The ultimate heat pump...

Mark Eatherton

Greetings fellow R Wallies. I am designing (in my head for now) the ultimate water to water heat pump. It needs a capacity of around 48,000 btuH, so I am thinking four one ton compressors piped in parallel through FPHX's. I'd LOVE to dedicate one compressor as the lead, and if possible, connect it to a variable speed motor controller (0 - 10 VDC control signal). I'd always start on high speed, then roll compressor motors RPM's down to match the load. I will build my own water storage tank out of 3/4" plywood and 4X4 treated lumber posts with a polyethylene liner. This open tank will have two coils in it. One for input (multi source) and one for extraction. The extraction loop will be glycolized, as will the input coil. The storage medium will be water, and I do expect to freeze this tank solid. My math works out that at full output capacity, and only utilizing 80% of the open tanks capacity (leaving room for expansion when mass expands due to freezing) I'd have 19 hours of reasonable COP (roughly 3:1) before I start getting into trouble, and in reality, this heat source will only be used to maintain my home at standby conditions, that being 40 degrees F. When I want it warmer, I enable my LP Modcon boiler. Anyway, to my questions.

1. I realize the need for check valves on the discharge of the compressors.
2. Is there a need for separate oil separators, reservoirs, etc like are used with larger compressors?
3. Is it possible to vary the speed of these little compressors without compromising their life expectancy significantly?
4. How big does the FPHXers need to be in order to properly move the btu's around.
5. What kind of water temperatures can I expect on the load side of the system with an EWT to the HXer of 40 degrees F and a corresponding source side temperature of 32 degrees F.

I do not intend to vary the flow on the source or load size (hydronic), and plan on maintaining a narrow temperature differential. I also realize you will most probably ask me for additional information that I missed on the first pass...

Thanks in advance for your input.

ME

Zman

Mark,
I am interested to see the responses on the heat pump side of things.
How are you modeling heat transfer and storage in a block of ice? What do think of the idea of burying a storage tank and filling it with an inexpensive brine of some sort.
I think it would be beneficial to buffer your storage from the fluctuations of the outdoor air temp.
Carl

Jack

"I'd always start on high speed". Mark I believe that would be quite a bit less efficient than rolling into it and building rather than bringing all that load in at once.

Mark Eatherton

Carl, this is a slight variation of the Thermal Battery Storage system, as seen at YouTube channel for the RPA. The developers have a program made by Trnsys for analysis, and have been testing it and fine tuning it. To quote one of the co-developers, "Energy is where you find it…"

The intent is to keep the tank at between 70 and 32 degrees F. Hence it will not be insulated. The lower the tank temperature, the better as it pertains to being able to recharge the tank with solar thermal, ambient air via a FCU, reverse snowmelt, woody biomass or whatever has a temperature gradient greater than the storage tank. By using a brine or other antifreeze solution, you lose the advantage of phase change. When water goes from 33 F water to a solid block of ice at 32 degrees F, it carries an additional 143 btu's per pound of latent energy that would not be part of the picture with a brine solution. Tank will be constructed in a tapered manner to allow the ice to push up and not exert stress on the walls of the tank. It will have two copper coils in it, one for input (bottom of tank profile) and one for extraction at the top of the tank profile. There will be antifreeze in the two coil circuits.

Jack, the only reason I think it would be wise to start the compressors on high and quickly drop back to low is to get things (refrigerants, oil etc) moving as it was designed for first, then back the RPM's down. If the refrigerant experts here say it is OK to start low and go up slow, then that will end up as my modus operandi. For the price I am paying for the 1 ton compressors, (less thatn $100 each at WW Graingers) I seriously doubt they have DC ECM technology on board.

As it pertains to the water side of the system, the circs necessary will be DCECM circulators, but will be small ones at that. Hence, no real advantage to soft starting them.

Just thinking back to the beginning of VS pumps, they always "kick started" them, and then rolled back to whatever speed the variables called for. I know just enough about refrigeration to know I really DON'T know enough, and know when to say when, and when to ask for help… H E L P !!!

Just so everyone knows, I have a good friend/former employee who is licensed in refrigeration, carries necessary certs and tools and will be my mechanic on the refrigerant side of things.

One reason I am going for staging is to avoid the need for a buffer tank (short space on the hydronic side of things) and to keep the compressor(s) running as much as possible.

Still learning after all these years…

Thanks!

ME

pecmsg

Copeland Scroll™ Variable Speed Compressor

Zman

Where will the tank be located?

Mark Eatherton

Thanks pecmsg. I will check it out.

The tank will be outside, under a deck in a semi insulated room. They usually do a direct bury (30" below grade), with no insulation for the tank.

ME

Harvey Ramer

Mark, setting up these compressors up in parallel with variable speed operation might be a bit more costly than a buffer tank.

Nontheless, I will donate my 2cents.

First of all, varying the speed of a single phase compressor will probably produce undesirable results. The COP's will be low and you would probably have overheating issues.

I would use a 3-phase compressor that is rated for inverter duty. Then you can control it with a VFD that converts single phase to three phase. The VFD will allow you to set an enhanced torque curve for compressor startup.

4 comp, 4 VFD's. Is your wallet getting lighter?

As far as evaporator control, my pick would be an EXV. It will maintain better control and it will likely be necessary for oil recovery. For an oil recovery, the compressor would ramp up to full speed and the EXP open the whole way. This would be needed periodically, based on compressor run time. The velocity of the refrigerant moving through the pipes is what brings back the oil. The further the compressor ramps down, the lower the velocity.

Since you can't control the BTU absorption rate of the evaporator, independent of the comp speed, you will need to control your head pressure in the condenser. This means you will have to control the flow of the "hydronic system water" through the flat plates in accordance with the rest of the refrigeration system.

Sounds to me like you need a PLC and a good programmer to make everything talk.

Harvey

don_9

I agree with harvey.getting the oil back to the compressor is key with your design.however as harvey mention a good plc n program you can time it to go to high speed after so many hours of low speed run time much like they do with inverter system.The only other issue would be a low pressure bypass timer to hold off the pressure switch until your saturation temps get above the pressure switch cutout setting.then there the issue of liquid getting back to comp as well so a accumulator may be in order.At this time however i can not add anything else to the conversation bc i have had too many shot of fireball.so with that said i will revisit this thread at a later date.Shoot teach what ever happen to keeping it simple?

icesailor

The concept will work well.

IMO, the plywood storage tank is the weak link and will break due to the expanding ice. Water expands above and below 39 degrees F. Even if all the coils are at the bottom, the cold water will rise to the top after the whole tank, open or not, will expand to a rate enough to push the sides out. No matter how much you brace the sides. In Colorado, go find a lake that freezes solid in the winter. Notice the shore. It will be very flat and shallow along the edges and clean sand. Then, it drops off like the edge of a shelf. You can see the edge of the ice where it stops. Where the vegetation stops and starts.

If you really want to experiment, consider a Polyethylene septic tank or something made of Polyethylene. Fill it only half full. You need to give the ice a place to expand to. Think of an ice cube tray. The ice is forced or allowed to expand UP because OUT is more difficult.

If ice is powerful enough to crush a trapped steel ship, it won't have any problem blowing out a wooden or steel box. Like all expansion joints. we give the expansion OUR choice of where to go.

Bury multiple tanks in the ground. Use the ground as a heat transfer medium. Sooner or later, the ground will be too frozen to carry any cold heat away. Like a buried LP gas storage tank where the draw is too high. The whole tank frosts up and the ground around it freezes. Really cool when it happens.

If you used half buried tanks, you could duct warmer outside air over the ice in the expansion spaces to collect gained heat from the atmosphere.

Or so it seems to me.

Mark Eatherton

Harvey Ramer said:

Mark, setting up these compressors up in parallel with variable speed operation might be a bit more costly than a buffer tank.

Nontheless, I will donate my 2cents.

First of all, varying the speed of a single phase compressor will probably produce undesirable results. The COP's will be low and you would probably have overheating issues.

I would use a 3-phase compressor that is rated for inverter duty. Then you can control it with a VFD that converts single phase to three phase. The VFD will allow you to set an enhanced torque curve for compressor startup.

4 comp, 4 VFD's. Is your wallet getting lighter?

As far as evaporator control, my pick would be an EXV. It will maintain better control and it will likely be necessary for oil recovery. For an oil recovery, the compressor would ramp up to full speed and the EXP open the whole way. This would be needed periodically, based on compressor run time. The velocity of the refrigerant moving through the pipes is what brings back the oil. The further the compressor ramps down, the lower the velocity.

Since you can't control the BTU absorption rate of the evaporator, independent of the comp speed, you will need to control your head pressure in the condenser. This means you will have to control the flow of the "hydronic system water" through the flat plates in accordance with the rest of the refrigeration system.

Sounds to me like you need a PLC and a good programmer to make everything talk.

Harvey

Harvey, Thanks for chiming in. Per the above, no my wallet is not THAT strong. I was checking out the prices of VFD's etc. I am also limited to 220 V single phase, so a 220 V single phase to a 208 3 phase inverter is possible, but probably more $ than I 'd planned on spending.

By going with 5 compressors, I am creating a 5:1 turn down, which by rights should cut way back on the short cycle conditions for a 5 ton unit, with or without buffer, no?

In an ideal world, all 5 compressors would be variable speed, but that is an economically unviable option at this point in time.

If I can find a manufacturer who makes a 230 V 1 PH variable speed compressor (still looking) I may consider purchase, or trade out for collected data from the field….

As for the PLC, already have one with plenty of useable I-O. It is the ENV control from Climate Automation Systems that I spoke to you about last week. If you can imagine it, it can be done… They (Kele) make a really nice controllable 120 V single phase speed controller that accepts a 0 to 10 VDC signal to vary the output, and the minimum speed is programmable. If need be, I could monitor any and all parameters available to ensure proper and safe operation (suction and discharge temperatures and pressures).

If I MUST control the flow on the hydronic side, I can do that with the same controller. I am waiting for a major pump manufacturer to come out with a DCECM circulator with 0 to 10 VDC input for control. Worst case scenario, it will cost me the cost of a conventional circulator and the available AC wave chopper controller, triac controller (PWM). I was just running on the premise that if I keep circulators running at maximum speed, drawing off maximum BTU's, that it would be better for the refrigerant side of the system. In the field, the only time I see a high head pressure issue is when I've lost a pump, or have flow issues on the hydronic side of the system. I'd be using DCECM circulator, just because it makes sense form a wire to water efficiency, but hadn't planned on controlling the pumps speed.

As for TXV's, I see Sporlan has an Electronic Expansion Valve (EEV) that could possibly be incorporated into the design in numerous ways and numerous places to ensure proper refrigerant and oil flow, but are limited to R-22 considerations. May have to do more research on that. I can also place the physical location of my FPHXers such that gravity helps get the oil back to the compressors. This is a long term project, so who knows what cool devices may come out between now and actual construction time :-)

The buffer tank is a matter of space. I do have some adjoining space to the mechanical room that is unheated, and I already have antifreeze in ALL of my space heating system, but then I'd have to move lawn chair storage etc. to another location, which is technically doable, but I'd rather hold off on that for an out if I can't come up with a decent mechanical configuration.

Thanks for the input Harv. Much appreciated.

ME

Mark Eatherton

don said:

I agree with harvey.getting the oil back to the compressor is key with your design.however as harvey mention a good plc n program you can time it to go to high speed after so many hours of low speed run time much like they do with inverter system.The only other issue would be a low pressure bypass timer to hold off the pressure switch until your saturation temps get above the pressure switch cutout setting.then there the issue of liquid getting back to comp as well so a accumulator may be in order.At this time however i can not add anything else to the conversation bc i have had too many shot of fireball.so with that said i will revisit this thread at a later date.Shoot teach what ever happen to keeping it simple?

Thanks Don. My refer mechanic did mention the need for an accumulator to avoid possibly slugging the compressors. I will defer to his expertise in that area. I'm just there to hand him tools and cut pipe…


Watch out for them Fireballs. They come out of no where and pack a mean holiday punch! Imbibe responsibly, and never TUI (type under the influence) :-) Thanks for the input and feedback. Much appreciated.

ME

Mark Eatherton

icesailor said:

The concept will work well.

IMO, the plywood storage tank is the weak link and will break due to the expanding ice. Water expands above and below 39 degrees F. Even if all the coils are at the bottom, the cold water will rise to the top after the whole tank, open or not, will expand to a rate enough to push the sides out. No matter how much you brace the sides. In Colorado, go find a lake that freezes solid in the winter. Notice the shore. It will be very flat and shallow along the edges and clean sand. Then, it drops off like the edge of a shelf. You can see the edge of the ice where it stops. Where the vegetation stops and starts.

If you really want to experiment, consider a Polyethylene septic tank or something made of Polyethylene. Fill it only half full. You need to give the ice a place to expand to. Think of an ice cube tray. The ice is forced or allowed to expand UP because OUT is more difficult.

If ice is powerful enough to crush a trapped steel ship, it won't have any problem blowing out a wooden or steel box. Like all expansion joints. we give the expansion OUR choice of where to go.

Bury multiple tanks in the ground. Use the ground as a heat transfer medium. Sooner or later, the ground will be too frozen to carry any cold heat away. Like a buried LP gas storage tank where the draw is too high. The whole tank frosts up and the ground around it freezes. Really cool when it happens.

If you used half buried tanks, you could duct warmer outside air over the ice in the expansion spaces to collect gained heat from the atmosphere.

Or so it seems to me.

Ice, ice ice… Like you, I have 38 years worth of experience in dealing with ice and its amazing power. And although I have thought this process out as completely as I can, mother nature has a way of humbling us and keeping us in our places. My plan for ice expansion control are as follows.

Expansion tube: I plan on putting an expansion tube connection off of the bottom of the tank using a bulkhead fitting. I am thinking of going with a tube from the bottom of the tank, with a significant increase in pipe size (like up to 4" if necessary) with heat trace tape to keep it from freezing for its entire length, terminated at the tank top level. Theoretically, this should handle the liquid expansion associated with the liquid expansion/contraction rates of the fluid (water).

Coil placement: I figure that my input coil near the bottom of the tank will give me a "point of last solidification" (bonus, new hydronic term!).

Trapezoidal cube design: The "cube" if you will, will be shaped like a trapezoid (3' X 7' on bottom, 4' x 8' on top, 4 feet high)
As the water solidifies, it will push the block upwards (or so I hope). All coils and expansion pipes will be connected over the sides and top of the tank, and the styrofoam lid (vermin/dirt control more than anything) will be free to move up and down with the expanding /contracting ice.

External tank Reenforcement will be 2" channel iron at 12" on center from the bottom to the top of the tank.

Liner for tank will be a smooth polyethylene liner which is conducive for creating a "glide" condition and avoid wear and TEAR on the tank liner.

Digging the tank into the ground would be ideal, except in my case I am sitting directly of decomposing shale, which is not conducive to hand digging, so the tank will stay on top of the ground. The company that is currently doing these uses a 2,500 gallon (or numerous depending upon connected loads) polyethylene cistern tank. They are using PE immersed heat exchangers, and their system is working quite well. Not sure how many freeze thaw cycles the tanks have been exposed to so far, but to date they have no documented tank failures. Check them out at Thermal Battery Systems, Inc out of Montana and South Dakota.

I actually have a lake, but it is about 500 ' , two right of ways and numerous private property owners away and its level fluctuates significantly throughout the season. If I lived on shore front property, I'd definitely look to the lake as a potential heat source, but the environmental studies and permits required before I'd be allowed to tap this resource are probably more than I care to pur$ue.

Long term, I have spoken to a local concrete septic tank manufacturer about the possibility of incorporating the input and extraction coils directly into the reenforced concrete tank walls. Again, the forces exerted on the walls of these tanks, even with steel reenforcement inside is a consideration. I am certain that these folks (for a price) could also configure the tank such that expansion would/could be pushed upward (trapezoidal shape), but having it buried in the ground doing push ups could be problematic for the finished landscaping above the tank. Would possibly need to incorporate "head space" in the tank to keep the ice confined and keep it from surfacing.

I do know that there are some fine examples of ice storage tanks used to defer peak load cooling considerations, but have not had a lot of free time to do the research necessary to seek that information out, yet.

I am enamored with this technology. Not so much with the production of ice, but more for the potential of being able to use any energy source with a temperature greater than the storage tank as a potential heat source. Attic heat, snowmelt slabs, ambient air, solar thermal (at 80% efficiency ALL the time) and other sources make this a very viable alternative.

The most expensive portion of GSHP is the loop field (vertical or horizontal). I think this system address that and many other heating related issues. Nothing will ever replace Mother Nature for an excellent heat sink during cooling, but not all of AMerica is built in cooling dominated climates. Even in low cooling load climates, a snowmelt slab could be the means of heat rejection at night, allowing this system to do some cooling.

Thanks for the thoughtful input. Much appreciated.

ME

icesailor

ME,

I'll try to remember all you have said.

My example of the lake and the ice wasn't for you to use the lake as a heat source, but to demonstrate the power of expanding ice.

Ice is one of the "stickiest" substances around. Take a brand new car, heavily waxed and leave it out in freezing rain. you still have to scrape the ice off and worry about scratching the paint. Polyethylene is extremely slippery. Ice still sticks to it. A trapezoid shape with the pointy end down is a good choice because the sideways forces are always let UP. If someone makes such a tank, it would be a good choice. They make Poly tanks that are round and may be cheaper. That is the only reason I mentioned the 1/2 filled round tanks. And it is easier to keep vermin out of a round tank than a big flat topped tank.

I think your ideas are very interesting and can work. Consider putting some solar panels in and a HX coil into the tank. Gather energy from the Sun and store it in the tank. Whatever heat captured will be a gain to the system. I don't follow such things but it seems that so much of solar panel energy is going to electric. It would be easy to put electric elements in the water and store heat that way. But you would know if it goes one way or another. With electric, there's no moving parts, no pumps to buy or fail. No antifreeze to deal with. Apollo 13 had a LOX tank with an emersion heater with Teflon insulation on the wires. A thermostat failed and overheated the wire, causing an explosion. You won't be dealing with that.

If it is windy, a DC motor, run faster than it is designed to run under power, becomes a generator. More available heat. You can have lots of fun.

You can have anything you want. Just not everything.

OBTW, the warmest ice will be at the bottom of the tank. Its that "Lake Effect" Thingy.

SWEI

Single phase VFDs are somewhat rare, relatively expensive, and come with significant limitations. Small 3Φ VFDs are inexpensive, reliable and incredibly flexible. At 3 HP or smaller, most will accept single phase 208/240V input power, and the smallest (usually 1/2 HP or less) can even be had with 120V inputs. Take your time finding the smallest 3Φ compressor you can and design around that. Most drives can be linked using EIA 485, typically via Modbus -- so if your control system can speak that you can do much more than you can using multiple 0-10V output ports. Many include onboard PID controllers, so you can push more of the logic down into the equipment rather than relying on the control system to do everything.

If I knew more about refrigerants, I'd almost certainly be designing my own as well.

Gordy

Mark my 2 cents is literally face value in this crowd, but about the storage tank. I dont know if the size is a real estate constraint, or design critera. Think above ground pool. I know mine has froze to the bottom and never a problem with it from expansion.

You could probably find someone who just wants it out of their real estate for free. Smallest I believe is 12' dia. Custom liner throw the winter cover on it to keep the dirt, and vermin out.

If the size is to big think circle in the design of your tank eliminate corners.

Anyway good to here your ideas your like EF Hutton when you post.

icesailor

Many pools use a vinyl liner. They don't break from expansion. The pool people usually lower the water level in the winter to get the water out of the pool piping that is underground.

Gunite pools can crack from frost/ice expansion. Think of "Frost Heaves". Gunite pools that crack can not be repaired.

But all pools, Gunite or a liner will achieve an equilibrium with the ground frost because the frost in the ground will counteract the frozen and expanding water/ice in the pool.

Gordy

icesailor said:

Many pools use a vinyl liner. They don't break from expansion. The pool people usually lower the water level in the winter to get the water out of the pool piping that is underground.

Gunite pools can crack from frost/ice expansion. Think of "Frost Heaves". Gunite pools that crack can not be repaired.

But all pools, Gunite or a liner will achieve an equilibrium with the ground frost because the frost in the ground will counteract the frozen and expanding water/ice in the pool.

Yeah you want to drop the water level about 1' below the top rail or the ice will heave it up.

Initially mine had a tupper wear lid to go over the skimmer, and a water,tight cap over the jet. This saved on water when refilling in the spring. But my buddy actually had a leak in his winter cover which allowed the water level to increase, and destroyed his pool due to heaving the top rail off without him knowing it, and collapsing the pool in the spring when it thawed. After that I just left those covers off let's excess water out.

icesailor

Pools are like boats. A hole in the water in which to pour money in.

Like horses. It costs a lot to feed and care for them. And all you get in return is an uncomfortable ride and a lot of ****.

Cheaper than a divorce though.

Gordy

The puppy

tim smith

I did a hvac system for a party cabana next to the swimming pool. They would have large party gatherings in the cabana and needed cooling. I used water source heat pumps and used the swimming pool as my condenser. It was a great heat sink.

Techman

Sorry for my late input. Check into how Supermarkets control their "compressor rack systems", that is multiple compressor racks. Danfoss had a Supermarket Refrigeration Class( Scroll comps ) where they discussed using just the 1st comp w/VFD, once the load is to much for the VFD comp, then #2 comp cuts in at full speed ( not VFD), and #1 comp ramps down. The #1 comp, the VFD comp, can ramp up above its rated HP as well as below its rated HP.. Flapping jaws here, a 3t comp can ramp down to 1 1/2t (45hz)and ramp up to 4 t (70hz) And as usual, oil return was a concern when the VFD was in its lower RPM ranges.

unclejohn

My 2 cents is why even bother ramping down a 1 ton compressor? Just stage them on and off.

Mark Eatherton

"My 2 cents is why even bother ramping down a 1 ton compressor? Just stage them on and off."

That's what I will probably do to start. Can always add to the logic later if short cycling becomes an issue.

Thanks for all the input.

ME

SWEI

A VFD brings more than just "ramping" to the table. The refrigerant cycle math is a bit above my pay grade at this point, but the performance improvement with cold sources is considerable.