Well pump and heat pump

Excellent post Valveman and very informative putting my setup into historical context. Admittedly I was a bit out with the pricing because I chose the CSV2W at $450 which is probably overkill for my app. This is a neat twist you have. Kudos to you. This is so simple and the simple ideas are always the best.

My view was a bit blinkered, focussing on the high flow heat pump side of things and completely neglecting the house water side. The PRV works fine for high flow. I realised I had to keep plenty margin to prevent dead heading. I never even considered the possibility or benefits of always preventing cycling under low flow conditions. My eyes have been opened and I now fully understand the advantage of your approach. After all, the best time to take a shower in our house is when the heat pump is on and there is then no pressure fluctuations.

No back to my original question, I checked the screen in the PRV and it was completely clear. I raised the pressure setting on it and still cannot get above about 43psi when the heat pump is running. With it opened up fully I would have expected the pump to cycle. Is there anything else that can fail in these PRV's to have them "stuck" regulating even when you adjust them fully open?

Failing this the problem must be on the pump side of the regulator. Maybe the pump itself, a blockage or a leak?

Bob NH said:

I would not be in a hurry to replace the heat pump with gas.

Click to expand...

Bob, our fuel prices here in Ontario are similar, the ratio is anyway. I got a guestimate quote last year to try to get an idea of how much it would cost to replace the heat pump. They suggested around $18000.

I can put in a high efficiency gas furnace and AC for half that. I would need to save about $500 a year for the next 18 years to cover that cost. We may not even stay in this house that long.

Maybe the quote was high and I can do better but at double the price it's hard to justify.

“Is there anything else that can fail in these PRV's to have them "stuck" regulating even when you adjust them fully open?â€

Yes, the kind of reg you have has from 7 to 25 PSI friction loss depending on the flow at any particular time plus some reduced pressure fall off. If you put a gauge before the reg you will probably see about 53 PSI while the gauge on the tank or heat pump is reading 43 PSI, even though the valve is set at 55 PSI or higher. When the heat pump and water in the house is used at the same time, the difference in these two gauges will be even more. If the reg you have is not the problem then either the water level in the well is 10' to 15' deeper, your pump is worn by a few per cent, or a combination of both. If you use a regulator that has less friction loss, the pressure on the heat pump side will increase. This would be a cheap way to get the pressure back up on the heat pump. Otherwise if the water is deeper, you need a larger pump. Or if the pump is worn you need one that will function like new.

The only reg I have with less friction loss is the CSV1. A CSV1-50 would probably hold you heat pump at about 51 PSI without changing anything else. I think Speedbump sells these for 60 something bucks. The CSV1 does not have reduced pressure fall off and has only a pound or two of friction loss.

You wrote;
“After all, the best time to take a shower in our house is when the heat pump is on and there is then no pressure fluctuations.â€

With the CSV1-50 and a pressure switch setting of about 40/60, you can have this kind of pressure all the time, heat pump running or not. (smaller tank makes it happen sooner)

The trade off we make to get these great flow characteristics is a possible leak. It does not always happen but, I am an honest used car salesman so, I always bring it up. The leak does not stop the valve from working and it is usually only about a gallon per weak at best. We like to install them inside the well casing (with added weight coupling) or in the storage tank if possible so, if it leaks, it leaks back down the well or back into storage. Some people have been known to just put a butter dish under it. It will usually evaporate faster than it drips. If I could fix this leak I would have years ago but, it is a great valve and the leak is just part of what causes so little friction loss.

In Speedbump and Gary’s defense, they have seen the CSV solve some real problems and they know it works. I appreciate them coming to my defense as I also have a real job and don’t have the time to keep up everywhere. Gary and Speedbump have considerable experience and are much closer to understanding everything about valves than anyone else you have heard from. When I see them miss a little point or two it makes me worry greatly about the understanding of the average person. For such a simple little valve the CSV has a very complicated explanation. This keeps the average person from getting the benefits of the CSV because of a lack of understanding. Engineers, supply houses, and pump men will not recommend something they are not sure they can explain, even though they know it works. I am trying to find better ways to explain it so I appreciate the opportunity to answer these questions. The negativity of a few individuals here is understandable but, it is not the product, it is my lack of ability to give a better explanation.

Thanks for the response Valveman. I will add a nipple before the PRV to allow me to monitor the pressure both before and after.

I'm guessing if I had a leak in the system I would see air. Is there any likelyhood a leak would go undetected?

The CSV solution is a much, much more elegant ,flexible and reliable solution than the PRV.
Consider this:-
Ideally you want to regulate the pressure somewhere BETWEEN your pressure switch ON and OFF settings. So with a 30/50 you want to regulate to say 45. This way you will supply a constant pressure (45psi) to your tank/house. This could be done with a simple PRV.

Of course with the above setup we will never reach the cutoff setting of the pressure switch and the pump will run continuously. This is good while we are demanding water because it prevents the pump cycling and we have constant pressure.

The downside of this is that when you stop demanding water the pump will continue to run, FOREVER.

So how do we get over this hurdle? Simple, we provide a bypass path around the regulator that water can "trickle" through thus over a short period providing sufficient pressure at the switch to turn the pump off.

Brilliantly simple and conceptually obvious. The design of this bypass, however, is critical to reliability. Incorrectly implemented, it will block. This is where Valveman's idea comes in. Now, Valveman correct me if I'm wrong, he did not wake up one morning, go to his toolbox, pull out a file and create a notch in an off-the-shelf regulator and start selling it as the CSV. This process took many months, many experiments and many destoyed regulators and prototypes to perfect a reliable implementation. This is why he won the patent.

Yeah, for certain apps the PRV and CSV are fully interchangeable. The problem with the PRV config is that you need margin between the switch cutoff and regulator setting and the only thing that can buy you this margin is demand. So it only works reliably with a predictable and relatively high demand (my heat pump for example). With low, unpredictable demand we can no longer rely on the PRV and must allow the pump to cycle. This is where the CSV wins hands down.

Dirtmover, if you do not have a check valve above ground anywhere, you can check for a leak by closing off all water use to the house and seeing if the pressure in the tank falls over time. If the pressure stays up, no leak.


Your description of a system with a standard regulator set at 45 PSI and a pressure switch setting of 30/50 is right on the money. The only correction I would make is when you describe no water being used and the pump just continues to run FOREVER. This is correct except that the pump is running against a dead head and will only last a few minutes before it burns up and shuts off for good.

You are also correct in that I did not just wake up one morning and devise a workable invention. I was born into a pump business and had many years of experience with pumps to understand what the problems are. Then I had several years of extensive training in electronics and computers and tried Variable Frequency Drives to eliminate pump control problems. Then I realized that the "counter intuitive" property of a centrifugal pump causes the power consumed to reduce by throttling the flow with a valve, the same as when slowing the RPM with a Drive. Only then did I go to work on valves. I tried dozens of different brands and styles of valves with multitudes of different bypass tubes and drilled holes and numerous ways of keeping the hole or tube from clogging. It was a eureka moment though, the first time I simply notched a valve seat. I noticed instantly that the water hammer and "hunting" went away. Then I tried every way possible to clog up the notch to no avail. At the time I did not understand how it worked, I just knew it did. My wife had to convince me to even try for a patent. I thought the idea was so simple that someone would have already thought of it. To my surprise there was nothing similar in the patent search so, the first patent was filed in 1993. The last 14 years I have spent with many trial and errors, figuring out how it works, why it works, and where it works. Recently many companies have come up with new variable frequency drives, and are finally catching up to where I was 15 years ago. Now I spend much of my time teaching classes on the pitfalls of the VFD and why the CSV is a superior means of pump control. So I hope you can understand why I am a little irritated with a few people on this forum who think they know a better way. I would be out of business if someone does find a better way but, it is not going to be the VFD, and it is not going to be easy and just fall in their lap.

Of course, there is no check valve and the pressure stays up so no leak! This leaves the valve and the pump. I'll check the pump side pressure at the weekend.

Nah, I didn't read the patent until this evening. I just tried to explain it as simply as possible in the hopes that different wording might make it clearer for others. The concept is fairly trivial, it's the implementation that is the clever part and you don't need to understand that to realise the benefits.

So do you have a CSV distributor in Canada?

We actually have a warehouse now in Quesnel, BC that is run by an extremely knowledgeable pump/valve man and good friend. They just distribute to pump supply houses in many parts of Canada. Let us know about the valve inlet pressure.

update

OK, so I made a couple of discoveries.

For some reason I believed that the well was 200' but I just found the drillers report and it is in fact 90' with the pump set at 60'.

I borrowed a clamp meter and I'm measuring a current draw of 8.5A. At 220V this is more like that of a 2.5hp motor.

Now I'm beginning to doubt my memory and may be wrong but I was sure that they put a 1hp pump in there 7 years ago. Unfortunately I can't find the invoice. For sure the pump it was replacing was 3/4hp beacuse I still have it.

From the reading I've done it would appear that 1/2hp should be sufficient to deliver the 7gpm required by my furnace from a pump set at 60'.

I can't understand why the guy that replaced the pump would replace a 3/4hp one with a 2.5hp one? It would, after all, appear to be around three times the required size. It was the same person that installed the original plus I would have questioned the size difference.

Assuming it is indeed 1hp could it have failed in such a manner that it's consuming more than double the power that it should, maybe a short. Could this be related to the apparent steady state pressure drop that I'm observing when the heat pump is running.

I did measure both hots and they are the same so no short to ground. I also verified the meter against a known load and it was spot on.

I didn't have time to get the parts I need to measure the pressure before the PRV and a new guage but will get round to it this week.

9 amps should be full load amps for a 1 HP 220 volt submersible. Pulling 8.5 amps means either your valve or load is restricting the pump, which will lower the amperage OR, your pump is worn and will not pull a full load. Try the amp test while using as much water as possible. The amount of water being pumped has a lot to do with the amperage. If it loads up to 9 amps, your pump is still doing all a 1 HP can do. If the amps stay low, pump is restricted or worn. If it is a three wire motor, it should have the HP on the label along the bottom of the control box.

The pump starts about 8.8A with the pressure at 30psi and drops down as the pressure increases. Its a two wire motor so no control box.

but, I was expecting:

hp = VI/750

or in my case 8.5*220/750 = 2.5hp. Why is this not the case?

For the 2hp motor in my dust collector I see a draw of 7A

7*220/750 = 2hp as expected.

Are pump motors not rated in the same way?

I just know off the top of my head that a 1 HP is 9 amps. And yes subs pull much higher amps than other motors, they always run in the service factor. I can give you the Brake Horse power equation then convert HP to Kwh and amps but, I know a 50 HP 480V horizontal motor pulls 55 amps and the same size sub pulls 76.

8.8 amps is close to a full load but, I’ll bet it started out new at 9 amps is a little worn, which would explain the slight drop in pressure.

Yeah, just found this. Franklin are quoting 8.2A full load and 9.8A max for one of their 1hp 4" motors http://www.fele.com/Manual/AIM_13.htm looks like it is a 1hp after all.

dirtmover said:

....... From the reading I've done it would appear that 1/2hp should be sufficient to deliver the 7gpm required by my furnace from a pump set at 60'.

I can't understand why the guy that replaced the pump would replace a 3/4hp one with a... .

Click to expand...

He also used a PRV that has a very large pressure loss all the while your furnance is flow controlled at a measely 7 gpm...

BTW that 1 hp set at 60 feet, it puts out a lot more gpm than the stated gpm. So IMO, he mistakenly thought bigger is better like many guys, and he also thought the PRV wouldn't wear out the pump and it seems he was seriously wrong on both counts.

Maybe you'll want to get into the number of bathrooms and your peak demand gpm in the house... and the static water level in the well etc..

Gary, you are right that he could use a smaller pump. Getting the exact pumping level and total demand could help size up a smaller pump. Especially if he used a regulator that has less friction loss and could do away with the other regulator on the heat pump itself. However, it is not the regulator that causes wear on the pump. Wear would be caused by the way his regulator is being used that makes it only work when flows of over 8 GPM are required.

The drillers report says the static level is at 12'. There are 3 bathrooms, 5 people. How do we convert this into a peak demand?

Also, how should I make the best of this setup? I don't really wish to replace the pump at the moment though I would consider once the weather gets better if there will be a clear payoff.

Gary, you've mentioned a couple of times that the PRV may not be suitable. I'm not sure that I understand the difference between throttling the flow using a PRV or CSV or simply putting the pump in a well where the water level is lower. For example, is dropping 43psi across a PRV or CSV not equivalent to pumping from 100' deeper as far as what the pump sees? We wouldn't be concerned about the latter so why should the former concern us?

We would need to know the pumping level while running the three bathrooms and the heat pump at the same time. Also, you are not losing 43 PSI across the regulator, you have 43 PSI out and probably about 55 PSI into the reg, so you are losing about 12 PSI across the reg. Other regs, like the one used to make the CSV1 will only lose about 2 PSI across it which would bring up the pressure on your heat pump by about 10 PSI.

valveman said:

Also, you are not losing 43 PSI across the regulator, you have 43 PSI out and probably about 55 PSI into the reg, so you are losing about 12 PSI across the reg.

Click to expand...

OK, that was a really bad example. It was supposed to be a hypothetical question but the numbers I gave matched my pressure measurement so you missed the point. I wasn't trying to make sense of of my observations.

The point was that generally in order to match an oversized pump to your demand you need to throttle the flow somehow. This is what the CSV, and my PRV does. Gary has suggested that PRV may have damaged the pump. I don't understand the difference in the loading on the pump between:
1. throttling using a PRV
2. throttling using a CSV
3. using the same pump in a deeper application where the head of water makes the pump less productive

He doesn't suggest that #2 or #3 could cause excessive wear so why would #1?

“However, it is not the regulator that causes wear on the pump. Wear would be caused by the way his regulator is being used that makes it only work when flows of over 8 GPM are required.â€

In other words the regulator is letting your pump cycle when you are using less than 8 GPM. That is what is happening when you see the pressure fluctuate in the shower and why you do not see the fluctuations when the heat pump is running with the shower. Although, cycling causes several symptoms that lead to the destruction of the pump and motor, a small loss of efficiency is usually not one of them. At 8.8 amps there could be little to no wear in your pump. Throttling will not hurt the pump nor will being in a deeper well, only makes it produce less flow. It should only pull a maximum of 9 amps even when the flow rate is maxed out. You would need to open up enough water usage to get the pressure lower than 30 PSI to do that. Then if it does not load up to 9 amps, you may have some wear, or a restriction on the inlet or somewhere in the outlet of the pump. If not then your pump just cannot build that much pressure at that depth to water with the friction loss and head of all the other equipment. Still wouldn’t know why the pressure would have changed unless the water table has lowered.

So, I took out the regulator and opened it up to check for wear or blockage. All was fine. Before going this far I did price up a repair kit at a couple of local pump shops. They were wanting between $50-$80 for the kit or $180 for a new regulator!!! I guess this puts some of the earlier statements about cost into perspective but I went back into the Home Depot to double check and sure enough they want $47 for the 3/4" and $35 for the 1/2" complete regulator.

I added a nipple for a pressure guage on the strainer cover, on the pump side, and added a second temporary guage on the house side for comparison and calibration.

So here are my observations

- the original guage is measuring about 3psi lower that the new one
- with the PRV fully open the pressure drop across it is not measurable within the accuracy of these guages, maybe 1-2psi (5%).

In the absence of the spec for my pump I have looked at a cross-section from other manufacturers and it would appear that a typical 1hp pump set at 60' with a drawdown, I'm guessing here, of 40' would appear to be able to deliver at least 15gpm @ 40psi.

Now, I'm seeing about 8gpm @ 43psi. I guess this has narrowed things down to the pump and/or pipe.

That would depend on what 1hp pump you have. My 1hp 20 gpm can do around 19 gpm at 40 psi.

bob...