Water Source Heat Pump Deep Pump Sizing

[valveman]

The lower you set the switch on the Mono Drive, the more flow the pump can deliver and the worse the bouncing is going to be. You could use a CSV set lower than the switch and that will keep it from bouncing but, then you don't need the variable speed at all.

The way I am talking about you can install as small a pump as possible for the heat pump. Don't really need much pressure, just volume. Then if you want to irrigate you use another pump to boost the pressure. Also could come out of the heat exchanger and tee off into an electric solenoid valve that feeds a ground storage tank. Level controls in the storage tank open and close the solenoid valve as needed. When the storage tank is full, water from heat pump goes right past the tee off and back down the dump well. Now you have a storage tank full of water and can hook it up a booster pump for irrigation needs. Don't need much pressure to fill a storage tank either so you can still use as small a pump in the well as the heat pump can live with.

[mbartosik]

The bouncing at low pressure really makes the MonoDrive a pain for GSHP.

I think that I am going to have the well guy come back and reduce the pressure, while measuring the pressure at the entrance to the return well. At the entrance to the return well the pressure only needs to be minimal (correct me if I'm wrong here). All pressure at the return well is litterally energy/money down the drain (correct me if I'm wrong).

My guess is that the pressure will have to be dropped to the minimum of 25 PSI that the MonoDrive will handle. Then it is likely to bounce all over the place. Then I need him to show me the variable speed working at the 25 PSI between 7 and 15 GPM (my heat pump requirements).

If it is unable to handle that I'll have to discuss with him what he can do.

[valveman]

It will do the 25 PSI all right, just going to keep bouncing unless you use enough water to get the pressure lower than 25 PSI. Bouncing a million times every 21 days, probably not going to last long.

[mbartosik]

I just spoke to the well driller.

He thought that the electrical energy required was proportional to the volume of water moved (i.e. gpm) and unrelated to the pressure. I explained that I believed otherwise (and so does physics).

He said that I should try reducing the pressure to about 30 PSI, and that having a little pressure at the return well is good, not for now but for the future if deposits build up in the return well, but only a little pressure. He also explained how to reduce the bladder pressure and the sensor pressure.

I didn't get a price from him for swapping out the pump and motor for 1/2 HP, but the implication was in the order of $1000 and I'd then have a pump and motor to place on web auction. So his mistake, my money.

With two pumps in a well, can they both use the same supply pipe or do separate supply lines need to be used? My idea here is to fit a smaller pump in the same well for the GSHP but sharing the same supply line from the supply well head to the basement. Fitting a new supply line from the well head to the basement below the frost line would be expensive. Then I could have the irrigation turn on the existing pump 3/4 HP, and the GSHP control the smaller pump 1/3HP or 1/2 HP. This way less gets sold via web auction. I would probably need a check valve to stop the higher pressure pump from backfeeding the lower pressure pump, and to only run one of them at a time. All getting more complex.

If the pump is forced to cycle until it breaks, then that might be a good thing just as long as it breaks within warrentee and in summer Then I'd maybe not have to fork out another $1000 to get something smaller.

[Bob NH]

I just spoke to the well driller.

He thought that the electrical energy required was proportional to the volume of water moved (i.e. gpm) and unrelated to the pressure. I explained that I believed otherwise (and so does physics).

With two pumps in a well, can they both use the same supply pipe or do separate supply lines need to be used? My idea here is to fit a smaller pump in the same well for the GSHP but sharing the same supply line from the supply well head to the basement. Fitting a new supply line from the well head to the basement below the frost line would be expensive. Then I could have the irrigation turn on the existing pump 3/4 HP, and the GSHP control the smaller pump 1/3HP or 1/2 HP. This way less gets sold via web auction. I would probably need a check valve to stop the higher pressure pump from backfeeding the lower pressure pump, and to only run one of them at a time. All getting more complex.

If the pump is forced to cycle until it breaks, then that might be a good thing just as long as it breaks within warrentee and in summer Then I'd maybe not have to fork out another $1000 to get something smaller.

The well driller is wrong. Power to a pump is proportional to the product of the flow and the head (pressure). Since a centrifugal pump delivers head proportional to the square of the speed, and flow proportional to speed, the power varies as the cube of the speed.

The characteristic curve of the pump scales as follows with speed.
1. Adjust the GPM by the ratio of the actual speed to rated speed.
2. Adjust the head at the new flow point by the square of the speed.

For example, if a pump delivers 10 GPM at 100 ft of head at the speed on the pump curve, then at 80% speed it will deliver 8 GPM and the head at 80 GPM will be 64 ft.

There will be some difference because low flow will reduce the losses in the pump.

You can reduce power a bit by throttling the pump, but you do this at a serious loss of wire-to-water efficiency because you waste energy in the throttling and you drive the operating point away from the most efficient operating point.

It is pretty clear that your well driller is out of his league when it comes to non-standard systems.

I would do some engineering to see if I could match the pump to the system. You can get 3-phase motors that are made to run with variable frequency drives, and small 3-phase converters that run on single phase are not very expensive. It is also be possible to get single-phase converters.

I would look at the system to find the pump that would provide the maximum required flow and head at the standard speed. Then I would operate the pump at a constant lower speed (no variation or feedback control) to provide the lower flow and head.

A simple pump in the size you need is not that expensive and you may be able to use some of your existing equipment. You can bolt a new motor onto most submersible pumps.

[mbartosik]

Thanks Bob.

Your post requires careful reading but is a lot of help. Care is required for accurate understanding (this is meant to be praise not a critism). Your comments are most useful.

I am sad to say that I spoke with 3 well drillers and they all proposed about the same system just at different prices, although one was 1HP! Today I spoke to the HVAC company owner that installed the heat pump, and my observation is similar, he said that he liked to see about 45 PSI at the pump to get enough flow. I told him that flow and pressure are different, and the heat pump required flow not pressure and his answer was "um, yes". "Um, yes" is not good to hear when you are paying over $30K plus the well costs, especially from someone who should know better since he does air duct pressure and flow calculations (most likely table lookups rather than calculations).

The local utility (which is state owned) is giving me a $4000 rebate ($800 per ton). They do this to encourage GSHP because of the efficiency. I know who are the right people to talk to at the utility, and I want to encourage them to require the pump sizing and flow and pressure settings to be reviewed by a qualified professional to qualify for the rebate, otherwise they are litterally paying to have energy pumped down a hole. They already do this for the solar rebates. My whole motivation is energy conservation (not the money).

Over the weekend I will try to lower the pressure. You point out that power is proportional to the cube of the speed, I had forgotten that (I last had to do such physics a long long time ago). The Franklin MonoDrive (what I have) has a range of 30Hz to 60Hz (or about < 1800 rpm to about < 3600 rpm) that I think means that the power range should be 1/2^3 or 1/8 (12.5%) to 1 (100%). Since input power is 240v * current * power factor, then I should be able to get in the order of a factor of 8 variation in input current (assuming a wishful linear efficiency curve for the inverter). So far (without adjusting pressure much, mostly adjusting flow) I have been able to vary the input current by about a factor of 2, so there should be a lot more variation available by adjusting pressure.

Also tonight while monitoring current I noticed the current sit at a steady 1A for several seconds, which is encouraging. If I can get to a point where it draws only 1A without cycling then I think that will be sufficient.

As I make adjustments I'll record pressure, flow, and current (and input power) and post my results here.

[Bob NH]

)
As I make adjustments I'll record pressure, flow, and current (and input power) and post my results here.

You should also try to measure pressure drop across the water side of the evaporator along with flow. That requires a pressure gauge on both ends.

You need to know the pressure drop across the evaporator, and across other elements of the water system, to correctly select the pump if you ever want to change it.

To get maximum efficiency on the water side, the pressure in the return pipe (which should go into the water in the return well) at the elevation of the water in the return well should be close to atmospheric pressure (0 psi gauge). That may cause the water pressure at the discharge from the evaporator (if there are no unreasonable restrictions) to be less than atmospheric pressure. You shouldn't have cavitation in the evaporator as long as the pressure at the discharge end is above about 5 psi absolute (Vacuum of 20" Hg).

If there are any valves in the water circuit, try running the system with all of them wide open and see what the flow is. If the flow substantially exceeds what you need for the evaporator, that is a sign that the pump is too big or is running too fast. Slowing it down to the point where you get required flow with no restrictions in the water circuit should give minimum power consumption.

Higher velocity on the water side will increase the heat transfer rate so the extra pumping power may be balanced by lower power for the compressor. There is a limit to that because the the evaporation side is often the limiting factor.

[mbartosik]

My original plan (until I found some of your posts on this forum) was to do some fine tuning of the power required by the well pump vs the improvement in heat transfer due to higher fluid velocity (making compressor work less hard). That is to say adjusting flow without concern for pressure, but when I saw your posts.....

My plan for this was to get two precise (but not necessarily accurate) logging thermostats, i.e. 0.1 F resolution with RS232 or RS485 data logging to measure the fluid temperature difference across the compressor. First ensuring that I know their relative offsets at the same temperature. Then I could use fluid flow, fluid temperature delta, and thermal mass to calculate the energy extracted from the fluid. Using a current clamp and RS232 logging I could measure the current draw on the combined well pump and compressor (need to know power factor or use a KW meter). Then I could tune the flow to find the "sweet spot" where most energy is extracted for least current draw (energy in).

The idea with RS232 was have my computer dynamically calculate the efficiency coefficient (COP) based on time averaged data because thermocouples don't have an instantaneous response. Effectively the computer would use the data logging to do a little integration in realtime.

If I manage to drop the well pump current draw enough (by dropping the pressure). Then I'll do this fine tuning. But it is clear that my main inefficiency is too much pressure.

Also if I manage to drop the pressure a sensible amount (down to 20 to 30 PSI) I might dig up the return well and fit a 'T' so that I can fit a pressure gauge there. The ideal would be to fit a pressure transducer so that I can rebury it (below the frost line) but leave the wires on the surface where I can measure with a multimeter. That's a load more work so it will have to wait.

If I do end up installing a smaller pump and motor, what do you think my chances are of getting a reasonable price at internet auction for the pair that I have now which are only a few weeks old?

[valveman]

I know Bob meant 8 GPM at 64 ‘ not 80 GPM and he is exactly right. The problem is that you lose head by the square of the speed so, you can’t slow it down enough that saving energy by the cube of the speed makes any difference. In other words you will save more energy by having exactly the right size pump compared to slowing a pump down with a drive or restricting the flow with a valve. Personally I like your two pump in the same well idea except, it may be hard to get them to fit. Done it lots of times in 8" and larger casing but, 6" and smaller is tough unless it is shallow enough for a jet pump instead of a sub. I think BobNH wrote about this earlier. Anyway the Mono Drive that he has is a single phase drive and the Sub Drive is a three phase drive. Only difference is that they now use a go/no go switch instead of a pressure transducer but, the effect on the power consumption is the same. Systems with pressure transducers can be slow reacting and not very reliable, usually sticking in the full speed position. So the go/no go switch solved some of the slow reaction problems and caused a few others. The bouncing and weaving is just how it maintains the speed required. What I hear from Heat Pump customers a few years into it, is that the energy consumption is less important than the life of the pump system.

I think my idea with a ½ HP sub in the well and a ½ HP booster on top is as good as the two pumps in the well idea but, will fit in 4"casing and work from over 50' deep . Instead of a 1 HP in the well that can be slowed down or restricted to maybe 3/4 HP, I will have a ½ HP restricted a little so pulling maybe 4/10 HP. Then the ½ HP booster will only come on if I need pressure for the house or irrigation. The ½ HP sub has already been running 24/7 for 10 years feeding the pond. Since I am already running the water, the heat exchange and the water supply to the house should be a bonus. This sub has already been running for ten years without shutting off, so I know that running steady and not cycling or ramping in anyway is what makes it long lasting. If the pump system will last a long time and only use 4/10 HP on average for the heat pump, it should be worth it.

[Bob NH]

In most heat exchanger applications with variable speed drives you can usually just set the speed with a potentiometer or manual control and let it run. The water side pressure drop is proportional to the square of the flow, and characteristics of the water circuit shouldn't change much.

The problem arises when you have variable demands on the pump supplying the evaporator. That is why it is hard to use that pump to irrigate or supply other needs while running the evaporator.

[valveman]

Yes you can set the speed with a potentiometer or manual control but, anytime you vary the speed you are burning energy compared to having the right size pump.
See http://cyclestopvalves.com/comparisons_13.html

There are also many negative side effects to VFD control that you will never hear about from the people who make or sell drives.
See http://cyclestopvalves.com/comparisons_5.html

Stray voltage has also become a major problem with VFD controls. The following link is just one of many you can find doing a search for VFD stray voltage.
See http://www.electricalpollution.com/

[mbartosik]

Here are some results of the input power saved by adjusting flow and pressure:

At 55 PSI and 15 GPM the current draw was 5.2A i.e. 1.25KW
At 55 PSI and 7 GPM current draw is about 2.8A i.e. 672W
At 45 PSI and 7 GPM current draw is about 2.4A i.e. 576W
At 40 PSI and 7 GPM current draw is about 2.3A i.e. 552W
At 35 PSI and 7 GPM current draw is about 2.2A i.e. 528W can only get 14 GPM max
At 30 PSI and 7 GPM current draw is about 2.1A i.e. 502W can only get 12 GPM max
At 25 PSI and 7 GPM current draw is about 1.9A i.e. 456W can only get 10 GPM max

35 PSI is the minimum required pressure to get the 14 GPM required for stage 2 of the compressor.
I could probably drop to about 18 PSI if I only wanted stage 1 of the compressor, and that might get me to 400W.
Since I only run stage 2 rarely, I might drop to 30 PSI. Running on stage 1 mostly at 35 PSI and 528w means that I'm wasting about 100w in excess pumping capacity.

Initially the installers configured the water side to 55 PSI and 15 GPM constantly using 1.25KW, wasting about 850W down the return well.
I asked for a second valve to be fitted to reduce flow when only 1 stage is used, this reduced power required most of the time to 672W.
Then I started reading this forum, and now I've got the input power down to 528W.
Total power saved so far is about 60% of pumping power, or about 20% of total system power (including heat pump load).

The Franklin MonoDrive manual has a table of limit switch and tank pressures. The lowest pressure in the table is 25 PSI, but it does not say that a lower pressure cannot be used. The lowest that I could likely go is 18 PSI for stage 1 compressor on, not sure how I'd run stage 2 then.
What do you think to running that MonoDrive at 18 PSI?

If I was to compare this against the energy used by using a fixed speed pump, if I was to use 1/3 HP (the smallest that I think that I can buy)...
1/3 HP == 0.25KW
assuming 40% efficiency between motor and pump losses, that would require 0.6KW of input power.
It looks like I'm already using less power than the minimal simple system, but I know that it is not that simple.
My other way to reduce power usage is to use a smaller (1/2 HP motor and pump), but still using the MonoDrive, the 1/2HP would be in a sweeter point of the performance curve.

[valveman]

It should not hurt anything to run at 18 PSI, if the switch will turn down that low. I also don’t see why you can’t use two switches. Just put in a relay that lets the higher pressure switch override the lower switch when using stage two. Also like BobNH said you can put a ½ HP pump on that 3/4 HP motor and lower the power a little more also. My question is, how are you getting these steady amp readings? Isn’t the amp meter still bouncing while the pump is maintaining the pressure you have adjusted the switch to hold? Also how much in dollars is saving 850W in your area. If I am thinking right this morning, in my area that would only be less than 10 cents per day, $3.00 per month, $36.00 per year. Sounds minimal to me, am I adding wrong?

[mbartosik]

I might use two switches, however, then the problem becomes the bladder.
The MonoDrive manual says to set the bladder pressure at 70% of the switch pressure. That means
18 PSI on switch set bladder at 13 PSI
35 PSI on switch set bladder at 25 PSI
I don't have a way to change the bladder pressure dynamically. If I used 35 PSI on the switch but 13 PSI on the bladder then I would risk damage to the bladder (a burst), and the other way round would be ineffective.

The bladder appears to be oversized also, I have an Amtrol WX203 tank volume of 32 gallons max accept factor 0.35 thus 11 gallons variation. The MonoDrive manual says that I only need 4 gallon tank total capacity (so x8 larger than needed).

As for the steady amp readings I should have said in the post. I averaged them, there was a clear seeking up and down, and I choice the mid point. For example (these are actual measurements):

At 40 PSI and 7 GPM input current varied between 2.2 and 2.5, the average appeared to be 2.3A.
At 40 PSI and 8 GPM input current varied between 2.4 and 2.9A, the average appeared to be 2.6A.
At 30 PSI and 7 GPM input current varied between 2.0 and 2.2A, the average appeared to be 2.1A.
At 25 PSI and 7 GPM input current varied between 1.8A and 2.1, the averaged appeared to be 1.9A.

There was a clear seeking up and down. I am unsure of the frequency, probably about 20 cycles per minute. The seeking didn't get any worse as the pressure dropped, I guess that's because this time I lowered the bladder pressure with the limit switch. I understand that this seeking is one of the one of the potential problems (e.g. worn wires). I think that they could use smarter software in the MonoDrive in conjunction with a lower limit pressure switch set just a couple of PSI below the other to almost reduce the seeking to zero with a good sized bladder.

Your $36 per year calculation - it may be right for you, but here is mine:

In my area electricity costs about 18c per KWh. I consider that to be under priced considering the uncharged costs (foreign energy dependance, wrecking the planet etc.).

So for a 100W saving in pump efficiency, assuming 30% duty cycle for 8 months a year (heating and cooling), that's roughly 2000 hours, or 200KWh.
Or $36 per year for just 100W or $288 for 850W.

Now since my motivation is energy conservation. My whole house will be powered by solar, so far with net metering (no batteries) I had 100% net solar power before the heat pump was installed. I have to look at the long term. Assuming a 10 year payback on solar, that $288 in one year is equivalent to about $3000 over 10 years, which is about the cost of installing 2 x 200W DC solar panels after install costs (before rebates).

As for changing the pump or motor or both to get to the "sweet point", then I have to work out what my likely power saving will be. I have to save at least 100W to make it a consideration even given the 10 year cost. So I'll try to find out the speed that the motor is running at my preferred load point maybe 30 PSI and 7 GPM, and then look at a pump and motor efficiency graph. To find the motor speed I will try to borrow a good frequency meter or an osciloscope (and look at the MonoDrive output wave form).

So if I was to change only the pump or the motor, is it the pump that should be downsized, or is it better to do both?
If only the pump is down sized will the motor then naturally demand less power, because of less brake power by the pump?
?

[Bob NH]

As for the steady amp readings I should have said in the post. I averaged them, there was a clear seeking up and down, and I choice the mid point. For example (these are actual measurements):

At 40 PSI and 7 GPM input current varied between 2.2 and 2.5, the average appeared to be 2.3A.
At 40 PSI and 8 GPM input current varied between 2.4 and 2.9A, the average appeared to be 2.6A.
At 30 PSI and 7 GPM input current varied between 2.0 and 2.2A, the average appeared to be 2.1A.
At 25 PSI and 7 GPM input current varied between 1.8A and 2.1, the averaged appeared to be 1.9A.

There was a clear seeking up and down. I am unsure of the frequency, probably about 20 cycles per minute. The seeking didn't get any worse as the pressure dropped, I guess that's because this time I lowered the bladder pressure with the limit switch. I understand that this seeking is one of the one of the potential problems (e.g. worn wires). I think that they could use smarter software in the MonoDrive in conjunction with a lower limit pressure switch set just a couple of PSI below the other to almost reduce the seeking to zero with a good sized bladder.

Your $36 per year calculation - it may be right for you, but here is mine:

In my area electricity costs about 18c per KWh. I consider that to be under priced considering the uncharged costs (foreign energy dependance, wrecking the planet etc.).

So for a 100W saving in pump efficiency, assuming 30% duty cycle for 8 months a year (heating and cooling), that's roughly 2000 hours, or 200KWh.
Or $36 per year for just 100W or $288 for 850W.

Now since my motivation is energy conservation. My whole house will be powered by solar, so far with net metering (no batteries) I had 100% net solar power before the heat pump was installed. I have to look at the long term. Assuming a 10 year payback on solar, that $288 in one year is equivalent to about $3000 over 10 years, which is about the cost of installing 2 x 200W DC solar panels after install costs (before rebates).

As for changing the pump or motor or both to get to the "sweet point", then I have to work out what my likely power saving will be. I have to save at least 100W to make it a consideration even given the 10 year cost. So I'll try to find out the speed that the motor is running at my preferred load point maybe 30 PSI and 7 GPM, and then look at a pump and motor efficiency graph. To find the motor speed I will try to borrow a good frequency meter or an osciloscope (and look at the MonoDrive output wave form).

So if I was to change only the pump or the motor, is it the pump that should be downsized, or is it better to do both?
If only the pump is down sized will the motor then naturally demand less power, because of less brake power by the pump?
?

Your economic/power analysis should be broadened to consider the cost of the solar panels and evaporator.

Two solar panels with 200 Watts capacity will collect how many kWH per year? At what cost per kWH?

Part of the problem comes from the pressure drop in the evaporator. You might be able to double the size of the evaporator (put another one in parallel) and cut the pressure drop by 2/3. That would also reduce the temperature difference between the water temperature and the suction temperature, increasing the Coefficient of Performance.

You should get rid of any device that causes pressure drop in the water circuit. Too often, system sellers cut costs on heat exchangers and pipes because the customer is not aware of how much that affects energy requirements.

If you put in a second evaporator, then you could consider switching them from parallel to series connection when you need less flow. Some hydraulic analysis would tell you if that would save energy in either the pump or compressor. Usually, more power for heat exchanger pumping will save some power in the compressor. You will want to be able to measure the suction pressure and/or the compressor current as you make changes on the water side of the evaporator.

With respect to the pump, I suspect that you will save more Watts per dollar by changing the pump head rather than the motor. However, that should be analyzed before making the purchase. The pump willl draw less power and require less electricity if you reduce the load on the pump.

A "correct size" single-speed pump with motor should no cost more than about $500. The rest is margin somewhere between the distributor and you.

If the discharge well pipe doesn't go all the way to the water level, then either it should be extended, or the casing should be sealed so that it will withstand a vacuum at the top. That will allow you to recover more of the elevation head when the water is discharged.