Radiant heating installations in Montana by Leroy C Horpedahl A THESIS Submitted to the Graduate Faculty in partial fulfillment of the requiremenrs for the degree of Master of Science in Mechanical Engineering Montana State University © Copyright by Leroy C Horpedahl (1950) Abstract: Radiant heating is not something very new. The Romans used it over 2,000 years ago, but it was not until 1908 that it was put on a commercial basis. There are many methods of supplying this form of heat to a dwelling. It may be done by passing hot air through ducts in floors, walls, or ceilings, or it may be accomplished by means of passing hot water through pipes in the same areas. This thesis concerns only the latter. Montana State College was granted a sum of $5,000 by the Danforth Foundation to be used toward the erection of a campus chapel. Various departments have volunteered their services, both professional and manual, to help make this chapel a reality at a minimum cost and at a maximum of student interest. My contribution is the complete design of a radiant heating system for the chapel which is included in this thesis. Radiant floor panels used with hardwood floors have been the desire of many citizens, hence a complete design and cost analysis is given with such a construction. This has been constructed and is in satisfactory operation. Much can he learned from the mistakes and suggestions of others, hence a survey of radiant heating installations in Montana is included in this thesis. In connection with the survey, several inquiries were made concerning proper floor coverings to use with heated floors, The results of the research conducted on various types of floor coverings are also included in this thesis. All calculations in this thesis were performed on a slide rule. The Libraries Montana State Universityo Bozeman Bozeman, MT 59717-0332 S B h e .University Special Collections Telephone o z e m a n 406-994-4242 GUIDELINES FOR USE OF MATERIALS The Merrill G. Burlingame Special Collections of the MSU Libraries was established to collect and preserve materials on Montana and its history. Books, newspapers, maps, diaries, letters, photographs, architectural drawings, and rare books are all a part of the collection. These materials require special handling and security. Their use is restricted to the Burton K. Wheeler Reading Room. University Archives materials are subject to the same patron use guidelines. Check-in Procedures 1. Fill out a registration form and present positive identification. 2. Place personal belongings such as coats and backpacks on our coat rack or on the window ledge. 3. Complete a call slip for all materials you want to use. Special Collections stacks and files are closed to the public. z . U se of Materials 1. Use care when handling materials to prevent damage. 2. Use pencils or word processors only for note taking. Pens are prohibited. 3. Do not trace or write on top of materials you are using. 4. Please to not change the order in which manuscript or photographic materials are filed. If filing errors are detected, bring them to our attention, and corrections will be made by our staff. 5. Please consult with staff for photocopy requests and procedures. 6. Use microform copy instead of paper copy whenever possible. Our staff will guide and assist you. Paper copies can be made from microforms. 7. Upon request, materials may be held on reserve; however, if they are not used within two weeks, they will be returned to the stacks. 8. Wear cotton gloves when handling photographs and negatives. Gloves are available from staff. 9. Researchers are responsible for compliance with laws governing copyright and literary property rights. Check-Out Procedures Materials may be left at your table. A staff member will unlock the door for you as you leave. All possessions are subject to search by Libraries staff. HOURS: Special Collections’ hours are 9:30 a.m. - 5:00 p.m ., Tuesday - Friday. We are closed on holidays. Rev. 9/94 R&DI&NT BBA2IRG IRSTALIATIQRS IR MORTARA by LEROY C. HORfBDABL A THESIS SubmLtilled to the Graduate Baculty la partial fulfillment of the requirements for the degree of Master of Selsnoe In Moohanleal .Engineering Montana State GmS^anj" Ei:SaIiffng“SoSmrtte© Jaxiy' Graduate/DlrisIon -2- TABLE OF CONTENTS ABSTRACT................ ...................... . ......... 3 RADINAT 1-EATING DESIGN FOR OANFORTH CHAPEL................. It. Calculations using g-" tubing.......................... 5 General specifications................................ . Calculations using 3/t " tubing,....................... Layout of control system......... ....... ............ icj Layout of panels using V> tubing. ............... ....20 Layout of panels using 3A " tubing.................... 21 Elevation views of Danforth Chapel.................... 22 RADIANT FLOOR PANELS /ITH VOOD FLOORS Calculations................ ... Construction................... Cost analysis.................. Layout of control system.............................. 38 Layout of panels using short headers.................. 39 Layout of panels using long l eaders.... .............. SURVEY OF MONTANA INSTALLATIONS............................ !p. FLOOR COVERINGS OVER RADIANT PANELS........................ 55 LITERATURE CONSULTED....................................... 59 Ar iENDIa ................................................... . 9 4 6 # ' Eadtairb heating Ig-. not .something very neWt. The Eomans used it oven E5OOO yeans ago, but it was not until 1908 that it was put on -a eamjaienoial basis* There are many methods of supplying this form of heat to & deoiling* l t m a y b<& dona.by passing hot air through duets: in floors, Walls5 or ceilings* or it may be accompli shed by means of passing hot water through pipes in the same areas„ This thesis concerns only the latter* Montana State College was granted a sum of §5a000 by the Panforth Foundation to be used toward the erection of a campus chapel* Various departments have volunteered their services, both professional and manual* to help make this.chapel a real­ ity at a minimum cost and at a maximum of student interest* My contribution is the complete design, of a radiant heating system for the chapel which is included in this thesis* Radiant floor panels used with hardwood floors have been the desire of many citizens* hence a complete design, and cost analysis Is given with .such a construction* This has been constructed and is in satisfactory operation* Much can be learned from the mistakes and suggestions of others* hence a survey -of radiant heating installations in Montana is included in this thesis* In connection with the survey* several inquiries were made concerning proper floor coverings to use with heated floors*. The results, of the re­ search conducted on various types of floor coverings are also included in this thesis* All calculations in this thesis were performed on a slide rule* .A R&DIAMT EE&TIRG 8ZB37EM PQR A PROPOSED DAI'JFORJPI O M i 5SL OH THE OAMPHS OF MOHTAM SfATS GOLLBG-E In tlie design, Ole any building in Montanal9 careful con*alderatlon must ba given to the type of heating system used with respeet to oost, msintalneaoe* and the ability to heat during extreme weather conditions => In a more specific cases such as the design of a chapel^ other conditions must also be Included, such as quietness of operation^, comfort^ and earIuslon of elewenta which would tend to deviate from the modern­ istic theme of the contemporary chapel* With these restrlc- tions in mind, radiant heating was suggested and the following do r>xfin. Is proposed* &7o complete systems are shown a -one with copper tubing and the other with 3/%." copper tubing. selection of the The tubing system was chosen because of the better heat distribution in the concrete floor with shorter spacing distances» Also9 the cost was slightly less* The first step in any heating design is to calculate the heat loss of the building* In radiant Installations* it la necessary to break up the heat losses into each individual room because the amount of panel installed in each room is a function of the heat loss for that particular room*. In this particular case, recommendations for insulation and windows were made in conjunction with the heat loss calculations at the request of the architecto Heat loss calculations for both single and double glass windows .were made to determine whether single glass would be feasible e As the calculations which follow show jj it would not be ix^aetieal here in Mcmtmia to use the single glass over such a large area0 First of all* frost formations would oeeur on the window with the resulting moisture on the walls and floor below* Secondly^ it would he impossible to heat this chapel under extreme conditions with just a floor panel using radiant heat without exceeding an 8^ °F floor temperature'o Wall or ceiling panels were not eon* sidered in this design because of the nature of their eon* Struetion0 She heat loss calculations with the necessary recornaend^ at ions and assumptions ^ follow 0 All heat transfer coeffi­ cients (XT) and resistances (R) were taken from the A SHFE Guide (19^9)« Heat !,ess Calculations Ac Sanctuary and entry I 0 Celling R Inside air film** oo . 0 ,****«,» 0O'0 no 0 e O1e o. » e 0 &0 0 e0 0 000 O % = 1/20*173 = 0*05 Btu/hr-3g,fto-*P Ceiling area = 53 % 22*| « 2 z 6 * 5 n 2|- s 2| ~ lli|.9 sq* ft*. Iieat loss * 1149 22 o05 % 90q * 5160 Btu/hr ~6~ 2o- Walls H I^xsj»c3-@ aX p xaIxn# ^ @».o*^ ^ *» .ft * Q 0.6 0 6 2^y^32 » fro-# #“ 0--*0*** **»* * *#**-ftft-.ft.ft*ft ... 1,055 25/32-" pine she athing»,,*&*,**. ,,*..*,.,. ,.* 0,98 I-Jp XiiILXX0X'ClI xjOOl o».0ft*f t . -f t -oftO-ft**« *iiftft-O.e-a,oi.*#»ftftftl-l|-O80 25/ 32” pine sheathing»«.*■» 0,0 ft. Oft98 25^^32 ^peclv/ooci0jo."*-*»a.< dft>*ft*e »* a.**0** -0* 0.ft, 1.055 Qntstdle aap ^ v ^ o. o. on** #-o- o^ a 0 - »^n- oo »- *.ft. 0.167 « ' H U = 1/1 9.6^3 = o05l Btu/kr-aq.fta- F Wall apea 9& % 33.5 + 9& x 19 + 9i a 22*5 = 713 ea. ft. Beat, loss > 713 x .051 Z .90 = 3270 Btn/hr 3. Doors (assnmed thickness I 3/l|.rf) B = 0,51 Btu/hr~sq,ft.*F Door area = 7 x 3 5 21 sq.eft» Beat loss = 2 a 21 x *51 x 90 = 1928 Btu/hr l{_. Stained glass window Tl 1 ,13 BW/Br-sq.ft,-*P Area - 9/ xc 6 = 57 sq.ft. Beat loss = 57 % 1*13 x 90 = 5^800 Btn/hr 5« Skylights U = »45 BtuZhr-SQftft00Bt Area = 5 X 2*5 X 2*5 ~ 31.25 sq.ft* Beat loss = 31.25 x .45 x 90 F 1,265 Btu/hr 6, Glass wall area. . 1 Area"^ 6-J x 43 t 2-| x 50 = 4^4 sq. ft.. (Assuming the use of double glass)" U - »45 Btu/hr-sq.ft.-°F -7“ Heat lass hol\. x ,i(-5' x 90 = l6>350 Btu/hr 7» Infiltration-— oracle method Windows Factor for average windownon-weather stripped— 39=3 cu. ft0/it. crack/hr Two.. „ . 2-|-1 x 3 { " 22 ft. of crack One... f.2%-* x 5 f = lj3 ft. of crack Total.............. «37 ft. 37 % 39« 3 = I W t cu. ft^/hr Doors Factor..».,110.5 on* ft./ft. crack/hr Two doors... .3 ’ x 7 1 - lf.0 ft. of crack IpO x 110.5 " l^i{-20 cu. ft./hr Total = I ^ f Wt20 = 5^874- an, ft^/hr Volume of sanctuary and entry 1180 X 10 = 11,800 CU. ft. This only gives -J air change per hr.4 therefore^ assume I air change per hr. Heat loss = 11S:8QO x *018 x 90 = 19 a100 Btu/hr m e r e *018 % .2^ x .075 and »;2k “ specific heat of air .075 ~ density of air^ Ihs./cu. ft. 8., Floor (ground loss) Heat loss a Btu/hr Gorlxng. •00. *.. ».* * c»........ »».,*»... WallS . . . . » 0 0 . . . » » e .. . . ..»*, * Doors . . Stained glass window.»„...»............ 5l8o 3270 I92S 5800 -8~ U-I^SS u?a3.X» esite. 1262 O6eefreeceeee^gfl ,16320 IllfXeX 'fcl’atios® e # e » » e e « » » e 4 , . i © cal g e *» » « »ji„„ „, »•, ,19100 »»<,»,,-»<,.**»».«,»o Btaa/lii5' A-S-Sujne ground loss 15^ of total heat loss?. 12# s 22873 = 7930 Btu/hr rotal % 7930 f 22873 a 6o»8o3 Btu/hr' B» Sacristy I 0 Gelling area Area = 17 z 11& - 7 x 3 - l6o sq0 ft* G “ o05 Btu/hr - sq.» ft* -,*? Heat loss ^ l6o x &02 X 90 = 720. Btu/hr S0 Wall area Area = 11& z 9%. 4 & % + 10& % % G = *05l Btu/hr w S q 11 ft* 4 261 sq. ft* '' Boor area = 21 aq. ft. Window area a 21 sq* ft. . •. ■ (Total area.s 261 - 21 - 21 = 219 aq. ft. Heat loss - 219 x »O/l x 90 = 1000 Btu/hr 3 o- Door area Area = 21 aq* ft* G - »51 Btu/hr ~ sq» ft* -°P Heat Ioaa = 21 x »21 % go = 96%. Btu/hr '-i-» Window area (assume double glass) Area = 2& % 8# a 21 aq* ft* G = *1x2 Btu/hr * sq. ft, -°F Heat loss = 21 x Ji$ x 90 = 820 Btu/hr 5t> Infiltration— crack.method Windows Factor— -39«3 on. ft./ft, crack/hr One.....2/' x 2*|**' — 9^2 ft. of crack 39*3 % 9*3 = 373 on. ft. Doors Factor-— 111./ -Cd* ft«/ft,, crack/hr One.*.*.3 *" x 7 + = 20 ft. of crack 110.2 s 20 = 2210 cu. ft. Total % 373 + 2210'= 2283 on. ft,/hr Volume of sacristy = 160 x 1.0 = l600 eu* ft. Orack method gives approximately two air changes per hr. Use Infiltration by .crack method for heat loss. Beat loss = *018 a 2283 % 90 - # 9 0 Btn/hr 6. Floor (ground loss} Beat Iossj. Btu/hr OoxImg 720 Walls * . « * * * . . . a * . * . . * * . * . , - * * . * . * - , . . 9 1000 Do ors ... no» * . * . * . # . **...*.. o- 9hlx V^mdOVTS, 39O Imxltration. * a *. ** o*♦ a *.*.**.. I4.I90 Total ~'J'J/TC Dtu/nx . * O 0 . * . . - I A - O . . * * * , * . . A * * * » O 0 0 0 a * * 0 * . . » , * < * A * * 1, . A * * * . A * . * . . . A * * . » *: * * . . o . . * o o O - O O a * * .... * 0 Assume ground loss = 12# of total heat loss 12# x 7724 == 1128 Btu/hr Total % 772^ + 1128 % 8882 Btu/hr Total Beat Loss For Chapel Assumed Use of Double Blass in Bast Wall Sanctuary and entrance 10 0 © '6 » ti- e d o a 2 Q1 vlftSt-Li^ o o tf» ■fr o d-dd'o'd tf- e o o o o odd O d d d e oo - a e . 6 - o d < |"'0 Oito.ddO O i!)O-O-^0 o o » -a o d * * * ttd o, e e -<i d ti er d o ft d d- o d .Stained glass v/indow^ o******** ^8OO Slc^w-vL-Xg^l.tjS1oe- e otie *r o o*iO- » d-ditotin Ao d I2l0^' I O o O iI a s s Ta I n f i l t r a t i o n * * , * * * » . , v/all A# o* ^ o o » 8 D ^ l O O r o-a 0 e d o p a o o^osioaao a * e 0 A d a e -O o a , * o o e * a a 1 9 1 0 0 T 9 1 0 J?oval &0 **a d o »» 0 * Oj» doe AO* a * do* 0 800 OI 33xi%/lir Saerlety 1 d 2 a OOXlXXl^ 0 d- ti i>-£• ft ft ti -* »■ ,0' fr ti *'» I'^ 1 1 S a * * » t i - o r d e « 34 oe d a e ft 0 0 e: 0 0. a 't o o o o Aftodftfto OOO r B a ' A f t o f t e - f t A f t o o e o - f t o a o » o e e- »a 720 1000 a b V J l n d O ^ T S >d e a eft. ft O ft ft oftft' ft O o o d V f t f t f t f t e 5<. Infll tration* •* + ******* v,*^*'**-, 6 ^“Floor d ft o ft A 0 ft ft ft c. 0 a 6-o * 0. «. ft ft o d * f l * 0 Total 0 d o ft- a 0 «>. • ft O -ft o 0 e ft a o d e d 0 d e f t d- !{1 9 0 1158 3: Btu/hr Total heat Iosa0* » * * * » t f « * * « o « « » o * 69,685 Btu/hr Total Heat Hoas For Ohapel Assumed W e of Single Glass in Fast Wall Sanotuary and entrance 5160 1928 3270 5800 5 S h y I X g i l t S o d e o o - o o o o o o 'oeOoOo-oft-o-o 1265 8-o GlxXSS wall OooegfrOo eg oOfrfrfr 000Oo- IpLlOO 7 0 Infxltration-o 0,0 ©0- o-o-o o.0 0 oo-°-00 -0 & 19100 O e r la a ig p I o o * * o @»# *■ # » ^ o < » o # @ 0 0 0 0 0 -OOOrS o , O 0 O o O O O e d o o O g O d O O O -O 1-O. 3 O IfifallsO O O O o O O O O O o - O Q O Of 0 O- >0 O O O 0-0 O o Ito Stained glass win-doe0«,-»«..0..*.*,0 2-0 6 CF F l O O r S t i o - O f r O - o o o f r d o f r O - o - O f - g - o -O- 0 - 0 0 ' 0 0 6 Total 0 OQ- 0 O g- 0 0 0 0 OOfrOO Btu/hr O O ' O f r O O o-0 0 - 0 Sacristy Oexling O f r O f r f r O - O 0-0 - 0 0 0 0 - 0 0 -6 f r f r O o O O O 0 Vfalls O o o 0 O 0 -O O O f r o OfrfrO- O O frfrd O O O O-ADoors O o ofr o f r f r o d O o d o e ofrfr.ofr 0 0 d . f r O O I 0 d 3 720 1000 <b If 0 xndon? Sfrfr O O - O f r g f r f r f r O a-fr-fro-fr-d-fr-poofr-fr 5 0 Infxl u-raxxon ©0°*00 *«♦■ og,0o»0 *.t* % 190 6 0 F lO O r o . o f r f r o o f r f r f r f r . f r f r - o o f r f r o o o f r - o o a e o 1158 T O vul O o f r o 8882 Btu/hr -O- frOo-frfri f r f r t t o o o o O-Ofr -O- * 0 0 6 Total heat loss 6 a a a ® -e v * a o*a 0 9 4 > W 5 Btu/hr -IiDealgik of Ploop Panels Hotel Heat losses based on the use of double glass'in all windows except stained glass mural* Design temperature of water is U O r F with a 20*F temperature drop through the circuit* •■ Depth of burp Is 2 inches * A 0 Floor panel - copper- tube 9" and. 6 ft 0*0» I*. Sanctuarj and entry Heat less - 60803 Btu/hr Panel rating used = $0 Btu/hr-sq*. ft* Floor panel area required -■ 60803 9 50 - 1215 sq, ft. Actual floor area = ll60 sq* ft* Use i I! copper tube 9 n and. S n 0*0* Length of tube required » I0I3.O x 1*3 - 1352 ft* 120 x 2^0 = 2^0 ft* Total** e • e « o '« 9 « 0- « -0 «- 0 «0«> 4 *..*** 1592 ft* 2* Sacristy Heat loss - 8882 Btu/hr ■ Panel rating used = 50 Btu/hr - sq* ft* Floor panel area required ~ 888-2 r 50 “ 177 sq* ft* Actual floor area ” 182 sq* ft* Use copper tube 9 H °»C0 Length, of tube required ~ .162 x 1*3 ~ 210 ft* Total length of tube, required - 1592 t 210 - 1002 ft, 3* Circuits Uoe oirouito of approaiimtaly 180 feet* Mo* of circuits = 1802 * 180 = I^Approzlmatoly Circuit #10„ r? ir a f«o-o«dfoeoOOCO OO0OO0 18k Pt* a-oOb'OOOCti-OOO-OOOOOOO 186 n u o o » oo-teco o Q» ocot t t eooo 173 ecteoooooooo. ^coe-oooo 181 n Mf-DCf>oofoo O O Q - o p t i r o p o p - a o t i O t e t e - o - o c 170 ii t u.Co <> o- o o <> O O O 1O P ^ O t e O O O 0 P Cr O oteteo 186 r? ?/'To o o e o o o o C f t e o t e o t e o o t e o t e t e o o o o t e 185 I? IQl Il ofiotitetyteOtkxko-teteooote-c-te >y'9O aOti,ti-OO t e o t e o ^ o t e - o t e o o t e O t e A O o o 182 If v/lO OGOOO o b- o O o O if c tete O'te^te-O - O O t e O 166 if ^JotalO O O Ote-teO-teteteOteCtete-teCftO-OOte Ptx o- y- e o fi? H D o- o o o <j o -< U is U n si M t? N -C o - =e o ti O a ‘/ / ' 3 O fi: o e tf !!-o Supply nnd return hoadors il.5<3 2" Ur0 I» pipe.-, If. ft a long oaohc 5 o Pump Tj.oo 1-|M standaixl or l|-tt Mgli head 6 o ITater boat or RociQBBnend use of hot water comforter similar to dOif.-li as supplied by General Fittings Oo0 and rated at lt.7O oq# ft. of hot water radiation (5 lb* stooati)„ 7 6 Valroo Use stop and waste valves on the return header to facilitate removal of air during the filling operation* Dao gate valves on the supply header* ?be general apoclfieatioua, H a t of materials, and cost fGllcnfO0 General Speolfieatiauaa Beat Ioaa from the chapel is based an the insulation quantities given 'beloiw Any deviation trill possibly change the entire s e t - u p heneo strict adherence is advised for aat** lofaotory operation of the heating Systos0 I0 Insulation: (1) Oeilingoa O-O^m rock wool or equivalent (2 ) Walls® <.<>.«a rock wool or equivalent (3 ) FlooroaaaaaI tt fiber glass insulation at opposed perimeter of slab® Extend 3 s inside along the rack Wall0 II,a. Windows: (I) Sanctuary„«„„double glowing throughout of a ^thormopaneN nature^ three operating -sections^ one over each door approximately 2 * 6 ” x 3 4 Orr and on© in the end bay approximately 2 8 6 H % 2' 0". (B) Skylights6ti®0.double glass (I8n airspace}® (3 ) SacrIsty00&«„double glasingj one operating section approximately 2 r 3 ff x 2 1 Gnc IIl0 Poors? (I) All doors are flush wood doors I"3Atn thick® IF0 Floors z (I) Ihe floor is to be constructed of a V 5 concrete slab poured on. 6 !,t crashed rock or extremely coarse gravel® The floor must be poured separately from the foundation and may havo- a concrete surface finiSh5 tile^ linoleum5 or wood nailed to sleepers embedded in the concrete® it A 3A A copper' tubing supply and return line shall bo imbedded in the concrete from the heating roomcto the east wall to be used with radiant base­ board* along the east wall of windows should it become necessary. VI. Construction of the heating system will be done b y the student# taking the radiant heating course during the fall quarter under the direction of Dr. Et F, Mulllkln* if agreeable. VII. Bill of material* and cost; I circulator (!&" atd. or 1&" high haad)....$ %9»00 I Indoor thermostat (Perfex line voltage)... 11»5>0 I converter (almllar to 00^*^* General Fittings Co.).... 73.20 I floating thermostatic trap ( 3 / % $ . * ) ^ . 1 ^ . 0 0 I temperature regulator (no. 9 2 @~LC Fulton Sylphoa^li*)......... 80.00 I expansion tank (20 gal* ) • . . . . * * # . • . 16•00 I automatio air vont•••»«••••*•«••••***•••«• 10 shut—off valves (1/ 2 ®)* ........* *.**«.*«♦•*. * l«30 13 •50 10 stop and waste valves ( l / 2 " 1 $ . 0 0 I supply W a d e r ()#. ft. of 2* W*I. plpe)^.... ^.00 I return header U|. ft* of 2* W.I. pipe)..... $.00 ft. of 3/%" copper tubing type L........ 20.00 90 I check valve »»««••••••••••»•• ««•*«. 6.00 I reducing valv# and I relief valve (combined) I altitude gage (thermometer included)...... 1800 ft. of i* copper tubing type 12.00 7*$0 282.00 4. shut-off valve* ^0*00 30 ft.. water pipe ® ^2^ per ft.*...*..* 12*60 Mxseel3.anions elbows, tees, etc ........... 8 IO0OO fJ1Ou a I V illa O O O O f l O O O O O O O O O O C O e O f l O fl I, O f l C C O O f l o O O O » ^O O 6 O O O O D The prises quoted above are subject to edueatio&al discounts which can be received If ordered through the Mechanical Engineering Department. I3C. The return header will be located In a pit slightly b e l o w .the floor level to facilitate drainlag the system* A sewer outlet will be located, in the pit. X* : The headers may be located along the north wall instead, of the south wall as shown on the drawing.* This will help prevent excessive heat in the utility room. Xltt • At each joint in the concrete, wrap each tube with felt or some cloth-like material to prevent the tubes from shearing off due to ground movement 0 Floor panel— 3/%" copper tube 12" and 9" 0,0. I. Sanctuary and entry Panel rating ■« j?0 Btu/hr Heat loss and panel area same as with i-n tube P Length of tube required - 9li„0 x I 200 % 1.3 * T o ba l c> 0 f l o o o . o 0 » on o oo o fl fl O O so o . 9ii0 ft a 260 ft 1*200 ' f t 2o SaG&la&jr Hoat Z033 and panel area oamo &o %ltb tube &@mgth of tube roqalzvd. = 162 a I = 162 fto #otal length = 1 2 0 0 * 162 - 1362 ft* 3* Olroultg Hoo G ircu ltg o f approzhuatoly 2&0 f t * Ho0 Ox c i r c u i t a «* 1362 4 2S0 s approxim ately 5 O x r c t L i ‘0 ,/Io o«croaai>oyiroa,oe»t>Q».cia-ao(y 2 7 9 f t <3 ^ -;'i-2 « ........ oAA s............. »e ooti.tiooco SQO tf Ei !fjP*ooeott^o-ooiS Ooqfeo-Obtao-Qb 00 ^ n < 1 Q 6 O O<ty0 0 4 0 t k O e O 9 C t i o x y v oo O ^ n >/'5Q»6*0.0,6 16«Oo-OtiO OO.titiOOQ«O.O1 273 ^ l IlOtnlo OtiO Otioabootbooa*otb% Gq eo&odo-IjyJt X1 Oo 4 o Supply and return imadors 0ac Stl Wti Xa p Ip o^ ii. ft a ©aeh Je I3ISiip Woo 1§" Otandard or 1%" high hood 6* Water hoator Hocoiiinicmd 1100 of hot vator coiwortor similar to no* GGk-.^ ao supplied by Goaeral Fittings Go, and rated at ^70 ag» ft* of hot trator radiation (J lb. otoam* 7o Valwoa Hoo atop and %aoto valvos on tho rotura. header to facilitate removal of air during tho filling opera­ tion* Hso goto valvoa on the supply hoador* Sbc coma gonorul spoolfloatiana ulll bo used for the 3/^" tubing ao for tho &" tubing* 17 Llmt of material* and eomt; I olrmulmtor mtd* or 1^* M g h heed) B & &*..***..$ ^9*00 I indoor thmrmomtmt (Perfmx line v o l t a g * 1 1 * $ 0 I oonrartmr (mimllar to GG^-4* General Fitting# Co**** 73*20 I floating thermoetatio trap UA")*.***....*.***..*** ' 4*00 I temperature regulator (no* 928-LG Fulton Sylphon-I^" 80*00 I expansion tank (20 gal* 16*00 I automati* air v e n t * . !$.*30 6 shut-off v&lv©a (3 A ” )*«.•*•***«»*•*•«*.»»»**»•*.*• * 11 *00 6 atop and w&afe® valyas* ••#•**•*«*•* «****».......... * ** 12. 00 I supply header (% ft* of 2" #* I. pip#)*.*..*.*.**.*. 2*00 I return header (4 ft. of 2" #* I* pip#)**.*...*..*.** 2*00 1490 ft* o f 3/ 4 * sapper tubing type L * . . 332*00 I check yalwa C1%" )*.*** ............ *.*## •* ***.*.*.«.*. 6*00 I reducing Tale# and relief *al*a (combined) &"..*.*** 12.00 I altitude gage (thermometer included)*.....*..*,.**.* 7*20 4 8at# valwem (1"& ).* ***.**.. ##.***. **, #.. 4o* 00 per ft*.*....***.*.*.***** 12*60 Mlmeellaneoum teem, elbow#, ete....................... 10.00 30 ft* !&" eater pipe @ Total. ,1691*10 The coat of material# far the two #y#tema 1# slightly in favor of the &* myatem, tut the added labor would put both of Wamm about equal. It i# reoemaanded that the &* mymtem be u w d because the cloaer spacing will give a more uniform panel - 18- t*mp*r*tur* and the smaller tubing gives more turbulent flow and henoe better heat transfer. This shape! has been designed by the students of the Montana State Gollege Department of Arohlteoture. As many departments as possible *111 eontribute toward its eonstruotion when sufficient funds are available* AIFi VENT SXdr.____u. i' REUEF VALVE t -- -t-'-i TEMPERMUKE REGULATOR 'T H T j CE GAGE f'."! . Y MATER -•Ti REDL :R G VAY E ——®,--‘ CHECK VALVE DRAIN CGCK — "&• L ROOM THERMOSTAT Y MONTANA, STATE COLLEGE BOZEMAN, MONTANA CONTROLS FOB DAAVFOFTH CMAPEL HfATiNG SY ST E M C R. SY TH. BY L. HOriPEUAHL 'd/Zo/sti C K g Y A P R BY DRAWING NO. M E 3/9 BRI.,G TUBE ' OUt JF SLAB AND CAP — TVC /4 :CTTS- TL5E: FC-. FJSS-; ; BALi/ ,T CkTSiCZ T E ' - - Z C F WALLS ElL Na INSIDE 4 > MWfFAL TEMP. +70 F. WCOL I ’i SLi LAT U N 4 'TMWEBAL WCO l IUSULATi JN F - rCm i" FISER-SLfiSE I NSULAlir H At EXPOSED P E M M E T E S VV'LDIWS IN EAST WALL AWC S A C B L S T f - DOUBLE GLAZZL ALi- Li- v SC OF out h COPPES TUS< BURIED 2 : MONTANA STATE Cu LLEGE _ _ BOZEMAN, MONTANA BADfANT HEATING FOB DALFiHTH PMcfOSED CHArEL Cx, g y AAL L T •CA.E: /F D n/v,;. ; -;n. M E 3 2 0 L..-J ' RriHj T'.irt'. ' ^ OF SLfk A-O CAP ---) J ■ >■ cum VY HE-DEF OiT-JCE TEriP -Z C sE WAL - S J " V iN En r 4 " MINERAL VVSOL I N S a. AT; J i FL. % I " FiBEE ~ Si. ASC. TNCLJ ' TV AT E X - J S VViM J S J j CAST VVAi- - AWO S A C f i i S T y — CJUS LE /.-L C.A.CI'!. J CF CJP.'' f p TL'B Bvr--EL z " e:Z:r'' . JV MAC MiV T M 5ATW/G FGA PFJ-CoEC C 4NF:iTrl Cp/iPFL DR. BY L. H Q R P E D A H L A^’ /i' CA. EY TSLfY SCALE % ^ /FT APF 2Y L-AW-LV1 S ,- I. M E SC/ aa LUlLT-Ut HOf 8- T. £ 6 U M W G W O L T U E L S C A L E I I A I I 0 H . W T ( I T / 4" - C- 0 " L [ V ( C A L E I T I 6 H T I I - 0" b .1 N- 1 " O _ U_ _ _ _ I » E L SCALE W .... ^ 6 " I Z 3 4 5 6 7 6 L M S i C N M A t i N M 6 ~IL T M I Y E. q J i u o M A S Z l-Il" 8’- I" 4 ’- 4“ 2‘- II" 5' - 6“ 5 6 " Z - 3" 5' - 8" i x x x x x x x x T I O W E L [ Y M A L E u L E ! T U E M O ti B A 6 N /*"= I'- 0" j-- _ U Z T Y M -IxZ Z x 4- M R C -TWEkMOOAUG - 3" HkC - U 3 COLUMN - Ix6 Il Il - A. " * fbOLT 'Y STAINEb GLASS T U E M O M N E H 4'- 5" 4'- 5" b Y ti. J. O T T E M J H. A---i f t T - L -- C 6 I I & C C A m SIZE Z 1- 9" Z 1- 9" 6'-d" 6’- 4" Z ' - 9" 6'- 4 M I L S T E i N V '1 l A" - r-O" u S T --------- ! M U L L I O N < E C T l OW L I * Il )4 -0 I A 4 U - f> G L - 4 ---- E F 0 2 JUW So! L - X ____ 4 ___ L- 4- E L E V A r i O H S M U H U SECTION C-C If Stl E C T OT 3 V 4I T -23HEATING FOR A TYPICAL MONTANA LOG BUNGALOW WITH HARDWOOD FLOORS radiant Most people associate radiant heat with a concrete floor slab. The result is that they don't consider radiant panels with hardwood floors. They forget that wall and ceiling panels even existJ In this particular design, wall and ceiling panels were not ,feasible due to the particular interior finish of rough logs, hence a floor panel was used. Pig. I - Monte Benson Residence The nature of construction of this residence was a full basement bungalow of 9" log walls, the logs being sawed on two sides. The cracks were filled with "Hydroseal", an asphalt base sealer, both inside and outside. The basement was finish­ ed in the form of bedrooms to be heated from the panel in the floor above. Tlie first step in the heating design was to compute the heat loss. The computations are as follows. " _ _ _ _ _ _ _ _ GooffIolenta of Hottt Transfer (IT) A s- Walls H I s IZlSliO Om f -LllSs c ti v o y f>u un <>Of, Od y y o o Dooo ti o d O O0606 2.0' 9 * l e g t r a il (assum o Bff average- th lo -k n o ss ) o-ol0a00 3 Oaide air fa Xlll0 d0 dCO 0u0a0.CO 0^an c0g 0y00OOvi 0Io Y H ^ I 0Z773 H % lAo.773 » *093 Btn/br ^ sq*. B s Windows Is Single glassed «■ H % I 0.13 B.tu/hr ^ sqe. ft.* *»°F 2 * IbenrtOpane H F 0lf.5 Bt’ti/br «< sq.*.- It90F 3* Single glazed (double bung}- « U a ^ Btu/br C 6 Ceiling sqs ft.9 H 1 -0' I u a x d,0 a x r f xlzii& q . *■&-# ^ o.9. » s * 0^ ^ ,0»- a »: 0 ^ .Cr.»:f> O&6 0 & 2* - 30 10 3© 6,s 5/8 x Ir fXooring-o.0.*-0*^«-.»%&* ^ ««-*. s* a■*0*0■» 2 roeIc v/ool.o o-0u^a<*o.0uUo99 0o-,<1a>>09^^9 y0u0y0 2 air anaoe 0c^ 0@060 o0000o.y000^*0«t>o-o»a 0ud 9 X. ^XZ3 S l Z O q 0n00 O'o-o-c/»--qqrV '»■:o"0o'u-»-0-§.■*yab e Built-up rsofing (assumed thielmesa 3/8" )*« 7 * O u ts id e a i r 0*8Ol rJ0IlO/ 1*100 Xff020 O6277 z H m * 0 0 &0 c 0 <, ^ 0 0 <>0 ^ „ u 0 » 9 »»□ 0 0 d o 9 . *167 R » IX Z o t H ? 1/11*378 = „09 Btn/br - sq& f t / ? Sineo tbo Z n air spade is used for Ventilating purposes# ealeuXato ttH lf from inside fe the air s.paee assuming tbo air spaee is at outdoor conditions* Use .tbo higher value of the. two* R . a x r f x lm 6 0% . 9 6 @0» y y.u». s 6 ,%9 9 » o-9 .*'»,».* * 606 2o.. 5/8tf fir xlooring-o, 0** 6sp-a«spo-.oini,*0sa)6..o'0 & « &-IS- *8OX 3 o’ 2 rook wool,>0&o &c *6■©0000 000 0»®».»0000 n„ 0,,0«36/4.03 k0 Outsido air film (still air)*»,**.».*»***** *6O0 . I-*.' i n s i d e R = W H % l/90i{.2 “ 0*106 Btu/br « sqa. Xt0oF Hao- H “ ol'O Btu/br »■ aq* ft*°F B D* Doe** (Aateal th&ekna#* I 3/S*) 8 » *$1 B$a/br - *q* f&**P R B# Bathopoom wall If & l/&t*2 * ,,Okie B W / b r *. *q* f%**P P. Ioflltratlom & # * * * & mdardbaag# pa* k** (groumd f l o w ) A m m a w i #1* a&wesg* p#*b** (baaaaaot ) 0. Baaaaaat **lla and flee*# B * *1 Btu/b* * #g* ft**F »##% Ieea Galeatatlon# Iaald* ^W#l#a taapa*#tu*a a ?o^P Outalla l*al#p teapaamtu*# ~ 20"P IaalRe Raaamam t -daalawlMmgwNm&ta*# + 70 P Outalda Rmaaamt daalge twqpmeatm* +bO*R A, Llalag Room I* ## H a A*a# e 20$ % 8$ ^ lb* a 8$ ^ 3 W * % b6*) - 3$ x 6*8* - 2* * %*7* ~ 2* * 8* * 162 mg. ft. Rmat lea# * 162 % »093 % 9@ = 1*3$6 Bte/b* 2. Plade## Aema (tbaeaqpen*) * #* x b*7* - 22*9 aq* ft. Sadt Ioaa - 22«9 % 90 = 92.9 Btu/br 3 3 26.9 aq» ft. Araa (alagle g&a&ed) = (28* .% Beat leas a 26*9 a 1*13 a 90 a 2^?%D Bto/br 3 o- Door Area a 3 * x 6 *8 " a 20 sq0 ft* a 918.Btu/hr Beat Ioaa a 20 % $0 k-v OeillBg . .Area a 20 z . % » .280 ft#.'. Best Ieag ^ 280.% 90 »1 a. 2g20 B w / h r 50 IBfiltration V & t w m a 20 X ill % 8 S 22lf,0 Ou0 ft* Beat Ioaa = .018 % 1120 % 90 = 181$ Btu/br 6 o- Total heat Ieaa “ 10*278 Btu/hr .Be Bedrecmi E@0. I :- Ie Walls Aroa = 13 3/% % & + 9 3/% JK 8 * 2 1/3 % 3 5/6 » 2 3/3 a 2 2/3 % 173 aq. ft. Beat less " 173 X .093 % 90 ™ lg% 6 Btu/hr 2.0 Windows , Area s 2 1 / 3 % 3.5/6.*- 2 1/3 % 2 2/3 = 15*2 sq. ft,. Heat less ” 15*2 X. lol3 %. 90 3- l ff5i|-5 Btu/hr ' 3» OoiliBg ' Area = 13 3/%-% 9 Heat 10. m = 1 3 % X 90 x = 13% aq. = l s2:05 Btu/hr -2 7 - k-c W l t M a ~ 13 3/4 ac 9 3/4 % 8 - 1070 Ort0, tto Baat Ieea = ,018 % 90/2 x 1070 - *81 a 1070 - 86? Btu/br 3a W t a l heat, loam & W & 3 Btu/hr 0» Bedroom Ho, 2 I* Walla . Area = 10 # 8 * 3 ^ 2 / 3 2 x 8 » - (2 1/3 x 2 2/3 ) 2 = lS? sqti ft6 Boat loss “ 167 x 90 x e,.093 “ lij-OO Btu/hr 2.» Wlmlows Area - 2 (2 1/3 % 2 2/3) = 12*4 aq* ft* Heat loss ■«, 12,4 x 90 X 1,13 —■126o Btu/hr 3* Ceiliag Area = 10 X 12 3/12 = 124.2 Sq. ft* Boat loss -s 124s2 x 90 X ,1 “•'1120 Btu/hr 4» 'lnflltratiom Volume 3 1 0 x 1 2 ^/12 X 8 - 995 ou, ft, Heut loss ” ,o:81 % 995 ~ 806 Btu/hr 3* W t a l beat loss * 4388 Btu/hr B',. Bathroom 1» Walls Area “ 8. x 5 ^ 2 x 2 2/3 34°7 sq* ft* Hoat loss ” 34®7 x 90 x. ,0415 ” 130 Btu/hr 2* Wlaaew .Area - 2 x 2 2/3 % 3+3 8q* ft* Beat logs s %. 90 % 1*13 % 239 Bto/hr 3* Ooiliag Ai?ea s 2 x T ^ 32 Sq.« ft* Beat Iaee = 3 2 ^ 90 x * l = 312 Btu/hr %.» i&fllta&tioa Valgme = 2 a. ? % 8 = 280 etu ft* Beat Ioae = *81 % .280 a 22? Bta/be 2» Total heat loss « 1211 Btu/ly? ■ E* ICitehoa ' I* Wstil Area = 9& a .8 * 2 1 / 3 ^ 2 . 2/3 = ? 0 a q * ft* B&at lees a 70 ac 90 % *093 = 286 Bto/br 2a Windoxr .x A2*@a. = 2 1/3 % 2 2/3 = 6*2 eg* ft* Heat less = 6*2 x 90 % 1*13 ~ 830 Btu/hn 3*. Ceiling Area * g& x 13*878 = 132 aq* ft* Beat Iesa ” 132 x 90 x *1 « 1187 Btu/hp i|.o Infiltmtien ' Volume s 9& x l 5l87& x 8 = lo22 on* ft* Beat leas = 1023 % *81 = 82# Btavfbr 5.0; Total he at leas = 3%#8 Btn/hr F b, Qfning room I* Walls Arm. = ll;& z 8 .* 9 7/12 x 8 - 2 x k 7/12 * 2 3/% x 2 2/3 = 3b63o2 s<&* ft* Seat loss ~ 163*5 x *093 X 90 “ 1370 Btu/te* Zs Oindeus Area (thozmepcmo) s a = 22*9 aq* ft* Seat .Iosa " 22,09 % j-0 x 90 & 919 Btaa/hr Aroa (alagle glased) 9 2 1/3 % 2 2/3 * 6*2 aq* ft* Seat Ioas % 6*2 a 1*13 % 90 =- 630 B t a / W 3* GolllBg . Area » li|.% x 9 7/12 - I39 sq0 ft* Seat .Ioaa * 139 % 90 at «1 ^ 12$%) Bta/hr k-o X^filtmtlea “ Ilii x 9 7/12 X .8 « 11X 0 ea*. ft* Seat less " o8l x XllO « 900 Bty/br 5* Total heat Ioaa t 508O Btu/hr 0* Sail I* Gelling .. Area “ 7 x 3 ^ 2§ x Ii ~ 39 aq* ft* Hoat loss. % 39 x 90 x *1 “ 391 Btu/hr 2» Total heat loss = 351 Btn/hr H 0 Baaemeht entraBoe I* Wall Area " 8 x 3i « 2 2/3 x 6 2/3 = 10*3 aqe, ft* 3&&at X&8# a 10*3 2. 90 % *1 ^ 93 Btu/hr 2* Boor . Area = 2 2/3 x 6 2/3 = 17*7 aq* ft* HeatlOGa = % 7 * ? X 90 x *51 = 811 B&u/hr 3* Total heat loss ” 90% Bta/ha? ■ *«30**- 1» BasemdiBt I* Walla Jkrea = 2 (36 as 7&) .* 2 (30%. ?) - 10 .(1 z li) s 9I2 Sqa. ft,* Heat leas = (.1) 2 x 30 # 9 1 2 = 9%72 Btn/hr 2 0 F lo e a * Area * 30 % 3 & ? I 080 &%* ft* Boat lea# & 1080 z *1 % 30 c 321*0 Bto/hr 3 o- WlBdiWS Area » 2 (I.% Ii-) 1.0 » 2$ sq*. ft* Seat, loss -. 90 x .23 x. 1*13 “ 23^0 Bta/te XBflXtBatIem Casswae als? ehmige/tir) .Hoat less “■ 081/2 X 7-cS .# 1080 - 3273 Btu/hr Total heat. Ieaa (groood. floor).*».*»**«**,*.6***30*767 Bto/hr Total ilea's, less (lsasoiaeiit}.&0 «.»o-o-*»* *.% ao,-*o;«« »0-Ca-01^.5:327 Btu/iu? fetal heat loss Chease Ja *■0- » . . . » %0 0 «. 0 a * 0 -1 <> 0 n „ * * $ <> 0 i [.3^29^[- Bta/hB Calculations .for Pipe length Assume panel output .to be %0 Btu/hr ** sq, ft. to the ground floor* Tb calculate additional panel, output necessary te heat the basements, do the following,$ Hoeessapy heat/hr - aq* ft* “ Ilt327/1060 ~ 13c/3 Btu/hr ** sq* ft* . ,*& 13 Btu/bp * sq* ft* Total panel output ko -t 13 “ 33 Btu/hp ^ gq. ft* A* hiring Room, (area = 280 sq6 ft*) fin tube-| 6" o$Gj 2 = S 26 ft* Pt* of W b G a B 0 Bedroom Ho* 'I (Area s %3hr sq.0, ft,) §* ta w s 6" o .c * F t* o f t a w = 7 0 7 3 /S J , # 2 = 25# ft* Ge, Bedroom Ho* 2 (area s 12i|, sq0 ft*) ' |w tabes 6” OtiG ti Pt* of tnbe F 6% 6/23 ^ 2 F 23% ft* D* B a t W o o m (area ~ 32 sq0 ft*) &" tubes. 6" OoOci Ftti.of. tube “ d.736/22 x 2 - 63 ft® E et Eitehoa. (area » 132 sq0.ft*) tubes B lt 0 *0* - ' 8/22 # 2 = 197 ft* Ft* of tube F ^ F.C Biaifig' room (area u I39 aq* ft*) i if .Wbos 6 n 0.oG* Ft, of tube F 7162/22 2 2 F 26% ft* Go Hall, (aroa F 39 aq* ft*) tube; 12* 0*0*. . Ft* of W W a c#6/ 2 2 a: 2 * 3% ft* H* Total ft* of tube f 1272 ft* I0 H-Ott of circuits - 1272/180 ft<,/circuit ~ -approx* 9 circuitsBoiler A used, hot water beater using am oil burner was used to sure installation cost and await the arrival of the natural gas pipe lime, to Belgrade* 'This hot water heater has a - 3 2 - capacity of 6o gallons which is too large but due to its large size, it had a low demand and hence a low price. Circulator Use a I?" std. B. & G. Headers Use two W.I. pipes (2") Of particular importance in this design is the floor con­ struction as shown on the drawing. Notice that a sub-floor is laid over the floor joists followed by a layer of water­ proof sisalkraft paper. One-inck by two-inch furring strips were laid on top of the paper on 12 " centers with two tubes between each furring strip. After this was accomplished, a layer of a poor grade of concrete was poured and leveled off k with the tops of the furring strips. Wlien the concrete was thoroughly dry, the hardwood floor was laid. Pig. 2 - Purring strips laid on sub-floor - 33 - Pig. 3 - Floor ready for layer of concrete The object of the concrete, which was actually 25/32" thick, was threefold. First, its purpose was to carry the heat away from the tubes by increasing the emlssivity factor. A shiny copper tube has an emlssivity of about 0.15 as com­ pared to rough concrete of 0 .95 * Secondly, it removed the air space which is a good insulator. Thirdly, the concrete acted as a grill to spread the heat uniformly over the floor. This installation was constructed and put into operation in January 1950 and maintained a room temperature of 70*F. while the outdoor temperature was - 35#F. with 120°F. water. At the present time the unit is heated by an oil burning con­ ventional hot water heater until natural gas is piped into Belgrade, Montana, the site of this residence. The owner, Monte Benson, is, in his own mind, convinced that his heating system is cheaper to operate and has a cheaper installation cost than any hot air system he could have installed. In -3 k - Fige k - Boiler and nearly completed return header addition, his heating plant uses a space about Ip feet square in the basement with no duct work to take up space and no convection currents to carry dust and cause objectional drafts throughout the house. "An added feature", says Mr. Benson, "is a warm bed to crawl into every nightJ" Since the interior is all finished in natural log, it would be extremely undesireable to have large quantities of dust moving through the air and settling on the logs. heat cuts convection down to a minimum. Radiant Interior decorating - i3 expensive. 35 - With no air ducts to darken the walls with the rising air currents, it is obvious that there would be a saving in interior decorating in any radiantly heated home. Also, there are no radiators or convectors to take up space and interfere with the decorating scheme. In this particular case, they would distract severely from the rustic log cabin theme. With the present system, it appears as if the fire­ place is the only source of heat. These are just a few of the many advantages achieved by using radiant panels. Fig. 5 - Interior of Benson Residence The materials used and their cost is as follows: Line voltage room thermostat......................... ^ 11.50 1600 ft. of copper tubing @ ft............... Sisalkraft paper............... .................... 232,00 12.00 Hot water heater.................................... . 150.00 Circulator Lg-" B. & 0. std...... .................. . . 58*00 300 gal. fuel tank with pipe and fittings............ i(.6,00 3 C A-XS* Cu . '5.(3. TJO..Cl V O ^OlX'b V'al'VG 0 ^ o o o u O -J O o o y o o o r/ o D 0 o o o O o o O <> o O \P 3*3.41^0 O ,0, a- e tt O » 0 »■ *. e * a * »-« O B- 6 ».b> « e 0- * B » » Ji1« |6 e * * « & » O- O » * V= Oombination noli of and roduoing Tcilvoy*»*4.***,.****» 2 WfrXo- boadonSf 3 Xto oaob C.2^ OcsBonn *** ^ B-BB u>Bo-b fraoBV BB-OV-^bb 0 ^ 4 S*Q 11*7# 3#00 3 O &o•» b-. 0-»-0.-»b *o-y-0»* »■o'b &b &^ »O-B-B-B # o^O dmr1x3iTig o'ujO-LjO0 00 =>0<>00oo 0y0 =>co0000 0-?000^%0o-^oo =>u^y ^0==:- 20fr00 S a n d C I ^ y a n d j ^ Mx s g q H c m o o n s a v * b - °= - a a -0 B » B - » 0 , CF -b . -b to- Br - a cog 5 ^ o i b o T i s •> 0 1 0 n 0 .-^ «=0 0 a 0 » o « ^ 0 o a y * o u * b tic f r 0 a B B B - V b - B B Br 4 » e a b o q 0 3 -t i O O 1 0 y QO X o t a l fr fr B fr- ■»-B- to,B -Q to 0 B -O- B B. O= 9 O -B- B -B 0' 'B B -B- B- O O B B B B B B--» to'B =O1to -» d B B' O -O til to- tt-V p 3 / 3 = 0 O O Most of the laboa nas dono by Mr* Benaona banco an exact labor cost analysis was not mrailable „. EonoTers assmnlng that two min conld install the ays tern in flto oight-hour days $t #2*#0 por honr-f. the labor coat nonld. be 13200*00? making the total installed cost #793 * 60 . AA estimate for a hot air system to boat the same hones was given as $900»GQ by a local dealer* It* of course3= would be accompanied by a large fur* m e n with duets taking up valuable space and a higher main** tainonco and operating coat with loss comfort end satisfaction*. Xwo different header systems are shown* Xhe short header system was used with all the valves located In the basement* This makes the filling operation and balancing much easier* Xhe "main-type"' system locates all the valves on the ground floor where they may be balanced or shut off* Xhoy are so. ar~ ranged that all valves are located in closets or cupboards* It Is necessary with this system- to rim. the headers through the floor joists which Is; rather undesirable o Xf the build­ ing Is cjuite Iarge3 it Is necessary to use the. itInainu system to keep the circuits from getting too Idng4 system was used ,in the. Benson residence,* The short ,header Sotiee on the draw­ ings that each room has more than one .circuit passing through* This gives extra control for obtaining the- room temperaturedesired* In any design*, each room should have -at least one- circuit of its own* In order to prevent overheating the basement 3 enough re­ sistance to heat flow was placed, between the sub-floor and the .basement celling to get the correct proportion downward 0 drawings follow0 The 38 S > § S to MONTANA STATE COLLEGE BOZEMAN, MONT A N A C O N T R O L S FOR BfNSON RESIDENCE HEATING S Y S T E M 12/6/47 CK. S Y APR Bf DRAWING NO. M E 324 r c:‘K a s s r - Y ' X . g p aras-o % -_ . o : 3 Z 3 -c. x N 7 .-' % I ICAL FLOOR ‘A H rrdw .IGL f l :"R t upper r u t- , 3 xx T v * ^ ,x - GONLrETE " FuR- U STR - - / \Z 'U b F 0 R ----- F.03R J - 15 TF x AT' --- CHfETHO LR CONSTRUCTION b o i U f; a r e a - SU PPLY HEADER RETURN H E A D E B DESIGN NOTES OUTSIDE TEfIR RZDeF. INSIDE TEMP. + 7 0 T WAi.uS 9 " LOOE SAWiO O.N TWO SiUES Al . window s W.T:-: .-W.' except on east - double hung CN LAST - TtjERMQPAWE CJHCLI T S CF '/£ ' Q C F F E R TUiAsN G 5.,^ T - C - - T . VA l C O N S U P 1-LY E N D CF EAC H C i H J U t T S T O P AX; . W A S T £ VALvT ON fit T-r.v t vL CV t-C. C i E C O j J VALVES, HEADERS A N D BOILER LCUATru V BASE M t N T A 1 L TUfiINu ON G "Ce N rEAS E X V - jT E rC-EN CM 3 " ALj HAiu LC- jL C E N TERS LAVER OF SISAi-KRAFT PAPER AeCVt S L BFu CCn ALL I V ' N MuNTAPvA STATE COLLEGE BOZEMAN, MONTANA BAOIANT MEAT,NG FOB MONTE BENSON FFSuOLNCE, BELGRADE, M O N T A N A DR. GY L._HORP-CDARL TP. Sr ---SCALE I '/2 = / FT. CK BY AFP. BY D R A W •'MG NO. M E 3 2 3 * W i B K M d52 S=ZB - IX» 76 .CL . C". >> l > H-V-DV \ FlOOR C h P P E A TUbPfG: ,NCRETE r J R P NG S T p i P S - / X Z. HEADER - Jub :cp i TYP :A_. FLO - C ' TON 8C%E S AREA (RASEM fN T) ) M z nn ; '• S u -Pu.-' . HEAEEP1 i - DESIGN NOTES OuTSjCE T e M R -£0°F INiSfCC TEMP. +TO^F W/__S 9"L0GS S A W E D ON T W C S I D E S A l L / ;C ' EXCEPT '-V £AST — CG1 UBkE .-PNG VV =U G O vVS ON EAST W A U L — T H E R M O P A N E A L - C RCUtTS O F ' / z " C O P 0 E R T U B I N G f" H E A D E R S C U T T H R O U G H J J - S T e A T C E N T E R BOIuER L O C A T E D N B A S E M E N T ALL T J G N S IN A" CENTERS c AC F P T K I T C H E N C M S ' C E t f T f R S . I I IN M ONTANA 5 TAT: COLLE-E BOZEMAN, MONTANA i BAEIANT H E A T i n j FE,R M O N T ' FEHS- V RcSfTfZZCc,BELOBACE,ML \ "ANA :CP.BY :TX 8 / L. HOFPEEAt - L IZ/ I S A fi C K. F Y IA Pr. S> i A SURVEY OF SOME TYPICAL MONTANA RADIANT HEATING INSTALLATIONS Since radiant panel heating is comparatively new, it seemed very worthwhile to make at least a partial survey of such installations in Montana to get personal reactions as well as technical and practical information concerning them. Much was learned from mistakes made and suggestions given by very cooperative and friendly citizens. The following is a group of brief discussions concerning each installation visited. Several others were surveyed but were identical to some already visited, hence left out of this discussion. Plumbers and architects were also contacted, when­ ever possible, for their views on the subject of radiant heat. Several types of construction seem adaptable to radiant heat. The residence of Professor L, 0. Brockman, Bozeman, is a two story duplex using walls of l6M of rammed earth. The lower story is half below and half above the ground, taking ad­ vantage of a hillside beautifully* The concrete basement floor area covers 1)4.50 sq. ft. and uses 6 floor circuits of copper tubing with its own circulating pump controlled by a room ther­ mostat. The second story of the duplex was heated by tubing celling circuits covering the same area. copper The tubes were on 8^M centers and were supplemented by radiant baseboards under the windows on the south wall. the entire length of the wall. These windows extended The ceiling circuits had their —lj.2— own pump which was controlled by a room thermostat. The system was fired by a natural gas 216,000 Btu input boiler located in the basement. Btu/hr. The calculated heat loss of the house was 75*000 Total installed cost of the heating system was about 12/a of the cost of the house. Comments by the Brockmans in­ dicated they would possibly like to install an outdoor thermo­ stat. Also, they had a problem of solar radiation due to the many windows on the south. A great deal of natural circulation was noticed which caused excessive heating. would have prevented that. A flow check valve The fuel bill for this house was $14.1.00 during the winter of 194-9-5>°* Another radiantly heated Bozeman residence is that of the Karl Swingles on Huffine Lane. This is a one story, flat roofed log home covering an area of 2550 sq. ft. and no limit on "picture" windows. At a cost of 10$® of the cost of the home, radiant heat was installed using I" copper tubing on 12" centers except in the living room where it was put on 6" centers. All circuits are in the If." concrete slab floor poured on 8" of crushed rock. The outside perimeter of the slab is insulated with I" of asphalted felt. ed with zonolite. The pumice block foundation is fill­ Each circuit is fed by a supply and return main header system and equipped with a balancing valve. The control system employs a room thermostat to operate the pump in­ termittently while the aquastat governs the flow of natural gas to the 250,000 Btu input boiler. Calculated heat loss of the - 43 - house was 120,000 Btu/hr. Sourdough Creek, south of Bozeman Is the site of a radiantly heated log cabin owned by the Nicholas Helburns. Two floor panels were used involving 1” W.I, pipe on 9" centers buried in a 4 ” concrete slab. Tlie floor slab was poured over a sub-floor of 4 ” concrete using 6 parts zonolite and 4 parts cement. No insulation was put between the floor and the found­ ation, hence an early crop of daffodils occurs each spring! An aquastat sets the oil-fired boiler temperature while the room thermostat turns the burner on and off. runs constantly. The circulator Cost for heating this 575 sq. ft. cabin, runs about #150.00 per year. As for comments, they were sold on floor heating, especially with children. Improvements suggest­ ed were the use of one-half as much piping under beds and run circuits under cupboards and closets too. They mentioned that certain waxes failed to harden on the linoleum due to its heated condition. Installation cost ran about one-eighth of the cost of the house. Another residence using log construction is that of Pro­ fessor Roy Wiegand, Huffine Lane, Bozeman. Ratio of the cost of heating system to cost of the house ran about I to 10. This 1900 sq. ft., one-story home has 10 floor panels which heat it and its basement. The basement, which covers half the floor area, houses the 100,000 Btu input gas-fired boiler and auxilliary equipment. Hardwood floors are used throughout the house -a- with -|-t! copper tubing placed on 8^" centers between the sub­ floor and the hardwood floor. Purring strips are placed be­ tween the floors to support the finished floor. At the time of construction, the tubes were left in air spaces between the floors but due to the low emissivity factor of the shiny copper tubes and the insulating quality of the air space, the house could not be heated although the water temperature was up to 210°P. The water returned from the circuits at practically the same temperature at which it entered. The only solution was to either tear up the hardwood floor and pour concrete or some similar material around the tubes, or take up boards and force the concrete under with a caulking gun. method was used with excellent results. The latter Of course the moisture caused the floors to bulge and warp but the heat brought every­ thing back to normal. A room thermostat controlling the cir­ culator and an aquastat to govern the water temperature com­ pletes the control system. Despite the misfortune of not put­ ting concrete around the circuits during construction, the TLegands are sold 100$ on radiant heating. The average years fuel bill runs about #120.00. The summer of 19^8 found Dr. H. F. Mullikin installing radiant heat in his now one-story Boseman home. Several sources had released information stating that Montana homes could not be heated radiantly by floor panels alone, hence this residence has ceiling panels to furnish the heat and floor panels to keep the concrete slab floor warm. Fourteen celling circuits of copper tubing were used together with 6 floor circuits of the same size. Each circuit was approximately 1^0 ft. long and was valved at each end to short headers located in the boiler room. This method allows all balancing to be done at one lo­ cation and all valves are out of the way from possible tamper­ ing by children. An indoor-outdoor control system is used employing a three-way mixing valve allowing the water to by­ pass the boiler when equilibrium conditions occur. An indoor thermostat operates the circulator intermittently. A day- night clock thermostat, to cut down the temperature at night, doesn’t seem to perform properly, probably due to the heat lag. Domestic hot water is heated by the same gas-fired 210,000 Btu boiler by means of a tankless heater furnishing an unlimited supply of hot water. Comments by Dr. Mullikin indicated he would just as soon have all the heat in the floor to cut down installation costs. Aside from installation cost, the Mullikins are well satisfied with radiant heating. Probably the most unique installation in Montana is that of IIirara Dotson, Rimini Route, Helena. His system requires no boiler because water from a natural hot spring is used, hence the only operating cost is the power to operate the pumps and controls. This one story residence is H O x ![J ft. or 5170 sq. ft. with an outdoor swimming pool which is heated by the dis- Fig. 6 - Natural hot water radiantly heated residence Fig# 7 - "Boiler room" for the above residence charge from the heating system. This eight room home. Pig. 6, with three baths employs nine circuits embedded in a concrete slab floor. The spacing and pipe sizes vary with the room and floor covering. The bedrooms, living, and dining rooms, all heavily carpeted, use l-§" cop ^er tubing on 1$" centers. This set of circuits has its own circulating pump and MinneqpolisHoneywell Modutrol Regulator which employs an outdoor bulb and a three-way mixing valve. The kitchen, utility room and bath­ rooms, combined, have a similar set of controls since they all have linoleum covered floors. The circuits use 3/lj.” copper tubing on 12' centers in these rooms and have their own circu­ lator. Tlae circulators operate intermittently from room thermostats. In addition to the circulators, there is a lift pump which pumps the water from the spring to the house. The Dotsons are extremely satisfied with their heating system al­ though it cost !7 ,800.00 or ll0o of the cost of their house. An example of an industrial application is the IivIH Co., Great Palls, manufacturers of drapes and curtains. Three floor panels of I-V1 7.1. pipe on 12" centers are laid in a concrete slab thick, A 3" main feeds the grid system of pipe layout with balancing valves in each circuit. The single-story building consists of a showroom, a workroom and an apartment, each with its own circuit. Cost of the gas to fire the 350,000 Btu boiler runs between #18.00 and 111}..00 per month. This cost includes the cooking and domestic hot water for the apartment. The building has 3110 sq. ft. of floor space with a 10 ft. celling. The control system is a room thermostat operating the circulator intermittently. Comments from one of the upholster­ ers indicated he had been a previous sufferer of rheumatism but since working on a heated floor, it had left him completely. The cost of this installation was '|l600.00. Another Great Palls industrial application is that of Pinskl Bros. Plumbers who have 10,000 sq. ft. radiantly heated. Their system cost #£>000, or about one-tenth of the building cost. The showroom, covering £,000 sq. ft. employs 6 circuits of I p 1 steel pipe on I8” centers with a 2" circulator, 3" over­ head supply main and a 2^" overhead return main. It operates on 120eP boiler water with a room thermostat to control the cir­ culator. The workshop comprises the remaining £,000 sq. ft. and has its own thermostat and circulator because a lower temp­ erature is desired for manual working conditions. Nine cir­ cuits on l6M centers are used throughout the workshop except for a small section on 12" centers. valve. Each circuit has a balancing During the month of February 19£0, the gas bill to heat the building was #l£.80. The boiler is rated at 9^0,000 Btu (steam) input of 4.8£0 sq. ft. of water. The workmen as well as office girls were very pleased with the heating system. A third Industrial application visited in Great Palls was the Rice Transport, a garage Involving huge doors to let trucks In and out for repairs. This seems to be an ideal application - 49- fov floor panels because the minute the doors are closed, the mechanics lying on the floor are comfortable. Instead of forcing all the hot air to the ceiling, the source of heat is right where it is needed. Garage mechanics don’t mind getting to work on cold mornings with radiant panels in the floor. Fifteen panels using Iijrf1 steel pipe on 1611 centers are buried in the 6" concrete floor. Each circuit has a balancing valve. A room thermostat controls the 3" circulator. A new addition has been built since the first part in 1945. The new section has Its own 2^" circulator with Its own thermostat. Minimum cost for heating this building with gas is >450.00 per year. Only the most favorable comments were received from mechanics as well as office employees. Unit heaters designed and In­ stalled in 1945 to supplement the system have never been used except to blow foul gases out of the garage. Installation cost was 9$ of the building cost. A Great Falls real estate dealer, R. V. Bowker, believes radiant heat is ideal for his offices. His one-story building 7 with four rooms uses I" W.I. pipe on 9" centers In a concrete slab. A room thermostat controls the pump while the aquastat maintains the boiler temperature at 1){.0°F. Cost of heating the 1125 sq. ft. of office space ran about p7•00 during Feb­ ruary 1950. Cost installation ran about $1.00 per sq. ft. of floor space. Archie Parks, near Vaughn, Montana, is rather disappointed —j?0— with his installation. Two copper tubes were placed between each floor joist under hardwood floors in an air space. In­ sulation was literally forced underneath to apparently bring the heat to the floor but to no avail. not emit its heat. The shiny copper would Mot knowing the cause of the trouble, his contractor finally put in radiant baseboards to supplement the over-abundance of insulation and tubing below the floor which, of course, couldn*t function properly. Mr. Parks warns against the use of aluminum casements in windows in Montana. His ex­ perience indicates that they frost and then flood the window sills with water when a Chinook wind appears. The control sys­ tem used was a room thermostat on the circulator. Previous to the installation of radiant baseboards, the boiler temperature was 200°P without attaining comfort conditions. boiler is set at li{j>*P. Mow, the Thus far the operating cost of this oil fired system has been very excessive and installation costs exceeded 10# of the building cost of this one-story, basement­ less residence. The use of zonolite plaster around ceiling imbedded cir­ cuits caused much difficulty in the Herbert Anderson residence of Great Palls. Since zonolite is an insulator, the heat did not conduct properly over the area and heating was difficult except in the basement where the circuits were buried in con­ crete. Two circulators operate the ceiling and floor circuits respectively. Ratio of the cost of the heating system to the cost of the house was about 1/10, Operation costs are about l/3 less than any system used previously. Lack of insulation around the concrete floor slab is probably the main cause of excessive fuel bills in the case of the Lee Amundson residence near Lewistown. throughout the house is another big reason. Lack of insulation One-half inch copper tubing on I^" centers is buried in the concrete floor slab. The circulator is controlled by a room thermostat. Despite fuel bills of from #28.00 to 3^.00 per month, it is difficult to maintain a comfortable room temperature. Ho de­ sign procedure was used in installing the system. Lewistovm has several very satisfactory radiantly heated homes that have been installed by Maynard Stapleton, a local plumber. Most of his jobs have been wall and ceiling install­ ations. One such system is that owned by Joe Matulys who built his home in Ic^t-?. The heating plant ran about of the total cost of the house. building with wood floors. =1,000.00 or l/l6 It is a 5 room, one story There are two inside wall circuits of 3/8” copper tubing on 10" centers and lip outside wall panels of 3/8" tubing on 10" centers. Tine outside walls are insulated with lp" rock wool and a double thickness of paper while the ceiling insulation consists of 3" zonollte and 3" rock wool. The boiler temperature setting is 120°F while maintaining a room temperature of 73*P. The average cost of fuel oil has been ,>63.00 per year despite several 6* x 10* double glass - windows. 52 - The floor area covers 1500 gq. ft. Mr. Mafculya states that there Is no lag problem with h.ia system and that the humidity is much higher in his house than in one using a conventional heating system. In summing up the survey, it was found that the majority of the installations were of the concrete slab floor type with embedded tubes and that Montana homes can, and are being, heat­ ed by floor panels very effectively and efficiently when in­ stalled properly. It is very evident that many residents suffered financially and physically due to guessing at tube sizes, spacing, and most of all, lack of, or insufficient, insulation. The proper use of insulation as to location and quantity cannot be overemphasized in Montana regardless of the type of heating system used. An extra layer may easily pay for itself in less than a year. Tiie use of double glass in windows cuts down widnow heat loss more than half, so, with wise selection of insulation and windows, fuel bills can become a minor rather than a major bill, As was pointed out in the survey, metal window casements are not practical in Montana be­ cause of the rapid heat conduction through them and the frost formations with eventual flooded sills. Cost of installation varied from about one-eighth to onesixteenth of the total building cost. Performance was not always a function of the cost. The controls used, as a rule, were very simple and inex- - 53 - Pig. 8 - Ceiling circuits using 3/8" copper tubing pj~ 9 - Bathroom heated with both wall and ceiling circuits of 3/ 8 " copper tubing pensive although some extremes were prevalent. A room thermo­ stat to control the circulator and an aquastat to govern the boiler water temperature was the system most used. Outdoor controls have been shied away from due to their cost, and, probably even more because of lack of understanding by the plumbers. Fuel bills varied considerably. In some cases they were a minor bill but in several cases quite outrageous. In every case the higher fuel costs could easily be traced to lack of insulation or improper installation or both. In no case could the lack of heat or excessive fuel bill be blamed on the mere fact that radiant heat was used. Such errors as the use of aonolite plaster with buried tubing and the use of s' iny copper tubing in an air space are sometimes costly, but often time it could be prevented if a reliable heating engineer were con­ sulted. There are obviously many applications for radiant heating in Montana. Probably one of interest to chicken ranchers is the application of radiantly heated brooder houses. set-up, the chicks are always on a warm surface. With this The heated floor keeps the litter dry and dries out the droppings. Young chicks are very sensitive to dampness, hence floor panels would be ideal for this purpose. This is just one of the many other possible applications of radiant heat. ~55** INDENTATION CHARACTERISTICS OP FLOOR COVERINGS USED OVER RADIANT FLOOR PANELS Since several inquiries were made concerning the type of floor covering to use with radiant panels in the floors, it was deemed advisable to set up a test panel to see what the indenta­ tion characteristics were In some of the typical floor coverings. A test panel was constructed such that the temperature could be set at any desired value and maintained constant for any desired length of time. A platform with four legs under it was con­ structed to simulate a chair. A metal glider was put under each leg and the platform loaded with a 200 pound weight. made the weight on each glider pounds. This Different sized gliders were used to see what its effect would be on indenta­ tion. The results of the test are shown in Table I. • Although the table tells the complete story of the test, it is interesting to point out a few highlights. First of all, indentation need not be a problem if the proper size and shape gliders are used under furniture. A good example of this is the case of grease-proof asphalt tile which is exceptionally soft under heated conditions. The indentation changed from .016" to e006" by just changing the glider size from to 7/3". The smaller glider was of the 3-prong variety with a more or less hemispherical shape, while the large one was flat with a nail protruding into the platform leg. It is obvious then, that careful consideration should be given to the selection of - 56 - gliders for use under furniture on radiant paneled floors. A few cents extra for proper sized gliders may save hundreds of dollars on floor covering. Secondly, by use of the table, one notices that one must be more careful with glider sizes with some materials. Grease­ proof asphalt tile and extremely dark colored asphalt tile are very subject to indentation whereas rubber tile has extremely small indentation properties. The composition of grease-proof and dark colored asphalt tile can probably be blamed for this undesireable trait. It is possible that the darker color has a slightly higher surface temperature. Since no accurate surface temperature recorder was available, this is merely an assump­ tion. Inlaid linoleum seems quite adaptable to heated floors. As shown in Table I, there was very little indentation at a temperature of 103°?. Since floor panel surface temperatures should never run over 85*?, it is obvious that the danger of indentation with inlaid linoleum is negligible-- with proper sized gliders. The new vinyl plastic coated linoleum also seemed to hold up quite well as far as indentation was con­ cerned. In conclusion, it may be stated that floor covering in­ dentation has possibly been over-emphasized. The table shows that indentations are very small at a floor temperature of 89°?. No radiant paneled floor should exceed a surface temperature of 85*?. The sun shining through a window may easily raise a floor - temperature well over IOO0F. 57 - This shows that despite the type of heating system used, due consideration should be given to selection of floor coverings and appropriate gliders used under all furniture. Floor coverings are not always uniformly con­ structed as is indicated by sample number 8 in Table I. One spot on the 9 ” x 9 " tile gave an indentation of .023" while another point did not show any mark at all. T/AGLE I - /MDENT/mON Or FLOOh OOVEhlNCO ON H HEATED C FLOOM U // UJ £C it CO rr < 3 tn z COVERING FLOOM SIZE GLIDEG // — GLIOEG 5 I Z E - % D E S C R IP T IO N TIME, HOURS T I L E ; G R E A S E - F P 1TOF D A P K GPEEN 48 26 .017 /03 ,016 103 "H E ' ' . 103 26 .0,0 103 20 27 .D C S .O C Z 103 97 48 Z2 .004 • /03 104 24 .002 O IOZ .0 0 4 3-4 ASPP/V..T TiLE; GREASE-PROOF D A R K GflcE A 26 .030 103 20 50 .0 1 0 .0 0 1 ZC3' 89 4 48 .008 103 27 .0 0 3 IO I 26 .OM 103 20 50 .007 103 .0 O 3 89 Z a TYPE A5PMALT ASPHALT 2-4 A S P H A l T 3 AfVD T I l c j GPEEN DARK GKfcSN GPEASE- PPOOFj RUBBErt T U E 7 C R E A M A S P A A l-T TiLE- M A R O O N 4 -4 - ' 7._[ .GFEASE-PRUnF, D A R K GREEN DEPTP1 IN C H TZMi, POURS TEMRj cF - 20 50 DEPTH, INCH .006 .0 0 3 TiMR, °F /03 6? 104 I 4-8 RUBBER TILE; C R E A M 22 ■ .002 104 24 22 . O /03 87 5" RUBBER TlLir ; C R E A M 22 .005 I 04 24 22 .002 .0 /03 si , 6 - RuBBiR TILE; C R E A M A S P H A L T T /CE; MArtDCN -- __ 26 22 '- : <6 9 22 ~2r& 22 TZLi; G R A Y • LE; IO 26— 22 26 -P 22 tl L U.fc.„ 1; V/NYL P -A O T lC ASPHALT LINOLEUM; /3 - ,f /4 I A S P H A L T zr IG I N 1L A l D , E; Y E L L B l UE L I N O L E U M ; INLAID, P R I N T LINOLEUM; IM-AfD, GRAY 'V I' /7 COATED, BASE RED TILE; D A R K LINOLEUM; BROWNj /04 : IO-Z .OlS .013 IOE I .002 24 2 Z v. ■ W-V ... 103 X 6' 23 .005 104 .OOZ as 46" 23 .Ci O O 7 04 88 46 .010 .010 I0 4 69 .006 I OI a? _ - ASPHA-T .004 .023 O I Ol 035 Oi £ /C l IO Z .017 .030 IOi IOZ i 47 2.6 23 4-7 .004 .0 0 1 23 O IC S IOI 68 87 zsZ S . 008 .0 0 8 I03 2T .Cl I .010 I03 25 as- .018 .012 103 IC3 27 4 7 . OOS- 2T 23" .030 103 I03 27 23 47 .020 .004 .OlS IOl .030 24 24 .004 .004 103 103 .002-.002 103 23 .003 23 .0 0 4 IOSIDS • To s I 03 2/ .006 Zl .003 I OI IO N 21 .006 ■ /0/ .0/5 24 Zl IOi 24 . O 69 83 89 IOJ " TiLE P A T T E R N - 59 - literature CONSULTED Adlam, T.N. 19^9. RADIANT HEATING The Industrial Press, New York American Society of Heating and Ventilating Engineers IQkQ Heating Ventilating Air Conditioning Guide IQkQ Waverly Press, Inc., Baltimore Maryland American Society of Heating and Ventilating Engineers April 1950 Page 99 Modern Church Has Radiant Heating, Air Conditioning (Heating, Piping and Air Conditioning) Keeney Publishing Co., Chicago Bell and Gossett. 19^9. THE B AND G HANDBOOK, SECOND EDITION Bell and Gossett Co., Morton Grove, 111. Chase Brass and Copper Co., 19) 4 8 . SUGGESTIONS FOR DESIGNING RADIANT PANEL HEATING Chase Brass and Copper Co., Waterbury, Conn. Heating and Ventilating Sept. 19)1.9 Page 108, Radiant Heating Proves Effective at -50°F in Alaska IIorae Heating Trial. Industrial Press, Now York Shoemaker, R.W. I9I4.8 . RADIANT HEATING McGraw-Hill Book Co., Inc., New York U.S. Dept, of Commerce 19kl INDENTATION CHARACTERISTICS OF FLOOR COVERINGS (Building Materials and Structures Report) BM873 U.S. Government Printing Office, Washington, D. C. LOCATION ME/^TjNG I N 5 T / \ U _ A T m 5 O W M E Z j? TYPE OF JSTRUCTURE BCZfHAN BOZEW V L.O. BROCKMAN /N M O N T A N A OF PANEL type TOTAL HEAT LO S S B T U / lift FUEL USED KARL S1A//NGLE EAR TH-Z STORY LOG BUNGALOW concrete BOZEHMN__ N. W E L B U R N LOG BUNGALOW CONCRETE FLOOR O IL BOZEMAN K. vV.EGANO LOG BU N GA LO W WOOD ON C O N C j r E GAS BOZEMAN H.F, M U L L l K j N WOOD BUNGALOW FLOOR 4 CEILING MBLENA H S . DOTSON WOOD BUNGALOW CONCRETF FLOOM GREAT FAILS GREAT FALLS LM_H CO. __ CONCRETE BLOCK CONCRETE FLOOR PlNSKI B80S. CONCRETE BLOCK CONCRE TE FLOOR GAS GREAT FALLS MICE TRANSPORT BRICK t GREAT FALLS GREAT FALLS floor £ ceiling fl o o r 7TeO p o GAC I Z O 1OOO GAS GS tC QO GAS___ NONE GAS TILE C O N C R E T E FLOOR GAS R.W. BO vvker BMJCK 4 T l L E c o n c r e t e floor GAS H. ANDERSON WOOD BUNGALOi*' CEilimg GAS VAiJG HN A. P A R K S WOOD BUNGALOW WOOD FLOORS O IL LEWSTO'A'N L. A M U N D S O N T I L E BUNGALOW CONCRETE FLOOR LEMS TOWN iJ. MATULYS WOOD BUNGALOW PlASTCM WALLS BELGRADE M . BENSON LOG BUNGALOW WOOD ON CONCfFTf OIL 4 s ; 3 0 0 OIL S T A T E COLLEGE ROZ C M AN, MONTANA MONTANA DR. BY L HORFfOAHL TR1£Y Cu. SY APR BY DRAWING NO. M E 3 2 6 ‘0 9 a/\OMNT — r INDEM TATiON CF FLOOR COVERINGS ON HEATED FLOORS 030 I y/Zl/s'O L C. HOP.PEDAHL ORf i WI NG N O . M E 3 3 8 ,0Z.F GLCOE ,020 I A /0 EN TAT ION, I N C H E S AT /(73 O 0 /5 .010 V uuDtn .005 Hi Tl tu u. 0 3 1 jT Of c Ci I-' < ¥ a. <s Ul > C TC > t- X o <5 CD c£ ir K C u_ S g CU < LU AD CL L? Uv Cl O c? O a: Q- Ul Vl 5 Ce ■f5 U LI o o K 2 Ul o -j > k'' h u. Ui <r V) Ui r Q EC O £ Lu v> < CC <J> Ui 2 U *' ..5_L SAMPLE' D fS C P i P Tf a V > -C Cr U5 O Z z I O O Ce < T A F c 2 Ce 15 F < Ui Cl U O Z CC G 2L EXPANSION TANK BAFFLES WATER HEATER I<ov TEMF y REGULATOR / THERMOMETER W ATER-TIGHT BOX PUMP T H E FLOOR COVERING WAS PLACED ON A '/4 STEEL P L A T E , FASTENED TO THE T O P OF T H E flOX A NO /N CONTACT WITH T H E WATER. NOTE: MONTANA STATE COLLEGE BOZEMAN, MONTANA SKETCH OF E Q UI P M E N T USED I NDENTATION T E S T S DR,BV L.H ORPEDAHL TP. Giy FOR 7/?o/T0 CK- a r a pp. er DKtVG. NO. M E 3 27 % (Sl i d e r . 3 -p r o n g glider STATE COLLEGE BOZEMAN, M O N T A N A MONTANA in . L . HORPEDAHL Q PR, B r Br scale I CO IT G L I D E R S U S E D >N INDENTATION T E S T S A P R . Gy- : F JLL S / Z E D R V G 1. N O . M E 3 3 0 9 4 6 4 4 ^ WONTANA state university libraries I III IIIIlMil I 3 1001 4 5 1 0 9 f N378 H78U cop . 2 H o r p e d a h l, L . C. R a d ia n t h e a t i n g i n s t a l l a t i o n s in M ontana ISSU E D date ----- N37f H784 — C 0/>. 2. 94644 TO