Home  5  Books  5  GBEzine  5  News  5  HelpDesk  5  Register  5  GreenBuilding.co.uk
Not signed in (Sign In)

Categories



Green Building Bible, Fourth Edition
Green Building Bible, fourth edition (both books)
These two books are the perfect starting place to help you get to grips with one of the most vitally important aspects of our society - our homes and living environment.

PLEASE NOTE: A download link for Volume 1 will be sent to you by email and Volume 2 will be sent to you by post as a book.
Buy individually or both books together. Delivery is free!


powered by Surfing Waves




Vanilla 1.0.3 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to new Forum Visitors
 Join the forum now and benefit from discussions with thousands of other green building fans and discounts on Green Building Press publications: Apply now.




Housing - New Build: Therm simulation suspended floor

Bottom of Page

1 to 16 of 16

  1.  
    • CommentAuthorward32
    • CommentTimeNov 13th 2014
     
    I have modelled a suspended insulated floor (vented under floor space) and while I can see some thermal elements (Foamglas Perinsul) don't do very much I am a little puzzled as to the results displayed.

    I created materials that match what I am using: Brick outer wall, Turbo blocks under floor U=0.11, EPS segment under floor U=0.034, Concrete beams with EPS infill U=0.034 and blown EPS U=0.34
    Boundary conditions are external wall = 10degC, Internal wall = 20degC and under floor = -10degC.

    Do the temperatures on the isotherm lines show real temperatures?
    I expected to see the internal wall and floor surface to be close to the start boundary condition of 20degC.
    It does highlight that taking insulation deep in to the foundations is unnecessary!

    I would appreciate any comments.
    •  
      CommentAuthorSteamyTea
    • CommentTimeNov 14th 2014
     
    Not knowing Therm that well, so may be guessing, but are some boundary conditions/interfaces missing 1.e. a ground temp
    •  
      CommentAuthorfostertom
    • CommentTimeNov 14th 2014
     
    There seems to be a strange lack of isobars inboard of 12.5C in the floor slab, but why do you say that internal temps are not close to 20C? Try changing the temp interval of the isobars.

    I don't think you have 'adiabatic' set for the 'cut ends' of the construction? Each set of cut ends should finish in a single adiabatic line, not staggered. Otherwise Therm treats the ends as a heat-loss surface, but without a boundary temp defined - v confusing. You have to be v clear which are the heat loss surfaces, and to define a boundary temp for each of them - and where such surfaces meet, their boundary temp must be identical, otherwise you get a local infinite temp gradient. That means that differing boundary temps must be separated by a stretch of adiabatic boundary condition.
    • CommentAuthorSigaldry
    • CommentTimeNov 14th 2014
     
    Get yourself a copy of BR497. Detail 4.3.2 is what is needed. If using this software for anything other than play – that documents the holy text!

    (useful additional information also in BR IP 1/06, BR 443, BS EN ISO 6946, BS EN ISO 10211 and BS EN ISO 13370)

    Depending on the junction, you may actually need 3D modelling if it’s a perpendicular junction – e.g. concrete beam and block, with the beam into the wall.

    The void and the ground should be included in the calculation if they’re not already.

    BS EN ISO 13370 sets floor breadth (b) at 8m, The floor for the model uses 0.5 x b, so should be 4m internally from inner face, the ground externally 2.5 x b, so 20m from outer face and same (20m) deep. Ideally take the wall construction (lw) at least 2m from the junction up.

    Adiabatic junctions for the furthest edge of the floor construction and top of the wall. Tom’s right about the non-staggered adiabatic boundaries.

    The temperature of the floor void below the floor construction for boundary conditions (Tu) comes from a U-value calculator which can determine that (based on the parameters in BR 497). Stroma's U-value calculator can calculate this alongside a floor U-value calculation.

    Temperature external 0°C with 0.04 surface resistance.
    Temperature internal through wall 20°C with 0.13 surface resistance.
    Temperature internal through floor 20°C with 0.17 surface resistance.

    For the boundary conditions to the floor void, floor underside and ground would be whatever Tu temperature that the U-value calculation states, with a resistance of 0.17 and the wall to underside of the void would be same temperature, but 0.13 resistance.

    For calculation of a psi value you use the sum of relevant flows (Q) in the calculations, but would only be considering the relevant ones, i.e. those through external wall and floor boundaries only.

    A bit of training is helpful – Peter Warm in conjunction with AECB I believe runs a day module on thermal bridging using THERM, or STROMA occasionally run training using HEAT software. The BRE haven’t run any courses as far as I can tell in a few years now, although I think BRE Scotland (Tim Ward) may if sufficient interest…

    Very easy to make mistakes which could give entirely wrong results...
    •  
      CommentAuthorfostertom
    • CommentTimeNov 14th 2014
     
    You understand that Adiabatic boundary means that no heat passes either way through it?

    It only corresponds with reality when it's perp to isobars which have settled down to run parallel (or rather, they do run parallel once the adiabatic boundary definition is placed) e.g. the inboard end of your purple slab. Isobars running parallel means heat's only flowing perp to them, not tending to spread sideways - in that case an adiabatic boundary is realistic.
    • CommentAuthorSigaldry
    • CommentTimeNov 14th 2014 edited
     
    Should end up looking something like this (If done as 3D, or 2D):
    • CommentAuthorEd Davies
    • CommentTimeNov 14th 2014
     
    Isotherms (lines of equal temperature), not isobars (lines of equal pressure), I assume.
    • CommentAuthortony
    • CommentTimeNov 14th 2014
     
    Might need a tiny bit of mitigation at the very corner of the floor?
  9.  
    I arrive at something like this
  10.  
    And energy flow
    •  
      CommentAuthorfostertom
    • CommentTimeNov 14th 2014
     
    Never mind BR497 which IMO is set up to enforce the vested ideas of the conventional insulation industry, with scandalous result such as the bent-science suppression of the alternative (if embryonic) findings of the multifoil industry, and ignorance of the effects of thermal massiveness especially ground-coupling.

    I've experimented in Therm with another approach to the ground effect, especially with downstand perimeter insulation in lieu of under-slab insulation, enclosing a chunk of subsoil that's in contact with the interior, via barely-insulated slab.

    Instead of making the top-of-subsoil boundary a flat 10C (surely ward32 you didn't really mean -10C?), I put the 10C boundary 1.5m down, with 1.5m of insulative subsoil above it. I made the top-of-external ground boundary equal to the average air temp for the month of the heating season that I was checking - and that was a little or a lot higher than 0C for all months. The results were very interesting - try it.

    That 1.5m figure is the depth at which ground temp is supposed to settle to near-constant all year round, at a temp equal to the annual average air temp. Both that depth and that temp may vary quite a bit from place to place - try varying it, or make it dip a bit deeper under the floor. Doesn't make a great difference - but is all the difference in the world compared to crudely (and unrealistically) saying all sub-slab subsoil is 10C and external ground surface is 0C.
    • CommentAuthorward32
    • CommentTimeNov 15th 2014
     
    Thanks for the replies and some useful pointers on declaring the boundary conditions in Thrm. I amended the boundary conditions so that Adiabatic breaks transitions from one temperature to another and it made a small difference. I did already have Adiabatic set for the sectioned ends so no improvements there. I changed the isotherm min, max and intervals from Auto but still got the same results. My results show the sort of heat gradient contours one would expect, and other contributor posts verify this, but I still cannot get the internal and external surface temperatures to start around about the boundary temperature.

    The image below has the following settings:
    External air temp above ground = 0
    External temp below ground = 5 (and adiabatic boundary separates the two)
    External air temperature below the ground slab = 0 (suspended & ventilated floor. Adiabatic separates boundaries)
    Adiabatic regions at section ends - left and top
    Internal air temp = 20

    I have the information I need from the simulation (thanks to confirmation from other posts) but I would like to understand where my error is using Therm.
    • CommentAuthorward32
    • CommentTimeNov 15th 2014
     
    And in case it is possible I have tried to upload a zip of the Therm file I have created if anyone cares to take a look.
    •  
      CommentAuthorfostertom
    • CommentTimeNov 15th 2014 edited
     
    First, you need to align the components of the two cut ends (slab and wall top) in a single adiabatic line, not stepped/staggered, which is causing perturbing and undefined heat-leak effect, visibly so at top of wall.

    Second, you need to have a block of external, conductive ground outside your wall. Realistic heat loss/flow is down thro slab, curving outward thro the found wall, curving upward to the ground surface. External wall/air face above ground, and ground/air surface must both be 0C. The underside of the block of ground should be 5C (if that's what you've chosen), linking up with a chain of 5C boundaries right way through to your underside of slab. It's not possible to do it by defining 2 boundary conditions at the external wall face, simulating ground at 5C below ground level and air at 0C above ground level, the way you've done it. If the two differing boundary conditions meet at ground level, creates an infinite temp gradient at that point. If they are 'separated' by a short stretch of adiabatic on the wall face, that's still a v concentrated temp gradient.

    Third, your internal surface temp looks OK - approaching internal air temp. It won't match because there's a surface resistance, which is greater for an upward horizontal surface (floor) than for a vertical surface (wall). The external surface temp is totally confused because of my previous para.

    You know there's a Therm forum http://windows.lbl.gov/software/therm/forum/index.html ?
    • CommentAuthorswindler
    • CommentTimeFeb 17th 2017 edited
     
    I want renew topic. I can download any attachments below and without 2 examples for me is too difficult to understood BS EN ISO 10211 C.5.1 or C.5.2.
    I must calculate the psi value of the detail with a suspended floor (in the basement is no air change). I want to understand whether it is worthwhile more to deepen external insulation to the ground.

    Attachment (calculation R values; detail.dxf; detail.thm, THERM 7.5)

    1. is the THERM file (model) good prepared ?
    2. i understood that PSI value can be calculated in two ways?
    3. i think EN ISO 10211 C.11 formula is the easiest way
    PSI=0,1248*6,210-2,55*0,12-0,5* 7,41 * 0,16= - 0,124 , this is good calculation?
    but in another way with coupling coefficients i must get the same result..but how?

    This forum is my last hope :confused:
    •  
      CommentAuthordjh
    • CommentTimeFeb 17th 2017
     
    Posted By: swindlerI want renew topic.

    I'd suggest instead creating your own new topic for a new discussion about a new situation.

1 to 16 of 16

  1.  
Add your comments

    Username Password
  • Format comments as
 
Back to Discussions Top of Page
   
The Ecobuilding Buzz
Site Map    |   Home    |   View Cart    |   Pressroom   |   Business   |   Links   
Logout    

© Green Building Press