Geothermal units locking out etc! Design troubleshoot?

Discussion in 'General Discussions' started by air2high, Nov 7, 2020.

  1. gsmith22

    gsmith22 Active Member Forum Leader

    I'm not sure what all the colored circles with numbers are (especially 1 through 4) and what has/hasn't been installed, but if it were me I would leave the loop piping in place and dig new trenches with new piping to fit in-between all the infrastructure you don't want to move. There is no requirement to run new pipe in the existing trench locations. Drain fluid (or don't not sure about any environmental concerns) and abandon piping in place. SShaw's point about flushing should not be ignored. If you can't purge air out of the loops with the big industrial pumps commonly used to do this, then your system won't work and won't be purged by the loop pumps. All of this is because the loop and header weren't designed properly (or at all which is probably more likely). They should be paying for all of this (new trench, new pipe, land disturbance, etc.) not you. I personally wouldn't want air source heat pumps in your climate.

    Edit: there are also slinky loops too which take up less area. no requirement to do out and back loops
  2. air2high

    air2high New Member

    Ok. Thanks. All the other dots etc are irrigation zones and sprinkler heads. I just drew on top of that irrigation map.
  3. gsmith22

    gsmith22 Active Member Forum Leader

    irrigation pipe is pretty easy to fix if that is the concern. I know because the guys doing the trenching for the header at my house hit three pipes. spliced in new sections to fix. I wasn't sure if the circles were big landscaping beds or some other non-movable infrastructure but it doesn't sound like it. It might benefit you to have someone (you hire) design a workable system and then provide that plan to them to build on their dime. This way you aren't getting saddled with some existing infrastructure that should otherwise be removed but will cost a lot of money to do so and so is being kept to minimize cost but is otherwise hampering the functionality of a properly functioning system. The loop pipes strike me as this. You don't want to remove them because of the mess, they don't want to move them because of the cost, but the existing piping can't be purged so it will never work right keeping them.
  4. air2high

    air2high New Member

    Just trying to think out of the box... is there any possibility of plumbing an external pipe/valve in the yard that can be used to flush the loops, thereby bypassing the length of header if they were to dig up and just replace the header pipe and possibly redo the loop connections? Or would that be an issue with pipe freezing etc.?
  5. SShaw

    SShaw Active Member Forum Leader

    If you had fittings and valves to isolate individual circuits, you could purge them one at a time. This is commonly done by DIY folks who don’t have access to a flush cart, by bringing all the loop pipes inside the home to a common header indoors. In principal, I suppose you could do something like this outside.

    There is another thing that looks odd to me about your loop, now that I see the drawing. Typically, loop circuits will originate from the same location, using a fabricated header with one input and one output for each loop. See photo of my headers.

    This header is small and designed to keep the flow in each loop the same. Based on your drawing, it looks like each loop is separated from the others along about a 125’ section of the 1 1/4” pipe. I believe this will cause unequal flow in each loop, with each farther loop seeing less flow. None of the pressure drop programs I’ve seen can deal with this. They all assume a header and equal flow in each loop.

    Might not be a big deal if you have enough pipe in the ground. Besides pressure drop, the other important aspect of a geothermal loop design is to match the loop size to the heating and cooling loads of the home. There are programs to do the analysis based on the location, soil type, and loop configuration. If yours is not a conventional design it would be hard to do that analysis.

    Attached Files:

  6. gsmith22

    gsmith22 Active Member Forum Leader

    freezing wouldn't be an issue because the loop piping needs antifreeze anyway to prevent it from freezing as the temp of the fluid drops below 32. 30 degrees is the common design point for loop fluid temp in winter/heating use so the loop fluid is typically water with probably something like 20% ethanol or methanol which will protect down to the teens. Maybe you would need more protection depending on exposure of pipes/valving. One of those antifreezes is easier to pump (can't remember which) but is also considered less environmentally friendly. NJ (in my instance) mandated the use of the harder to pump antifreeze because it is considered less impactful environmentally should a leak in a pipe occur. PG is also available but no reason to use for antifreeze because it really increases pumping power needed (wondering if they used that too to compound your header pipe issues).

    As SShaw noted, if you have to dig up the header to loop connection point anyway (for reconfiguring) maybe that is the point where you could install this valve system. But I think you would almost need like a chamber or something to get inside of because of all the pipes that should be coming together here and valves, etc ( I doubt you want this sitting on grade). I'm thinking like a precast concrete box used in utility work or maybe a septic tank type structure set below grade with a lid at the surface. You would need to get power to it for the flush cart but maybe not a big deal if relatively close to the house with extension cords. But, it doesn't look like you have a common point of connection for each loop if your drawing is correct - you would need the header like SShaw pictured in this box so you get access to each loop and valve it off from the others and the header for flushing. I think you may not be able to make any of these decisions until you get it dug up to see what you are working with. All the more reason to have your own consultant working on your behalf while this is going on.

    You are unlikely to find a program pre-setup to model the loop lengths if what is drawn is correct because no one with any competency in the geo field would build it that way (not that it is surprising at this point). Maybe if you had some kind of obstruction or limited area you would specify different lengths recognizing the unequal flows. But wide open yard with a blank slate - no, no one would do that. To get flow in each loop, must go back to first principles and solve simultaneous equations for flow and friction loss on each path. Flow will divide in parallel plumbed systems based on the resistance the flow sees (ie friction loss) along each path. so solving simultaneous equations to find both flow and friction loss at the same time is the only way to do it because friction loss is based on flow and flow is based on friction loss (having fun yet!). So you don't have 4 loops working simultaneously, you have like 3.3 (guestimating) because of the unequal flow. I think you need to know what you truly need in terms of loop before determining if what you have can be used (maybe you only need 2 so 3.3 works fine; maybe you need all 4 so no good). Which goes back to what I said in post 23. The itch to use what you have when it isn't ideal or something much worse than ideal. I would figure out what is ideal and then determine if what you have can be worked into the system without sacrificing function or ability to maintain.

    Edit: when I was doing parallel flow paths for college course work, it was all nomograph/chart type stuff too. it was real tedious so no one purposely chooses to use different length parallel flow paths if they don't have to just to make the math easy and the construction difficult to screw up (just order 8 loop pipes the same length). Maybe this is why they choose to use all 1.25" pipe everywhere. Search for "close header" and you will probably find many examples of what SShaw posted becuase that is the way to do it.
    Last edited: Nov 18, 2020
  7. air2high

    air2high New Member

    The well driller/installer is not entirely sure how it is all plumbed together. I am not confident that they know if from the trenches that the pipes all come together somehow at a common point where they enter/exit the header over a short distance or if the they are connected to the header at a point where the trench ends. Unfortunately I didn’t get a chance to take photos during the build when the entirety of the trenches were open. So... last I understand is that their plan is to come up with multiple scenarios for when they would dig up the header, then review what is all in the ground after digging it up, take it back to engineer to formulate a final plan, and then enact said plan.

    I guess another thought instead of plumbing together an access for flushing out near where the trenches are located is to instead possibly bring each trench (that likely has 2 loops plumbed in series) back to the house for a total of 4 separate 1.25” pipes where you could then create a manifold and short header in the basement, which then is plumbed in parallel to the 2 GSHPs like SShaw alluded to. Then you could flush each trench by itself from the basement via the manifold where you could valve off individual trenches? However, I’m not sure if this would work or how it would then change pressure drop and resistance due to there being even more length of 1.25” pipe to account for.

    As far as if the loop in the ground is sufficient for heat exchange - I believe the length of loop for a 7 ton closed horizontal system is appropriate (barring issues with soil conditions etc) from what I understand, although I’m not an engineer or geo expert by any means.
  8. gsmith22

    gsmith22 Active Member Forum Leader

    Its hard to rule of thumb your loop size only because you have two out/back loops in each trench and it isn't clear if they are connected in series with each other or are plumbed in parallel relative to the header. The temp of one loop influences the temp of the other loop because they are so close together. Assuming they are staked, the bottom loop is probably more effective than the top loop (or part of the overall loop if plumbed in series). If you leave the 1.25" loops in place, and bring them back to the house, then your loops in trench 4 are definitely going to be a lot longer than the loops in trench 1 so you will cause a non-uniform flow issue. Looplink RLC can model the stacked loop piping - its what I would look to for guidance.

    I don't know if this is how they did it, but that spaced trench layout would make better sense being a reverse return header like what is usually done with vertical loops spaced ~20ft apart. Regardless of what they did, that would be the configuration I would do with that trench layout rather than running the loops all back to your house. With a reverse-return header you would get equal flow in the loops. Without running any numbers, I'm guessing that dividing flow by 8 (all loops are parallel plumbed) won't get you much velocity in each loop making heat transfer poor (low renyolds #) so probably not optimal. If you connect the two loops in each trench in series, you get overall flow divided by 4 so velocity goes up in each loop for better heat transfer but I believe SShaw indicated purging air is an issue back in post #16 with this layout. If you can't purge the air, the loops won't work right. I would imagine the purging air from 8 parallel loops would be even harder than 4 parallel loops given the many flow paths. End of day, system has to work and while it will obviously be painful to excavate/abandon existing loop piping what is the point of keeping something that won't allow the system to function even if the header gets corrected.
  9. ChrisJ

    ChrisJ Active Member Forum Leader

    I would look in to this idea. Having a vertical pipe down to a shut off for each loop. I also think the 4 loops should be redone as 8 loops.

    It's done everywhere for municipal water supplies using a tool with a long reach to operate the shut offs. Of course that means 8 locations where a pipe would be flush or just under the surface of lawns and gardens.

    I'm not sure Reynolds number is all that important, I believe Docjenser has said that with the use of variable speed pumps, that slow the flow in the loop, still have good transfer.
    Last edited: Nov 19, 2020
  10. gsmith22

    gsmith22 Active Member Forum Leader

    you would need more than just ball valve shutoffs, you need hose connections to hook up the flush cart to for purging the air out of the specific loop. I think you would probably have to have a manhole to get into to be able to shut off valves and connect up flush cart hoses. I don't think you could bring all of this to the surface - maybe a box extending down 4' to loop piping at an absolute minimum.

    If hell bent on keeping the 1.25" loop piping, then still need to confirm or make the header reverse return and much larger. I hear you on the Reynolds number but frankly I wouldn't want to be the experiment that tests docjenser's theory as everything everywhere says get Reynolds number into turbulent range (by definition above 3000) to assure turbulent flow for proper mixing at design rating (ie max flow on coldest day of year). Without having run numbers, probably get bare minimum turbulent flow with 4 loops (21 overall gpm / 4 = 5.3gpm per loop) but not 8 loops (21 overall gpm / 8 = 2.6gpm per loop). Sure at lesser flows on non design days don't have turbulent flow but you also don't need all that loop at temps above design point. If you have less than turbulent flow at design point, then you surely have much less than turbulent flow at all other times.

    Edit: just remembered, the original post says the trenches were 8ft deep so presumably one set of loops laid at 8' and the other set at 4' in hole. That complicates having separate loops with their own valves and hose connections for flushing and access to them. How in the hell did they dig 8' trenches and keep the dirt stable so someone could get in there to install the pipe? There are more questions than answers unfortunately with this install
    Last edited: Nov 19, 2020
  11. ChrisJ

    ChrisJ Active Member Forum Leader

    Don't need to flush outside, shut all but 1 loop off, do 1 at a time from header pipe(larger) in basement.

    8 ft trenches without falling in depends on soil, worked for us the guy digging said it was like digging in cement....

    Attached Files:

    gsmith22 likes this.
  12. gsmith22

    gsmith22 Active Member Forum Leader

    that is a tremendous amount of soil to move. i can now see why no one wants to undig the loop pipes. pain all around
  13. air2high

    air2high New Member

    Another thought. Is it standard to have plumbed in the basement a valve at highest point to release air? Or perhaps to have an air separator device plumbed in at that highest point?

    And also, what about a device that burps or adds water into the system when it senses lower pressure, therefore keeping it adequately pressurized? Then if you see that water gauge running up you know that something is not right and have to investigate why it keeps needing water.

    Not sure if these are more commercial precaution measures or if residential systems have these things in place to help maintain/troubleshoot the setup.
  14. gsmith22

    gsmith22 Active Member Forum Leader

    You fundamentally have a pipe size problem - your header pipe diameter is too small and your loop pipe diameter is too big. This creates two issues - 1) air cannot be purged from the system during the initial install because flush cart pumps simply don't have enough flow at the head loss of your header/loop pumping arrangement and 2) the system doesn't work well (or at all) in normal operating flow also because of the same flow vs head loss of your header/loop pumping arrangement. I don't think any of those things you mention solve your problem - they are nice extras on a properly functioning system. Think of them as tinkering around the edges. For instance, my pressurized system has none of those things and doesn't need any of them to function well. Could they make it better? Maybe. Would I notice a difference in function? probably not.

    If you don't want to dig up your lawn, you could always do a vertical well system. Its footprint is a lot smaller (probably could put 4 wells in a 20'x20' square grid) and you only excavate from house to wells for header. Abandon existing loop pipe and header pipe in place and keep your yard mess to a minimum. Don't quote me on what you actually need I was just guestimating at how many wells to give you a sense of how small this could be. No exterior boxes, valves, etc. as was being discussed above. The existing plumbing is all wrong and keeping any aspect of it will likely cripple a workable system.
  15. air2high

    air2high New Member

    Thanks. I wasn’t asking to see if these measures would solve the overall problem. Just more as added measures when it is fixed that could be added to allow for better future troubleshooting etc. in addition to redoing the ground pipe.

Share This Page