Why would well pump cycle on and off every second?

Dear jim:

And your tank doesn't mask the problem?

It is 2 gallon "pressure tank". Have you *ever* seen this size for water service? I have seen them for compressed air service.

You are entitled to your opinion. Fortunately, I don't have to hear it.

David A. Smith

Dear Harry K:

Do you have any experience/history with soft starts being used with jet pumps? Since the problem is brought about by the "starting torque" of the motor, if you slowed the startup, the "cause" of the problem disappears (and it might even pay for itself in reduced starting currents). I just don't know how effective (long lasting) they are for high rpm motors...

Something like the Telemechanique (Square D) Altistart ATS01, or something of its ilk... (McMaster Carr only has three phase soft starters available).

Thoughts?

David A. Smith

Air dissolves into water pretty fast -- a few weeks for two gallons of air. That was the other reason for bladder tanks besides smaller size.

Bret Cahill

No, afraid not. I see no reason it won't work and should fix the problem. It doesn't add more equipment to the system as several others in the thread want to do.

Harry K

There is an assumption that those who are reading posts in an engineering newsgroup have sufficient knowledge to understand enough of the thread so as to be able to ask for clarification of a post, and to not make stupid comments until they do.

you know, Dougie, if you have little experience or background in the subject matter as you appear to have, and your reading skills are as poor as yours, you should let the engineers who do this regularly, speak. Especially engineers with decades of experience in the particular subject and who have several well pumps of their own (two shallow installed 1981 and 1984 and one deep 1978 at the lake places, one jet 1970 at the lodge - still running with no problems, precharged to 30 psi.)

see below for cites refuting specific neophyte comments

Who says? For what purpose? One can select a precharge for the parameter one desires to maximize: for max usable volume, then precharge closer to low set point, and for max damping of pump-ripple in the water in order to minimize parts wear or to get maximum control stability, precharge at a lower setting-- or one can precharge somewhere between, as a compromise.

There are four reasons that 38 psi precharge with a lower set of 40 psi is narrowly considered and the recommendation is poor engineering. And I don't care WHICH junior engineer made that recommendation.

First, few people have a water pressure gauge that has better hysterisis than 2psi, let alone accounting for the system hysterisis, so calling for

38 psi precharge for a 40psi switch setting is total BS from an execution standpoint.

Second, a precharge of 38 psi on a 40 psi start? So you can have pump ripple cycle all your control parts? A precharge set above 60% of upper pump setting would only be made by a pump salesman or switch salesman looking to sell more repair parts. 38 psi precharge on a pump starting atr 40 psi is BS from a basic design reliability standpoint.

Third, set for 38 psi so you can have how much water stored under pressure at 40 psi? Enough to keep the pump from running except on draws above the rule-of-thumb of a gallon? A five gallon tank gives a gallon reserve at 25 psi precharge 40-60 psi, and the same five gallon tank gives a gallon and a half reserve at 38 psi precharge 40-60 psi.

Try the old P1*V1=P2*V2 equation, using the differential from 40 psi and 60 psi.

(Ppchg*Vtank^1.1)/Ppres = Vair-vol^1.1 and then use (Vair-vol 60)-(Vair-vol

40) to find water stored in the operating range

for a given tank V between 60psi and 40 psi, V usable =[(Ppc/40) -(Ppc/60)] Vtank

Fourth, experienced engineers avoid control settings near system spring constants unless the controls have dampers or they latch. 38 psi in an air bag is close enough to the switch setting of 40 psi such that oscillations in the 38 psi air volume from the moved-water momentum changes can adversely affect a non-latching switch. And they do.

I also refer you to the Fluid Power Engineers Handbook, Parker Hannifin, page a-3.

The pump in this thread is closer to isothermal operation, and so I quote

" pre-charge..to about 1/3 to 1/2 system pressure. "

I have gone to a higher precharge than 1/2 for specific design applications, and I knowingly accepted the loss in the other parameters.

What, in your world P1*V1 = P2*V2 does not apply?

20psi air pressure*Volume air before water = 20psi air pressure* Volume after water

means volume of air is constant. Same volume means no water can displace air.

Apparently in your world, you can compress 20 psi air with 10 psi water and change its volume without changing its pressure. Patent it , quick.

Neat trick in your mind, but not possible in the real world.

"As water enters the tank, the volume decreases". That "water enter the tank" cannot happen below the 20 psi of my example, as I said.

perhaps I should have written "tries to fill" rather than "fills"

If the pump isn't developing

Gee, that is EXACTLY what I said. "The compressible air will stay at 20 psi until the water pressure rises above 20 psi"

If the pump is

Which is what I said.

You really should not be posting to engineering newsgroups other than to ask a question. Read carefully again the comment by one who has done this for decades.

After the initial fill, as water is withdrawn from the tank,

In spite of the fact that I was addressing force interactions and you are fixated on precharge and have wandered off the path, I will comment on your error.

Contained air is a spring, with a spring constant. The higher the precharge, the higher the spring constant. All pump-fluid-control systems are spring -mass systems that interact. When pumps false start, it is a sign of unintended interaction, There is no discussion among engineers that changing a spring constant changes the interaction, only discussion as to what amount.

Either way, the pump

apparently the original poster's comments refute your statement. His pump shut off after a second at what appeared to be 40 psi. Off at 40 psi. Not off at 60 only, as you claim. In the real world, precharge pressure can and does affect controls.

FWIW - you have put a second air spring in parallel with the original air bladder, and by

1/total spring=1/spring1 + 1/spring2,

you lowered the total system spring constant

(and in so doing, you reduced the peak force the system now sees from the moving water)

Adding a second air chamber in parallel to the first air chamber had the same effect on the system as lowering the first tank's precharge in order to lower the system spring constant.

That's a typical recommendation by manufacturers of well pumps and pressure tanks.

[snip]

Obviously your experience with the reliability of residential well systems is purely theoretical.

If you think that 40 and 60 are the right numbers to use, you don't know as much as you think you do. Hint: pressure guages read -zero- at an actual pressure of 14.7 psi. [snippola]

I think I pretty clearly said that it does.

Read what I wrote.

Ya know, I'm not a mind reader. If you don't say what you mean, then you should expect to be misunderstood.

fine...

No, that's not what you said. But there's not much point in arguing about it.

So whose pump is false-starting? The OP's problem is that it *stops* when it should not.

Obviously not, if you've read the entire thread. The pump was shutting off _at_60_ due to pressure surge, not at 40 as it _appeared_ to be.

The OP's problem was not the precharge pressure; as noted by others in this thread, his problem is an improperly installed system. As you would know if you had any _real_world_ experience with residential well systems.

No one here thinks that the problem is caused by the mass of water that needs to be accelerated when the pump starts?

That the initial surge of torque in the pump needs to get the mass of water between the pump and the tank in the house moving, thus the reason for the higher pressure (seen at the tank) at switch-on?

That moving the pressure switch to the tank end would definately solve the problem, but that adding a snubber (the OP chose to use a 2 gallon tank) before the mass of water would cure the problem too, and the OP didn't have to dig a trench that he didn't want to?

If the air in the tank is going to be absorbed into the water so fast, what keeps normal snubbers (vertical bits of pipe in your house that stop hammering when you shut off the taps) from filling up with water?

Al...

< No one here thinks that the problem is < caused by the mass of water that needs to < be accelerated when the pump starts?

Most agreed to that near the beginning of the thread.

I thought a check valve could let a little water through each cycle but I forgot,

1. the check valve in conventional systems is at the bottom of the well, and,

1. preloaded check valves are loaded shut, not open.

. . .

< If the air in the tank is going to be absorbed < into the water so fast, what keeps normal < snubbers (vertical bits of pipe in your house < that stop hammering when you shut off the < taps) from filling up with water?

Here are some suspect theories:

1. Maybe the water doesn't flush through or exchange very much.

1. The water is already saturated with air.

2. Very little or zero air is necessary to stop hammer. You really only need to adjust the volume.

I used a bladderless 40 gallon tank for 4 years and every 6 months I had to drain the water out because the air somehow escaped.

The water was brought up 50' in turbulent two phase flow and should have been pretty well aerated for a vacuum.

Bret Cahill

Perhaps you had microscopic leaks? Big enough to allow gas through, but too small for water... I've designed quite a bit of stuff for underwater and given enough time things can get through... Usually it's the welds, but once I had leakage past some hydraulic purge caps that were pressed into service as air bleeders... I got a couple cups of pure tasting water into several electronics pressure cases that were down ~50m for 2 years... and this was in the Bosporous.. salty water... The water was so pure we never had any corrosion or even electronics failures in the cases...

Anyways, I'm guilty of applying my mind to the OP's problem, though I've never installed a well pump... And as this stuff isn't anything new, he really shouldn't have asked for help here.... He should have gone to the place where he bought the well equipment...

But it is fun to think this stuff through and learn a thing or two. And perhaps he wanted to avoid "the right way" for reasons of his own, in which case this is exactly the right place to come and get ideas from...

Al...

Well said! Mr -- apparently knows more than the engineers at the pump/tank manufacturing plants.

Harry K

They do. It just takes longer than would one would expect. The water in the tube is pretty much captive and very little exchange is made. Given long enough the tube will water lock and need to be drained.

Harry K

The air didn't "somehow escape" - it dissolved in the water.

This line of thought has merit, I reckon. A start current limiter would have something of the same effect. Though if it is contactor driven, it's as easy to get burned contacts as a malfunctioning bladder tank, I suppose.

Then again, something like a light/motor dimmer/controller [cheap!] could readily be rigged, and the solid state, hands-off use might well be very reliable, barring lightning induced transients, or some such.

Brian Whatcott Altus OK

from the original poster:

see below -

What you note (accelerating a mass of water) would more likely cause a cavitation problem - what I addressed is the already moving water (being relatively incompressible, the water is already moving in under that one second at start-up, since its response time is much like a steel rod connected to the impeller) hitting the compressible air and rebounding, and that dynamic rebound energy being added to the static pressure energy, causing a force spike that the switch sees as misreads as static pressure- that interaction a fairly common problem in fluid power.

No, there is always an initial surge of torque with a pump, even when unloaded ( due to the resistance of the rotor and/or water-mass accelerating). If the mass of water is too high for the pump, usually it will either cavitate (make noise and eat impellers) or run hot at start. I haven't seen an undersized or improperly designed pump cycle the start-up.

Yes, there are many physical ways to remove the symptom and stop the pump from cycling on start. Once a change is found that works at removing the symptom, then one should see what change in the scientific parameters the physcial change made, so as to check if the fix added a future problem. Occasionally a fix in one area creates a problem in another, and sometimes the fix could have been done more cheaply and the cost of that experience can be amortized over furture problems

I don't believe I ever claimed that air was entrained in water at residential pressures. The snubbers don't fill up with water at residential pressures.

Some facts -

I wrote

Dougie wrote

Original poster Nomad wrote

and later, Nomad said again

which explains why Dougie has problems with understanding the poster's problem

nuff said.........

Harry, I have no problem challenging engineers on the breadth and scope of their recommendations. My questions have on several occasions caused industry "rules-of-thumb" to be flat-out changed. As a chief engineer and principal engineer for almost 30 years, I am not cowed by an engineer because an engineer was hired by a firm or because he works there. To increase the firm's engineering team's outlook and experience and knowledge is often part of my job. Ironically, if the engineer always does their job right, they never have any problems nor any non-book feedback to learn from. Almost all work I do for a firm is done at the firm's request, done when a firm's engineering teams are unable to solve their company's technical problems. And often as not, they missed fundamentals and relied on their own misinterpreted experience.

Engineers are human. Like all humans, they give their best guess based on their knowledge and experience, and like all humans, their knowledge and experience may be misapplied or it may be inadequate or through poor management they may have misinterpreted their own experience. The better engineers ask, so as to interact and expand their understanding. Others.........

immediately,