struggle against itself on the other side of the check valve. Yep.

You really made my day!

I'll post back after testing tomorrow.

waynefromnscanada

A similar start stop start stop start stop problem can occur if there is too long a pipe between the pump (with pressure switch mounted at the pump) and the storage tank (NOT the cause of your problem, but a heads-up for others).
In the case of a long pipe between the pump and tank, when the pump is running, water is flowing at relatively high speed through the long pipe to the tank. Finally the pressure reaches the shut off point, the pump stops, but the water in the long pipe, just like a freight train, keeps going for a bit before coming to a stop. In the meantime, that still rushing freight train of water drains the pressure at the pump where the switch is located, causing it to restart the pump even though the tank is already up to pressure. Like a water hammer in reverse. This also can be cured by locating the pressure switch right at the tank instead of on the pump.
Again, this is not the cause of your problem, but is something others may run up against and wonder what the heck is going on.

Aguarancher

Good catch Wayne, That one brass check valve should service the whole system, You might want to remove the one on the dc pump as well. I like to keep all pressure switches at the pressure tank. If both of yours were, your system would work as you have it. So either put a pair of square-d pressure switches on a tee at your tank or remove both check valves. BTW, I'm a believer in ck valves as they keep the load off the pumps.

Ralph Day

Ditto, good catch Wayne.  There's a reason there's a pump for your avatar.

Ralplh

Hairfarm

Hi Wayne,  

FYI, An earlier email to Flotec technical support just got back to me today confirming exactly what you said.

thanks again to everyone for their input.

Hairfarm

Aguarancher said:

BTW, I'm a believer in ck valves as they keep the load off the pumps.

Do you find that having the pumps under constant pressure wears them out quicker?

waynefromnscanada

Hairfarm said:

Do you find that having the pumps under constant pressure wears them out quicker?

I've often thought it would be nice in theory to have no pressure on the pump when it was not in operation, but in reality, in normal domestic operation, unless the pump is leaking, long term (decades) of experience shows it really makes no difference. In fact, even with a tiny leak, if there is no water under pressure to replace the leaking water, air can make it's way in as the water leaks out, leading to a possible airlock. In over 50 years living in a rural area where each house and farm has it's own private well and water pump, those with a check valve on the output of the pump are in fact reserved for submersible pump systems where the pump may be 100, 200 or more feet below the surface, and the check valve at the surface serves to reduce the back pressure on the water line leading up from the pump far below. I've never seen a check valve on the output of a surface mount jet or piston pump. In these cases, the water is held from flowing back into the well by the foot valve. Although the valves within a piston pump for the most part stop the water back flowing, there is usually some small leakage back into the well, thus a foot valve is needed there too.  I've never seen a surface mount jet or piston pump damaged in any way by having tank pressure supplied to it when not running, just doesn't happen. And that's why pump manufactures so often mount the pressure switch directly on the pump, just like it is on your jet pump, just like it is on my 60 year old twin piston pump. Remember - - adding a check valve to the output of your pump created problems that didn't exist without it, because the pump was designed to always have tank pressure applied to it.

Hairfarm

waynefromnscanada said:

Hairfarm said:

Do you find that having the pumps under constant pressure wears them out quicker?

I've often thought it would be nice in theory to have no pressure on the pump when it was not in operation, but in reality, in normal domestic operation, unless the pump is leaking, long term (decades) of experience shows it really makes no difference. In fact, even with a tiny leak, if there is no water under pressure to replace the leaking water, air can make it's way in as the water leaks out, leading to a possible airlock. In over 50 years living in a rural area where each house and farm has it's own private well and water pump, those with a check valve on the output of the pump are in fact reserved for submersible pump systems where the pump may be 100, 200 or more feet below the surface, and the check valve at the surface serves to reduce the back pressure on the water line leading up from the pump far below. I've never seen a check valve on the output of a surface mount jet or piston pump. In these cases, the water is held from flowing back into the well by the foot valve. Although the valves within a piston pump for the most part stop the water back flowing, there is usually some small leakage back into the well, thus a foot valve is needed there too.  I've never seen a surface mount jet or piston pump damaged in any way by having tank pressure supplied to it when not running, just doesn't happen. And that's why pump manufactures so often mount the pressure switch directly on the pump, just like it is on your jet pump, just like it is on my 60 year old twin piston pump. Remember - - adding a check valve to the output of your pump created problems that didn't exist without it, because the pump was designed to always have tank pressure applied to it.

Good point.  

A question for you. Since I removed the check valve from the discharge of larger AC pump can I still leave the check valve on the smaller DC pump? Won't the DC pump still sense pressure from its suction side since there's no longer a check valve on the discharge of the larger pump?  In other words when the dc pump is building pressure won't that same pressure make its way back through the discharge of the larger pump and out its suction, then into the suction of the small pump allowing the pressure switch to feel it? The water pressure would be traveling in a loop through the large pump and into the small one's suction.

Hope that make sense. If you look at the video you can see what I mean.

If not, do I need to isolate the DC pump with ball valves or will it work as is?

thanks

waynefromnscanada

Hairfarm said:

A question for you. Since I removed the check valve from the discharge of larger AC pump can I still leave the check valve on the smaller DC pump? Won't the DC pump still sense pressure from its suction side since there's no longer a check valve on the discharge of the larger pump?  In other words when the dc pump is building pressure won't that same pressure make its way back through the discharge of the larger pump and out its suction, then into the suction of the small pump allowing the pressure switch to feel it? The water pressure would be traveling in a loop through the large pump and into the small one's suction.

Hope that make sense. If you look at the video you can see what I mean.

If not, do I need to isolate the DC pump with ball valves or will it work as is?

thanks

Haha You are understanding things!
Indeed, once the jet pump stops, full tank pressure will be applied all the way back to the single common brass check valve at the water inlet coming through the wall, so yes indeed, the little DC pump will just circulate water backwards through the jet pump instead of adding water to the tank. Two ways to fix this. One is to have individual check valves on the INLETS of each pump so they can only draw from the source and not backwards through the other pump, the other way is to use manually operated ball valves on the outlets of each pump, OR the inlets of each pump, OR if you wish, go all the way and put ball valves on both the inlet and outlet of each pump so each one can be totally isolated both to prevent circulating the water and to allow removal of either pump while the system remains operational using the other pump. I'd tend to go with the latter. Good catch.
BTW, the little DC pump almost certainly has internal valves preventing the back flow of water, just like piston pumps do, thus preventing the jet pump circulating water backwards through that little pump. Jet pumps like yours on the other hand do not have any such internal valves, so water can easily move backwards through it if it's not running, unless prevented by an external valve of some sort.

inetdog

waynefromnscanada said:

Haha You are understanding things!
...
... Jet pumps like yours on the other hand do not have any such internal valves, so water can easily move backwards through it if it's not running, unless prevented by an external valve of some sort.

Unlike submersible pumps, jet pumps need to be primed to allow water to be pumped under pressure down to the jet down in the well. That constant prime can come from the pressure tank on the outlet side of the jet pump, but that is only practical as long as there is also a properly functioning foot valve at the bottom of the well to keep standing water in the pipe running from the jet location to the pump.

waynefromnscanada

inetdog said:

Unlike submersible pumps, jet pumps need to be primed to allow water to be pumped under pressure down to the jet down in the well. That constant prime can come from the pressure tank on the outlet side of the jet pump, but that is only practical as long as there is also a properly functioning foot valve at the bottom of the well to keep standing water in the pipe running from the jet location to the pump.

That's true both for jet pumps set up for "deep well" use with two pipes going down in the well as you describe, where the actual jet is down in the well. It's also more or less true for those set up for "shallow well" operation as shown in the video supplied by "Hairfarm", and uses only a suction line, the jet being contained within the actual pump body. If the well is shallow and if there are no leaks in the suction line, this back flow can be prevented by an inline check valve, similar to the one just inside the wall on Hairfarm's installation, although a proper foot valve almost always provides far more reliable operation, AND normally have as part of them, a screen to prevent the uptake of gravel and small rocks etc.
The only time I've seen or installed a shallow well pump (jet or piston) without a properly operating foot valve, is when the water supply is at cottages at a lake and only used during the Summer. We've also done this with submersible pumps installed in lakes by removing the built in check valve in the outlet of the submersible pump. In these cases in these parts of the world, such Summer use only installations are drained back to prevent freeze damage during Winter. In these cases, no foot valve is used, just an inlet screen, and a check valve is installed near the pump inlet, or in the case of submersible pumps, at the tank inlet from the pump. Come Winter shut down, a valve which has been installed on lake side of the check valve is opened to allow atmospheric air to enter the suction line, allowing it to drain back into the lake, preventing freeze damage. Of course the in cottage pump and indoor plumping must also be drained, OR filled with plumbing antifreeze until Spring, but it negates taking connections apart and pulling the suction line from the lake. The suction line need only be buried a short distance at the water level, enough to prevent shifting and expanding and contracting lake ice from damaging it.