Washington DIY on Olympic Peninsula

Discussion in 'General Discussions' started by jyoke, Jun 25, 2019.

  1. jyoke

    jyoke New Member

    Hello All,

    DIYer here. I've been working on the design of my heating system for the house I'm building and am getting ready to start purchasing and digging. I would appreciate it if anybody cared to give me some feedback on my plans.

    Broadly, it will be a water to water heat pump with horizontal trenches as source and radiant floors as load. It is a 2800 sq-ft log home with expansive windows on NE and NW sides. The house sits on top a ridge exposed to occasionally nasty winter wind. I followed the Manual J and came up with 38,500 Btuh heat loss at design temps. The trenches will be in "sandy-gravel" soil, well drained, but the location (Olympic peninsula, WA) gets a fair amount of rain. The load is heating dominant. Many people around here don't even have air conditioning, although I plan on putting in connections for chilled water to air handlers so if I want to install them later.

    I'm looking at getting a 5 ton HP. Either a Bard GW060-1S1A1RCCX or a ClimateMaster TBW060-AGD01B0BS or its equivalent Comfort-Air.

    I used the 09 IGSHPA Ground Source Heat Pump Design Guide to come up with my trench design. 4 trenches, 2 pipes/trench (1 loop/trench), 6' deep, 24" wide, 10' between trench centers. Trenches would be 265' long but if I add 10% for safety then aprx 290' long. Trench pipes (4 parallel branch loops) would be 3/4", with a 190' home run (x 2 pipes) of 1-1/2" header pipes between the trench field and the house.

    I expect I will go with a non-pressurized circulation system. Haven't selected the pump(s) yet. I calculate that I will have about 35' of head loss at 11.25 gpm.

    I would very much appreciate comments or suggestions of any kind.

    I have a few specific concerns:
    The IGSHPA books (I also have their soil book) present a very broad range of diffusivity and conductivity values for 'sandy-gravel' soil, depending on the moisture content. I really don't have a clue on this. We usually get plenty of winter precipitation, but the soil drains easily, and what if we have a dry winter? I used the values at the dry end of the range. If I use the values at the wet end, the trenches can be at least a 100' shorter. How to account for moisture variability when moisture affects the calculation?

    The 5 ton heat pumps I'm looking at are somewhat oversized. A 4 ton heat pump would be marginal and I don't see a simple way to provide supplemental heat to a radiant floor system. I don't mind paying extra for a bigger heat pump but the whole point is to have an efficient system. I'm hoping the 2 stage Bard will do this for me. Any thoughts? Are there any other 2 stage W-W HP's that sell direct to consumers?

    The Bard HP Performance Data "http://www.bardhvac.com/digcat/S3364_TechDoc_CD/TechDoc-PDF/S3465.pdf" has a column called 'Heat of Absorption' in their heating performance table, under 'Source'. Does anybody know what this is and how it relates arithmetically to the Heat Capacity, Power Input or COP? Note, this is not Heat of Extraction, as defined by the IGSHPA text, the values don't match what that would be. I contacted Bard and the technician that I talked to at first couldn't provide an explanation but then emailed me that Absorption and Extraction are the same thing. I may be an amateur but I can see that the math doesn't work for these values as Heat of Extraction. Can anybody add clarity to this?

    Anyway, sorry for the long post. I appreciate any comments at all.
  2. jyoke

    jyoke New Member

    Wishing to get some feedback still. I am a complete novice.

    To compound this, it looks like I will be the first in this county (Mason), to install a geo system. A couple of weeks ago I went to the county's building dept to check on any potentially required setbacks and they said STOP! They requested that I submit all my calculations and designs and get their approval before I proceed. I don't know if this is typical in other states/counties or not. In any case, I gathered everything I had into a presentable form and submitted it, but they admitted I was the first and that they had no familiarity or understanding about what I am proposing to do. I'm in a holding pattern now waiting for the county's review and decision.

    I sure would like some feedback from somebody with experience.

    Since my first post I have decided to increase the size of the 'home run' pipes to 2" dia. as that would reduce the head loss by more than 4'. I will be doing the pipe work by myself. Never worked with this type of pipe before. Hopefully going with 2" will not prove to be too much of a difficulty.
    I've uploaded my designs if anybody is interested.

    Attached Files:

  3. wing

    wing Member

    Welcome to the geo energy universe.

    Get a professional HVAC engineering consultant to run manual J and work out your radiant floor design. You mention cooling so you will need to provide for ductwork, the HVAC expert can design that as well. You definitely will want to execute ductwork during construction, not afterwards.

    That is unfortunate that your home's glazing is on the north side of the house. What are you planning to reduce radiation heat loss through these windows.

    For a 5T heat pump. you are going to need 15 gpm flow at the usual 3 gpm per ton guidelines.

    Good move on increasing your header pipe size to 2 inches, that will save quite a bit on power requirements to circulate your closed loop, particularly at 15 gpm. Laying 2 inch HDPE is a beast, takes three or four people.

    I would look at adding at least two more trenches for your 3/4 pipe, i.e six loops of 500 to 600 feet each. Your plan is for app. 2000 feet of 3/4 pipe or 400 feet per ton. That is marginal at best. The cost penalty for the additional two trenches is $600 for hdpe pipe and app. 4 hours digging time per extra trench, i.e very nominal compared to your total investment.

    Keep it simple and really no need for anything more than a single stage heat pump with buffer tank.

    What is your plan for connecting the outdoor manifold. I recommend you have a professional perform the fusion work.

    Radiant floor is a lot of time, effort and money. Are you going to embed your floor loops in gypcrete. Radiant floor also has impact on floor truss design, framing and finish flooring options. You might be shocked how much it will cost as a system, I certainly was.
  4. jyoke

    jyoke New Member

    >Wing, thanks for your comments. Here are my thoughts and some more info in response.

    I'm fairly confident of the heat load calculations. I bought the Manual J 8th edition book, studied it, set up a spreadsheet, and did the calculations. I also purchased a limited one-time use license for AccuLoad and used this commercial software to double check my results. They matched, so I'm at least as confident as I would be if I had hired some random engineer I didn't know.

    For possible future installation of AC I am going to install, before the walls are covered, insulated supply & return waterlines to 3 points in the house were I might in the future want to have AC air handlers. The house is an open loft to great room type of cabin, so there is a very large open space and then 2 enclosed bedrooms. I would put 1 larger air handler in the loft projecting forward into the great room area and a small air handler in each enclosed bedroom. Although I have not studied these I envisioned that I would be able to find air handler units for these locations that did not require ducting.

    I know what you mean about the windows, but the location of the home, its orientation, and placement of the windows were all driven by the great views. A significant motivating factor to undertake the challenge of installing a geothermal system was to compensate for the inherent thermal disadvantage of these windows. They are high performance double pane with a U-value of .26.

    One of the things that has confused me about the online discussions of heat pump size in rated tons vs number of trenches and pipe length is that the actual heat load of the house often seems to be missing from the discussion. The 060 series of heat pumps are nominally considered to be 5 ton units right? But at my design load conditions of 30°F EWT-source and 111°F LWT-load the Bard GW060 outputs about 45,000 btuh or 3.75 tons. So do I need to size the trenches and water flow for 5 tons or 3.75 tons? Furthermore, my actual heat load at design is 38,500 btuh or 3.21 tons- shouldn't that be what I'm really sizing for? The IGSHPA book's equations using the actual load and capacity values result in just under 2000' of 3/4" pipe utilizing 1 loop per 250' long trench x 4 trenches. The configuration of my property allows me to make the trenches as long as 300' but I would have to cut down some nice trees to put in more than 4 trenches at 10' O.C. So, what I'm inclined to do is stick with the 4 trenches but run them out to 300' for a safety margin. This would be 2400' pipe.

    A potential complication is that my property reputedly sits on hard-pan. Digging these trenches and laying the pipe is a big question mark for me. The topsoil is glacial till- sandy-gravelly with many rocks. A local septic designer told me that I would be digging into hard-pan, apparently formed by eons of glaciers sitting on top of acidic soil, causing it to harden (but not turn into rock). An excavation guy told me that it can be done with the right excavator. He recommended a John-Deere 135, which I plan on renting. I guess I will need to bring in sand too, to bed and protect the pipe before backfilling. I'm trying to figure out these details now. I was going to connect the pipes with socket fusion and rent the tools from the pipe manufacturer in Tacoma that I'm getting the pipe from. I'll get a few extra fittings and make some practice joints first.

    The house has 3 floors with radiant as follows: basement- tubes in concrete slab, main floor- tubes in gypcrete, loft- above floor tube & aluminum plate. The slab was already done by a GC before I took over the construction (I'm a marine engineer and I was out at sea initially). He made so many mistakes, I figure I can't do worse.

    Thanks so much for comments.
  5. wing

    wing Member

    The chances of this system working as designed is virtually zero. Sorry to be so blunt.

    The Bard GW060 data is based on ground loop flow rate of 15 gpm. If you try to run less than this you wont get the rated heat output and likely you'll be experiencing lock out faults during the winter.

    You state that the topsoil is sandy / gravely which is not a good scenario for heat transfer with a undersized ground loop compounding your woes. 2400 feet of pipe would be inadequate to marginal for a 5T even if you had water saturated muck to trench. Sandy / gravely - I would expect performance to be poor and you will likely be freezing up your heat pump during the winter.

    After all that time and money to install pipe then make your fusion connections a DIY project after a few practice attempts - seems like a really bad idea. If there is a leak, you have a huge mess to deal with and your house will be without heat until you dig everything up, try to figure out what is leaking (not easy) and effect repairs. As you generally hook up the manifold 'first in, first out' there is already a big tangle of lines in the proximity of the manifold. Maybe your wife is the forgiving type, at least I hope so.
  6. jyoke

    jyoke New Member

    Wing, thank you for your comments.

    In the first place, the Bard GW060 is rated for a 13 gpm flow, both source and load, not 15 gpm. Check the manual I linked to in my first post.

    Secondly, the IGSHPA calculation, using very conservative soil properties of .44 btu/ft hr °F conductivity and .42 btu/ft hr °F diffusivity gives me a total pipe length of less than 2000' across 4 each aprx 250' trenches. The calculation from the website LoopLink RLC also indicates 4 trenches, and pipe aprx the same length will work. You didn't address my earlier point about the actual heat load of the house being just 38,500 btuh. Regardless of the heat pump nominal rating, I don't need to take more heat out of the ground than I need to put into the house. I wish somebody would comment on my actual numbers instead of just reciting rules of thumb that they've read in a forum.

    Thirdly, you make connecting this pipe sound like open heart surgery. I remember a chief engineer dangling me by my heels down into a bilge while I twisted around and, using a mirror, did an overhead weld on the underside of a leaking 6" steel circulator pipe. This was about 40 years ago, on a king crab boat in the middle of the Bering Sea while it was blowing 65 knots and the boat rolling something fierce. We finished the trip and season with no further needed repairs to the pipe. By comparison, connecting plastic pipe, with precise modern tools and methods, on land, at my leisure, after a full night's sleep, after studying on Youtube, and practicing, doesn't sound too tough. Kids today, I tell ya...
  7. gsmith22

    gsmith22 Member

    If you are doing this yourself, and are willing to research and learn, the IGSHPA's design and installation manual will be invaluable for you - the first book #21025 at the following link:


    You mention a IGSHPA book in your first post but I'm not sure if that is the same as the one I linked. Mine is copyrighted 2009 with 2nd printing in August 2011, ISBN is 978-0-929974-07-1.

    Seems like you have the house heating/cooling load via Manual J under control. I also like this website for Manual J calcs: https://www.coolcalc.com/ Its approved by ACCA and is free to use. You only have to pay if you want a printout to submit for a permit and even then its $15 for 5 printouts. I found it very easy to use, can do both block load and room-by-room with as much detail as you would like to include.

    There is a ton of soil information online here: https://www.nrcs.usda.gov/wps/portal/nrcs/site/soils/home/
    You might be able to use one of the historical soil surveys or the web based application to narrow down the type of soil in your area. It might help you with interpreting the soil information in the IGSHPA soil and rock classifcaiton field manual (assuming this is the one you have). Otherwise, your decision to use the worst thermal/driest soil value will only cause your calculations to spit out a longer loop length than you probably need. I'm on the east coast and generally we don't have the west coast's dry summers/wet winters - rain is even distributed year round. Wet soil (and rock) is usually your best friend due to both its movement through the soil as well as water's high specific heat capacity. If you do get a dry winter, could you run a hose over the loop area to keep it wet? (have no idea how realistic this is) If it was me, and the possibility of a dry period is very real, then I would use the dry value to design and hope to never get it and have the system just run better but at least you are covered.

    I have forced air, so I haven't used water-to-water heat pumps, but there must be some way to provide supplemental heat to it - maybe electric resistance heat for the load side water similar to an electric hot water heater? I'm just not familiar so I don't know the answer. THe IGSPHA manual does run through water to air and water to water examples so I am guessing the info is in there.

    "Heat of Absorption" is only on the heating tables under the source side - it has to be the same as "Heat of Extraction". You will note that on the cooling tables, the names changes to "Heat of Rejection". It is implying the direction that heat is going. I don't see how Heat of Absorption can be anything other than Heat of Extraction. Maybe the "math" is different for a water-to-water unit and you are calculating heat of rejection using a water-to-air unit calculation (rather than water-to-water)? Again, I'm less familiar with the water-to-water units and their calcs.

    With regards to heat pump capacity/naming, they aren't like an A/C or furnace unit because their source of heat (ground via the loop) has a variable temperature. A/C uses electricity and furnaces use gas/propane for input energy that have known/non variable energy source values. For a water-to-water unit, its output capacity will depend on the source (ground loop) entering water temp and flow rate as well as the load (hydronic piping) entering water temp and flow rate. Of these, the source entering water temp will vary the most over any season. For heating, its highest temp occurs in Sept/Oct and steadily drops through the winter as more and more heat gets extracted. But, at lower entering water temp, less heat is available to be extracted so the unit's capacity drops too. So it wouldn't be uncommon for a 4 ton unit to have less than that heating capacity at low entering water temps that will occur at the height of the heating season. As a general rule, you will get a heating capacity less than the unit's nominal rating and a cooling capacity at or greater than the unit's nominal rating.
  8. jyoke

    jyoke New Member

    An update...
    I got the county's go-ahead on my plans. Pipe is on order (enough 3/4" pipe for 4 each 600' loops). Should begin digging late next week.

    I have been worrying about the logistics of bringing in sand for bedding the pipe. My soil is glacial till including silt, sand, gravel and rock. There is plenty of rock to damage the pipe by contact in the ground or impact during backfilling. I am willing to bed with sand but I see a logistical nightmare getting sand from where the dump truck will have to deposit it to the trenches and evenly distributed. I have been thinking about constructing 'A' frame structures covered with screens over the trenches and then just using the excavator and dumping the backfill over these screens until there is enough fine dirt under and around the pipes to protect them. I would make 40' or so worth of structures and just use the excavator to shift them down the trench as I backfilled. Before I put in the rest of the backfill on top I would wet this fine soil with water to make sure it filled in around the pipes. I haven't read of anybody doing anything like this. Is this a crazy idea?

    I've spent my life out on the ocean and don't have enough experience with dirt to make a guess as to how this might work. Would love to hear if anybody has any thoughts.
  9. ChrisJ

    ChrisJ Active Member Forum Leader

    I have 5 @ 200' trenches with 1" pipe. The guy doing the digging would sort of sweep the rocks away with the bucket, then scoop dirt with little to no rocks, dumping the dirt in the trench as close to the bottom so dirt & rock never fell very far.

    Covered the pipe with about 18" of dirt, then we layed PCV water pipes with holes drilled every 2'. Brought each trenches pipe to a manifold then one pipe to the house with a hose connection. I can introduce water to the bottom of the trenches if I feel the soil is getting dry.


    Attached Files:

  10. jyoke

    jyoke New Member

    Thank you gsmith for your comments. The IGSHPA manual you mention is the one that I have and on which I based my trench calculations. I thought it not terribly well organized nor its formulas well supplemented with examples (compared to the terrific 'Modern Hydronic Heating', which I used for my hydronic plans), but it did give me enough info to give me confidence (perhaps misplaced?) to do the design myself.

    I checked your link to Coolcalc. Cool indeed. It looks great but I don't need it at this point. If I have a chance, maybe I'll play with it later to double-check what I've done on my own.

    Thanks for the link to the nrcs website. Lots of data. Very interesting. So far what I found more or less corroborates the assessment I already made about the type of soil I have, 'Hoodsport gravelly sandy loam'. So far I haven't found the thermal specs for this specific type of soil, so I'm proceeding with my conservative guesstimate of .44 btu/ft hr °F conductivity and .42 btu/ft hr °F diffusivity.

    In all the stuff I've read on hydronics, I haven't run across anything comparable to the quick and dirty supplement that a water to air HP has in the form of electric heating strips. The cabin will have emergency backup heat in the form of a propane stove as well as a wood-burning fireplace. If it really turns out that I'm under sizing my system and that I will often need supplemental heat I will have to rely on these backup sources until I can make some fundamental changes to the geo/hydronic system.

    The Bard technical specs really don't make sense. Heat of Extraction is defined in the IGSHPA texts and other literature that I’ve found as:
    HE = HC x ((COP-1)/COP)
    So, in a typical example from the Bard Performance data for this 5 ton heat pump, with
    Source EWT = 30°F, GPM = 13, and Load EWT/LWT = 90/97°F,
    the Heat of Extraction would be:
    HE = 45,150 BtuH x (3.66 – 1)/3.66 = 32,814 BtuH

    Yet the table lists the corresponding Heat of Absorption for these values as 44.09 MbtuH. More than 11 MbtuH higher than the calculated value for Heat of Extraction. Something's wrong.

    If you look at the data for similar water to water heat pumps from ClimateMaster, WaterFurnace, etc, their values for Heat of Extraction work out mathematically. Bard's value for Heat of Absorption does not, at least as a value for HE. Nor does Bard list correction factors for antifreeze in the same way as other manufacturers, especially, the lack of a CF for power. Actually, with Bard's HP data, you can't even perform the IGSHPA HP calculation worksheet. To do it, I used CFs from ClimateMaster HP data. Why doesn't Bard follow the standard way shown in the IGSHPA manual? Frustrating to me- I enjoy the calculation side of this project and look forward to pasting temperature, pressure, flow and power sensors throughout the installed system to compare performance to my calculated expectations. Bard, are you listening?
    I would switch to another manufacturer but for the fact that I haven't found another 2-stage in my size range that is available direct to consumer.

    With regards to the heat pump capacity rating, I got that their actual output depends on source/load entering/leaving water temperatures and flows. I was just pointing out that logically, the GHEX size must be proportional to the heat load. The x# trenches per nominally rated HP ton could be no more than just a loose rule of thumb, at best. If I've got the size/type of my loop field wrong, please express it to me in terms of heat load (what the HP will actually be required to do, rather than its rating). 4 trenches, 300' long, 2 pipe/trench, 13 gpm, didn't do it for a house you've experience with? Please tell me the house's heat load. Its heat pump rating seems only relevant to the extent whether it's been confirmed that it exceeds the heat load. Admittedly, I'm a novice. Please explain if I've got this wrong. (Speaking rhetorically, not directly to you gsmith).

    Anyway, thanks for the comments.
  11. jyoke

    jyoke New Member

    Chris, thanks for sharing your details. I'm leaning toward trying to work something like this into my system. From the gov website that gsmith linked to above (thanks again gsmith), I see that my soil is considered to be 'droughty'. Although this is most likely considered to be a summertime condition, it's another bit of data to give me apprehension about consequences if we have a drier than normal winter.

    I'll probably be putting in some kind of pipe system that can be used to add water. Ideally I'd like to get it connected to the run-off from my gutters. What diameter is the PVC pipe in the trenches? What diameter are the drilled holes? Have you had a chance to test its effect on EWT?
  12. ChrisJ

    ChrisJ Active Member Forum Leader

    It was 10 yrs ago... Pretty sure it was 3/4" PVC, the holes were 1/8" every 2' for 1st half the length of the trench, 1/4" every 2' for 2nd half of the trench.

    I have let the water run for 3-4 hrs near the end of the summer, that dropped my EWT only a couple degrees.

    My trenches go up hill a little so I wouldn't be able to use gutter water easily.

    Was not very scientific... maybe with all your reading of hydronics you could come up with a better system of piping with holes in it.
  13. gsmith22

    gsmith22 Member

    The PVC pipe system/idea is certainly a lot more elegant/functional than my hose idea - very cool. Why not set it up like a pressure regulated mound style septic system - not using waste water but the pumping concept? Collect your runoff in a cistern and then pump it through your pvc pipes laid out like a manifold in the trenches. Pump and manifold concept is used in non gravity mound septic systems to better distribute the effluent which is the collected water in your cistern for your system. This way you get water to the far end of the trenches and not just closest to the cistern.

Share This Page