Well Water Assisted ASHP

[pilko]

Hi everyone,
My heat pump is a Carrier 3 1/2 ton air source. I also have a deep well that can produce at least 10 gpm at 50 deg F.
I would like to pump the well water through a coaxial heat exchanger to heat a portion of the refrigerant in parallel with the outdoor coil.
The proportion of refrigerant passing through the coax will be controlled by a motorized mixing valve depending on the outdoor temperature.
If for example the coax is able to reduce the water temperature by 10 F at 3 gpm it would be giving up 18000 BTU to the refrigerant.
Is This feasible?

Thanks,

pilko

[Peter_1]

This is a very good idea but why not install the coax coil in serie with the aircoil and removing the 3-way valve. If you want to extract the heat out of the air, switch o the fan and switch off the pump. If you want to use the well, start the pump and switch off the fan. We have done this already.

[MikeHolm]

Hi Jeff, how's it going?

[pilko]

This is a very good idea but why not install the coax coil in serie with the aircoil and removing the 3-way valve. If you want to extract the heat out of the air, switch o the fan and switch off the pump. If you want to use the well, start the pump and switch off the fan. We have done this already.

Thanks for the response Peter.
I had thought of putting the coax in series with the o/d coil but I would need to run both fan and coax since I think I can only pull 18000 btu from the water.
I also thought that if they were in series then the second coil in the refrig flow
(heating or cooling) would have low efficiency as the delta T refrig to air
or delta T refrig to water would be low.
BUT ---I am an amateur and could be making incorrect assumptions.

@Mike ---doing good, -- good to hear from you

pilko

[mad fridgie]

There are problems with your design, shift liquid feed (liquid) not the vapour side.
You need to ensure that your water evap can not freeze.

[pilko]

There are problems with your design, shift liquid feed (liquid) not the vapour side.
You need to ensure that your water evap can not freeze.

Thanks for the reply mad.

Did You mean like this?.

[mad fridgie]

Sorry not much time.

No

First consideration is protecting your heat exchanger, so it does not freeze.

Best with two expansions valves, forget modulating the vapour.

You need to look how the energy is going to flow to determine design.

So going back to the first point you do not want the refrigerant temp in the water evap to fall must below 0C, so theres a starting point

How did the defrosting go without the fan running?

[pilko]

Sorry not much time.

No

First consideration is protecting your heat exchanger, so it does not freeze.

Best with two expansions valves, forget modulating the vapour.

You need to look how the energy is going to flow to determine design.

So going back to the first point you do not want the refrigerant temp in the water evap to fall must below 0C, so theres a starting point

How did the defrosting go without the fan running?

Thanks for the reply mad,
I was planning on monitoring the temperature of the coax discharge water and using the
info to control the mixing valve. In other words control the amount of refrig through the coax to keep the water temp above about 35 F.
( still defrosting without fan running --- big improvement )

[pilko]

@ Peter_1 and mad fridgie,
I would really appreciate your comments on my replies to your posts.
I am niether a refridge Eng. nor an expert in thermodynamics.
Any further help would be greatly appreciated.

[The Viking]

Pilko,
What you will find is that us time served fridge guys prefer to run everything according to the KISS principle.

As soon as you add electronic controls and uneccesary valve we will start to think about what would happen when these fails. If the failure means that there would be a total and expensive system breakdown, like a compressor failure, we will discard that solution and look for simpler and more reliable options.

When it comes to the question of why regulate on the liquid instead of the gas, in it's gas phase there are a lot of variation within the refrigerant especially if you are looking at a mixed phase as you would find between the expansion device and the evaporator, in the liquid phase on the other hand the refigerant will be solid.

Why in series and not in paralell? Your end game is to get the refrigerant as warm as possible before it returns to the compressor, hence a longer distance to travel in the warmer mediums makes sense. This also requires less controls and mechanical valves and a failure will not by default blow the compressor up.

I hope this will have answered most of your questions,

[pilko]

@ Viking,
Thanks for the information.
Just to be sure are you saying install the coax as I have shown by the black rectangle in the attached drawing?

[The Viking]

Pretty much yes,

You have to decide which conditions the system will be designed for, if it is for when the well water is warmer than the ambient temperature the coax should come after the outdoor coil as the refrigerant flows in heating mode.

Also, make sure there is something in place to protect the coax from freezing the water inside it.

[pilko]

Pretty much yes,

You have to decide which conditions the system will be designed for, if it is for when the well water is warmer than the ambient temperature the coax should come after the outdoor coil as the refrigerant flows in heating mode.

Also, make sure there is something in place to protect the coax from freezing the water inside it.

Thanks Viking.
My well water is pretty stable year-round at 10 C. In the winter the water is almost always
hotter than the air. But, if I put the coax after the coil (based on flow in heating mode)
then it will be in gas, which if I understand correctly you were opposed to in post #10.

[The Viking]

Ah,
No, what I and the mad fridgie was against was trying to regulate and balance flow by a valve in the gas or rather the gas mixture line..

Also be aware, even water at +10ºC will freeze if the heat-exchanger itself get too cold.

[pilko]

Thanks Viking,
"Also be aware, even water at +10ºC will freeze if the heat-exchanger itself get too cold."
You're quote brings up another issue:
--With a series system, if I lose the water flow for any reason, or if I do not have enough water I would have to quickly turn off the heat pump and drain the coax. I would not be able to restart until the coax is fully drained and dry.

[pilko]

Hi again,
I have moved the diverting valve to the liquid and put exp. devices in each branch.
Providing I protect the coax coil from freezing, are there any other problems with this system?
If not, can someone help me determine the size of the coax coil?

pilko

[mad fridgie]

Hi again,
I have moved the diverting valve to the liquid and put exp. devices in each branch.
Providing I protect the coax coil from freezing, are there any other problems with this system?
If not, can someone help me determine the size of the coax coil?

pilko

Looks better, still no protection against freezing.
This can be either an EPR (this goes on the outlet of the co-ax evap) Open on rise of inlet pressure. (the pressure/temp never drops below a preset) or hot gas by-pass, a valve that goes between the discharge and between the EXP and the coil. (similar sort of result)
The motorised valve, not sure if readily available.

[pilko]

Thanks for the response mad,
To protect the coil against freezing I was planning on doing the following:
1 Maintain the coax coil outlet water temp at 37F by loop control of the diverting valve. (lower temp less refrig.)
2 In the event of either loss of water flow or the coax coil outlet water temp falling below 35F then automatically shut down the well pump and dump the water from the coil.

pilko

[hyperion]

The refrigeant quality motorised diverting valve may be a problem to obtain. You might want to consider a simple electrically operated solenoid valve to give a simple on/off control of the refrigerant for each circuit. This could then be controlled by the output from your water temperature sensors.

[mad fridgie]

Thanks for the response mad,
To protect the coil against freezing I was planning on doing the following:
1 Maintain the coax coil outlet water temp at 37F by loop control of the diverting valve. (lower temp less refrig.)
2 In the event of either loss of water flow or the coax coil outlet water temp falling below 35F then automatically shut down the well pump and dump the water from the coil.

pilko

Controlling the water temp (at these temps) is after the fact, you not controlling the surface temp of the heat exchanger. For example, if you start to freeze the coil, heat transfers reduces, reducing the temp drop of the mass, however at the point of transfer ice could still build.

Instead of a coaxial , you would be better with a more open type heat exchanger, in its very basic form a pipe in a bucket. This way ice build up is not a major problem. Something like a shell and coil.
http://www.vaportec.co.nz/main/?page_id=211
for example.

[pilko]

@ hyperion,
I would prefer to use proportional control. For example when the air temp is <-5C I would send more refrig through the coax and when the temp >+5C I would send more refrig through the outdoor coil. (I have attached valve info.)

@ mad fridgie,
I was thinking of using a coax because I can get a double walled, vented model, which would be safer if I return the discharge water back into the top of my well.( I have attached a link.)
http://www.vaportec.co.nz/main/?page_id=209

[The Viking]

My personal opinion?
- Too complicated and likely to blow a few compressors in the R&D process before you get it right, sorry.

But, I do have a question...
If you got access to well water, why keep an ambient air coil? Why not just convert it to a GSHP?

[pilko]

My personal opinion?
- Too complicated and likely to blow a few compressors in the R&D process before you get it right, sorry.

But, I do have a question...
If you got access to well water, why keep an ambient air coil? Why not just convert it to a GSHP.

My well can only produce about 3 GPM at 10C. About 5KW

[pilko]

@ Viking,

"My personal opinion?
- Too complicated and likely to blow a few compressors in the R&D process before you get it right, sorry."
You make very good points.
If I go with series operation I think I should add a second reversing valve to ensure the air coil is always the first in line. In heating mode,the air is nearly always cooler than the water and therefor should give up it's heat first. In cooling mode the air is nearly always hotter than the water and therefor should absorb the heat first. --- attached heating and cooling modes.

[JackPreacher]

Hello Pilko,
You just have to look at my thread to see what happens to a HX that freezes at the interface - catastrophic damage. Viking is correct, use an open loop heating transfer...you could incorporate the coax somehow to transfer well water to the back of the fan unit so transferring some btu's across the fan coils.

Good luck

"Massive Refrigerant Leak into plate heat exchanger"

[JackPreacher]

Maybe an old car radiator??

[mad fridgie]

@ hyperion,
I would prefer to use proportional control. For example when the air temp is <-5C I would send more refrig through the coax and when the temp >+5C I would send more refrig through the outdoor coil. (I have attached valve info.)

@ mad fridgie,
I was thinking of using a coax because I can get a double walled, vented model, which would be safer if I return the discharge water back into the top of my well.( I have attached a link.)
http://www.vaportec.co.nz/main/?page_id=209

Used plenty have couple in my workshop. Heat transfer is reduced quite a bit.

this still does not resolve the low refrigeration temp issue.

You need to install a constant pressure device on the water evap. along with other devices, you can dramatically reduce bursting the evap.

[pilko]

@ JackPreacher
"Maybe an old car radiator??"-- Tried it this past winter with a new $100 truck rad. --- not enough delta T between air and water. Also the rad restricts the air flow, so what I gained from the water I lost from the reduced air flow.
@ mad fridgie
"You need to install a constant pressure device on the water evap. along with other devices, you can dramatically reduce bursting the evap." --- can you explain this more fully to a non-refrig guy?

pilko

[mad fridgie]

The pressure determines , the temperature, you do not want the pressure/temp relationship to drop much below 0C,
When load is low, the pressure will fall,so will the temp. So in the evap you can either hold the pressure back or you can add pressure (as vapour not a liquid), thessekeep the temp with in the evap at set point regardless of what the pressure is elewhere

[pilko]

Would this system be better? The compressor would be safe but the water/glycol EX would still be at risk of freezing the water, although it could be simply a car radiator submerged in a flow-through tank of water.

[JackPreacher]

That's more like it. I have to build a similar array for my fell water but will construct a copper array as old vehicle radiators don't like sub-zero temperatures even if they're not iced up.
Is there any way of introducing a safety device in your spring water/glycol circuit in case a pinhole leaks occurs in the rad that reduces pressure and increases the possibility of ice build up in your Glycol/Refrigerant HX?

[pilko]

@ JackPreacher,
I read your thread and I am learning a lot from it -- my condolences.

"Is there any way of introducing a safety device in your spring water/glycol circuit in case a pinhole leaks occurs in the rad that reduces pressure and increases the possibility of ice build up in your Glycol/Refrigerant HX?"
---I am a bit thick, can you explain why a leak in the rad in the water/glycol circuit would cause reduced pressure and ice build up in the Glycol/Refrigerant HX

[JackPreacher]

... can you explain why a leak in the rad in the water/glycol circuit would cause reduced pressure and ice build up in the Glycol/Refrigerant HX

Just theoretical forward planning - if there is a leak the glycol will leach out into your spring water reducing the circuit's capacity for ∆T across the HX, therefore reducing the temperature at the interface of glycol & refrigerant causing temps to drop below the freezing point of the glycol [mix ratios being the defining factor] and then, well............welcome to my world!!

You could design in a gradient that would drain out completely if a leak did occur hence avoiding/delaying ice building in the glycol & refrigerant HX........?

The pipe in a barrel [kind of Vikings suggestion] is exactly the same but I personally would feel a lot happier with a brazed copper array as opposed to a vehicle radiator......at least then you can easily inspect the array and add to it if it needs more surface area.
Do you think all your arrays/HX would need to be common grounded to avoid pinholes forming?

[JackPreacher]

It is quite uncanny the similarities between your proposed project and my TOTALLY WRECKED project - if anything I'm pleased my disaster is making you think things through a hell of a lot more than I did !

[pilko]

@jackpreacher,
What if I monitor the water outlet temp and stop the glycol pump if temp < set point?

[mad fridgie]

you install the right valve to do the right job.
Or install pipe work in a tank, with the well water running through and use it as an ice builder if it gets to that point. (which is very easy to control)
Every time you add another loop you loose efficiency, which the whole point "efficiency"

[pilko]

@ mad fridgie,
I have great respect for your knowledge and your advice. Your advise has helped me very much in the past. The problem is, on this issue, I do not understand what you are recommending due to my lack of knowledge in the subject of refrigeration and thermodynamics.
If you or someone else can explain in simple terms I would greatly appreciate it.

[The Viking]

Hi Pilko,
I think you just nailed it when it comes to both Mr. Mad's and my own reservations to this project.

To pull anything like this of, any manufacturer would be prepared to spend a small fortune on R&D, they would have access to experienced R&D engineers with plenty of experience and they are still likely to calculate on blowing several compressors before they get it right.
In your case, is it really worth it for you?

What you are trying to achieve is well outside the norm, personally I have been in this industry well over 20 years and most of that time I been employed to resolve problems and find the unusual solutions and I'm not sure I could get your idea to work.
Sorry.

For starters, your variable speed pump... Why would it be controlled on the well water's leaving temp?

Anyhow, if your well currently not are able to deliver the heat you require what can be done to increase this? A beefier well pump or bigger bore lines will most likely be cheaper than a few blown compressors...

Or as Mr. Mad said, put a few coils in to a bathtub with the well water flowing through it and then install this coil in series with your outdoor coil. Will the potential energy saving be worth it? Who knows but you are more likely to end up with a system that actually works.

[pilko]

@ The Viking,
thanks for your response and patience. There is confusion on my part, created by me by posting differing systems.
When you quoted --- "To pull anything like this of, any manufacturer would be prepared to spend a small fortune on R&D, they would have access to experienced R&D engineers with plenty of experience and they are still likely to calculate on blowing several compressors before they get it right.
In your case, is it really worth it for you?"
-- were you reffering to my system in post 30. In that post I mentioned a radiator submerged in a water flow-through tank? --- Also the coils are in series.
--"For starters, your variable speed pump... Why would it be controlled on the well water's leaving temp?"
----I was planning on using a closed loop PID controller so that as the water temperature dropped towards a set-point, the glycol pump would slow down to give less cooling and maintain the temperature.
-- My main question is, would the circuit in post 30 keep the compressor safe and minimize the risk of freeze-up to the bathtub HX ?

[mad fridgie]

tubes in a tub/tank are designed to freeze with out causing the pipes to burst, more commonly called an ice bank. The water/ice can expand away fro the tube into the space in the bath.
An earlier question, on fixed refrig system (which have) the system will always be at an equalibrium. The equilibrium and refrigeration temp is determined by the load (what you are cooling), if you do not have enough the refrig temp will drop (the reason for you adding the water H/E), but this not big enough on its own, so to absorb energy from both the air and water, the coldest source will determine the temp of the refrigerant.
Remember you dealing with latent heat, so give or take the refrig temp is constant.
I have no problem with your parallel circuit, because I know you have smarts to control, but you have cover the freeze issue which is real, in an inclosed heat exchanger. I do not see the point of loops cooling loops. You are unlikely to get a major benefit, maybe a minor.

note to Viking, just "Mad" is OK

[pilko]

@ mad fridgie and Viking,

--- Thanks for all your help. Am I still not understanding, or will the attached system work?

--- I'm thinking I should add the second reversing valve to ensure the air coil is always the first in line. In heating mode,the air is nearly always cooler than the water and therefor should give up it's heat first. In cooling mode the air is nearly always hotter than the water and therefor should absorb the heat first.

--- Temperature probe in the water detects when temperature drops below set point and stops compressor.