Kentucky Cold water input will I freeze?

Discussion in 'Maintenance and Troubleshooting' started by ReallyOddJobs, Nov 16, 2016.

  1. ReallyOddJobs

    ReallyOddJobs New Member

    Hey all, I got a couple of questions about geothermal and I hope someone here can answer them.
    I work at a water treatment facility in Lexington Kentucky and we have an output line that gives 17.5GPM of water on demand but it comes out of one of the cooling tanks from our processors so it is a constant 44 degrees. We have a supervisor office on-site and one of my guys got ahold of a 5ton Bryant Evolution Geothermal unit that we want to hook up to heat and to cool the office. We have drainage already so pump and dump should work. I see I need 3 Gallons Per Minute per ton so I got enough water for shure but is that temperature too low? Groundwater is in the 50's so don't know if 44 is going to cut it. Will it either not give me heat or cause a freeze up. How much temperature can a unit -pull out- of water per minute as a maximum? If this works Anyone here in the Lexington area want to hook it up? It's already in the closet. I got power and ducts done. I got access to more pipes, valves and actuators than you can shake a stick at so I'm looking for someone to rig up a thermostat and get the unit set up for best heating and cooling options. I looked through the maunal and it looks like you need a special degree just to read half of it!
  2. mtrentw

    mtrentw Active Member Industry Professional Forum Leader

    44 degree water should cut it. With open loop applications and water at 50 or better, general requirement is 1.5 GPM per ton. The 3 GPM figure is more for closed loop. My closed loop 3 ton does fine at 30 degrees enetering water temp (EWT). Just needs plenty of antifreeze.

    With 44 degree water, you'd want to target 3 GPM per ton. With 44 degree EWT, you may get 36-38 degree LWT. You need to ensure your LWT stays a few degrees above freezing.

    I'd expect one of the experts on here will give you an idea of delta T based on the unit you mention and flow rates which will be much more precise than my guesswork.
  3. ReallyOddJobs

    ReallyOddJobs New Member

    I figured I was right but I always like hearing it from someone else for sure.
    My pea brain will never figure how you all heat a room with near freezing or below freezing water but I guess that's why I don't make the big money.
  4. ChrisJ

    ChrisJ Active Member Forum Leader

    The magic of refrigeration! There is 5 more degrees of heat in 30 degree water then 25 degree water.
  5. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    I agree with the above math that increasing GPM per ton results in reduced waterside delta-T, and that is helpful in the case of 44*F EWT.

    3 GPM / ton may not alone guarantee success. I advise carefully and accurately monitoring refrigerant pressure in the watercoil while in heating mode. Even if LWT is 36*F or higher, refrigerant could be boiling at a temperature 5-10 degrees below 36*F, and that could result in spot icing within the refrigerant / water coil. Icing both insulates the gaps between water and refrigerant tubing and reduces water flow. Both effects lead to more icing, and the condition rapidly snowballs.
  6. mrrxtech

    mrrxtech Member

    Todays Geothermal Units have a Low Temperature Trip for loop water outlet of 30 degrees for an open loop system.

    If the loop is a closed loop with freeze protection, there's a wire jumper on the card that you can cut to open which changes the Trip to 20 degrees.

    My Trane 4 Ton Unit drops loop temperature by 9 degrees when in the heating mode. With a 5 Ton Unit Loop water temperature should see a larger drop but stay above the 30 degree trip.
  7. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    I can't speak for all systems but on well designed ones you'll find that the freeze protection sensor is actually on the refrigerant line at the water coil rather than the leaving water line.

    It's a subtle difference, and one might ask why infer water temperature from refrigerant temperature? Consider that if water flow stops entirely (say from loss of pump or user error with regard to a water valve...) In the case of a no-flow situation a leaving water temperature sensor will be influenced by ambient conditions - the air temperature in the compartment which it shares with a warm compressor. Icing within the water coil would not be "seen" by a leaving water sensor.

    So what does that mean to ROJ? The possibility of low water temperature trips even with leaving water in the mid 30s. As to the actual probability of that occurring I will defer to my professional colleagues to the north...we don't see much in the way of 44*F water down here in Florida.

    Going a bit further out on a limb - If necessary (nuisance trips at 29.9*F with 36*F leaving water) it may be possible to tweak the refrigerant charge and / or metering device to elevate saturation condition within the refrigerant side of the water coil.

    It would be unfortunate to go through the trouble of installing this thing and have it trip out. The temptation at that point might be to switch to the 15*F freeze protection point intended for closed loop systems protected by anti-freeze...don't do that.

    As for cooling, if I had a steady year-round supply of 44*F water I'd look at a fan coil - simple water to air heat exchanger. Google "Multiaqua" - they are made in South Carolina, and in my neck of the woods distributed by Trane commercial. Small, light cheap, and simple.

    MRT - Please continue to learn, but hold off on teaching at least until you've read the entire chapter.
  8. urthbuoy

    urthbuoy Well-Known Member Industry Professional Forum Leader

    Done right, and you're fine. On suspect system like this, I also put in an aquastat that intercepts the compressor call at a certain low temp EWT.

    We've done a fish hatchery that is similar to your demands. The difficulty is actually diverting only what you need with minimal equipment. If you can do it just with a T and proper pipe size, you're golden.

    Also, I've frozen a coaxial in about 20s with airlock and unmixed antifreeze helping to contribute. Doesn't take too long at all. No damage though - thankfully.
  9. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    That is not the case. Usually a larger unit has a larger flow rate, meaning that the delta T stays the same. Only if the larger unit has the same total flow rate as the smaller unit will the delta T go up.
    ROJ, in water without antifreeze, we usually start seeing ice formation within the coax coil around 38F discharge temp, since the metal surface area of the coil is much colder.

    Keep you flow rate at 17.5 gpm to keep your discharge temp above 38F and you should be good to go.
  10. Eric Fletcher

    Eric Fletcher New Member

    I don't have the technical experience of some other posters in this thread, but I do have real-world experience with heat pumps and extreme cold.

    When our installer recommended a pond heat sink for our GHP in 2006, I also wondered about freezing. After all, I live in Quebec where -30C/-22F is fairly common during winter, and our pond is usually covered by 2-3ft of ice. However, owners of other units he'd installed in the area seemed satisfied, so we agreed. In the first year, our energy use was cut almost in half.

    The system worked well for 6 years, with lots of heat during winter and almost free A/C from the quite cool pond water during our hot summers. During particularly cold parts of winter, we noticed that occasionally the unit would revert to electric resistance backup. When I asked the installer's service guy about it, he told me this was a normal automatic fail-safe for when the closed loop temperature dropped too low.

    But in 2012, a severe drought affected our pond's input source. After seeing our electricity consumption skyrocket in January 2013, I broke a hole in the ice: the two 4x8 heat sink panels were encased in ice! The normal steady flow of water into the pond had completely stopped, so there was no longer sufficient flow to melt ice that would form around the heat sink plates. Despite that, the system was still able to extract some heat — but it was very inefficient, and too readily shut down the loop to revert to electric resistance.

    The original installer gave us a very good deal on drilling two 380 ft wells (in solid granite) to replace the pond system, and we have had no problem since. Unfortunately, our installer was not so lucky: drilling our new wells and changing our connection was the last job he did before declaring bankruptcy.

    As an aside, if anyone could use two Slim Jim SJ-05T Lake Plate heat exchangers in excellent condition, let me know! ;)
  11. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    The fact that your system uses electric supplement heat during cold spans does not automatically indicate an issue, but can be a common design. But it would be unusual to shut of the geo system completely altogether. Normally, there is a large amount of heat below the frozen ice, but it requires circulating water via convection, which is usually the case even in ponds without a steady input of water.
    While a spread out pipe on the bottom of the pond is usually very efficient, the surface areas of the Slim Jim is usually not large enough for the rated capacity unless you ensure even more water circulation around the pipes, in your case probably via the water input. However, if you take that water flow away, ice might build up due to much colder pipe temperatures temperatures.

    The fact that the panels were encased in ice means that there was not enough convection and the surface area was too small and not spread out enough, which is a slim Jim issue, and not a pond loop issue in general.

    Here is a pond loop in Buffalo NY, a bit warmer climate than quebec, but irrelevant for a pond loop performance. 10 ton system. As the graph with the annual water temp indicates, entering water temp never drops below 40F during the entire heating season. The difference is a spread out heat exchanger in the pond.
    Last edited: Nov 25, 2016
  12. Eric Fletcher

    Eric Fletcher New Member

    Our unit is designed to shut off the pumps when the incoming fluid temperature in the closed loop reaches 18°F to avoid damage to the internals. The backup electric resistance component was added to keep the house warm in such cases (and to get immediate heat if we ever needed it).

    You are right about the heat exchanger freezing due to lack of circulation in the pond. In our case, the water was only ~7' deep, so the tops were only just below the ice. In retrospect, we could see that the pumps would normally start again after natural convection and some flow in the pond melted ice that was forming on the plates — but the reduced flow let it build up to a point that it was permanently encased. Water is densest at ~39F (probably the lowest temperature at the bottom of our pond), so removing 6F drops the plates to 33F. With no circulation, that quickly allows water to get colder, so ice forms pretty rapidly.

    That's an interesting pond loop setup in your link! That pond is much bigger than what we have (as is the building).

    We considered using tubing spread out on the bottom of the pond, but were concerned about muskrats chewing through it. This is a serious problem locally with field tile drainage: the pests get into a drain line and chew through it to get out — letting soil wash in and destroying that line. The Slim Jim units are compact, and work very well if they can be placed in deeper water where there is more capacity to prevent icing.
  13. mrrxtech

    mrrxtech Member

    My neighbor has a pond and is getting too old to continue stoking his wood & coal burner. I've been trying to talk him into doing a hobby Geothermal install for the cost of materials. I'll keep you in mind for the 2 heat exchangers, plus my brother may get tired of using pond water in his system. Maybe he will be looking for heat exchangers. I'll drop him an email to see what he thinks.

    Take a look at adding a 500 gallon propane tank and a furnace back up, or gas logs in the fireplace as I use in the winter. Electricity isn't getting cheaper for now.
    Last edited: Nov 25, 2016
  14. Eric Fletcher

    Eric Fletcher New Member

    Our heat pump works very well now with 750 ft of drilled ground loop replacing the pond exchangers. We also have a wood stove with practically unlimited supply of good hardwood, so I won't likely need a backup. As well, our electricity rates in Quebec are amongst the lowest anywhere. We use a lot less power now than we did 10 years ago, and typically use the wood stove as a pleasant backup rather than a necessity.
  15. mrrxtech

    mrrxtech Member

    This old back won't be cutting & hauling wood except as a charity for a neighbor who needs my help collecting his fire wood.

    I dropped my chainsaw on my knee and took out my Cordura Chaps, geans and long underwear before the chain stopped moving. I had a white scratch on my knee but I could feel the pain that might have been if I hadn't let go of the trigger a split second before the chainsaw dropped.

    I figure one chain saw accident could be a costly as a few winters of heating with electric and propane.
  16. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    If you have 18F antifreeze, and you have 39F water by definition in the pond, you are obviously not exchanging heat very well, presumably due to the small surface area of the highly conductive plates. It is not the first time I have seen or heard that of the slim Jim plates. We have done many ponds, mostly smaller and shallower ones too. They work great if you understand the physics behind it. The pond stratification is a great thing to get a good amount of connection going and to circulate the water, but the surface area is simply too small when it comes to the Slim Jim, at least the ones I have seen.
  17. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    From what I've gathered though not personally experienced, Slims work better in cooling-dominated climates where warmed water naturally rises up and away from the plates. I'm not sure how that would work in my AO where ponds, lakes, rivers hit 88-90*F without additional heat from a geo system.
  18. urthbuoy

    urthbuoy Well-Known Member Industry Professional Forum Leader

    Word on the street around here was the same - good at cooling, nah at heating. So I have never seen one in the wild.
  19. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    Should't the same be the case for heating, where cooled water below 39F raises up and away from the plates?
    In the case where we changed out a Slim Jim versus a conventional slinky pond loop, the 5 ton rated slim jim was roughly supporting a 2.5 ton load.
  20. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    I suspect that the density increase resulting from cooling water below 39*F is marginal (and constrained by the hard lower limit of 32*F) relative to operating in cooling mode.

    In addition, once icing begins on any part of a submerged heat exchanger trying to collect heat, it can't help but spread and thicken, further insulating the heat exchanger and worsening the effect.

Share This Page