re: diy dehumidifier SOLAR.KnowledgePublications.com

re: diy dehumidifier
19 may 2005
stretch wrote:

>...a calculator or computer will not measure wet bulb or relative humidity,

a computer with sensors can. check out the u mass team house at the 2005
doe dc decathlon contest in october. 

>but once the data is collected could convert it with the proper software.

radio shack sells "data collectors." here's some "proper software" for an $8
casio fx-260 calculator: ventilate when ro < rie^(9621(1/(460+to)-1/(460+ti)),
ie ro < rie^(9621(1/(460+to)-1/530) if it's 70 f indoors. for example, if it's
70 f with 80% rh indoors and 80 f outdoors, ventilate when ro < 0.8e^-0.336
= 0.57, ie the outdoor rh on the $20 radio shack display is less than 57%.

>...most people need someone with a strong hvac background to determine what
>they need.  even most a/c contractors don't use psychrometrics enough
>to figure it out.

i've noticed that few a/c contractors have strong hvac backgrounds.

>a properly controlled exhaust fan can dehumidify with 100x less energy
>than a dehumidifier.

>...that depends on the outside ambient.  if it is very humid outside,
>running a fan will make it worse. what you say may be true in massachusetts,
>but here in south carolina, a residential exhaust fan does more harm than
>good about 8-9 months per year.

raleigh looks fairly humid, but a properly controlled exhaust fan might help
from october (w = 0.0081) through may (w = 0.0099). july is dampest, with
w = 0.0149 on a 78.1 f average day with 68.1 and 88.0 daily min and max. an
"airtight" house with 15 cfm of natural air leakage (vs. 2.5 cfm in canada)
and w = 0.0120 indoors would need about 24hx60x15x0.075(0.0149-0.0120) = 4.7
lb/day of dehumidification from air leakage plus about 2 gallons per day from
human activities (andersen's estimate for a family of 4), about 21 lb/day.

on an average july day, 1920 btu/ft^2 falls on the ground and 750 falls on
a south wall in raleigh. we might have an epdm rubber liner with a passive
greenhouse-type solar still with shallow licl lakes separated by dry epdm
beds to act as water collectors and parasitic air heaters, like this,
viewed in a fixed font like courier:

                   |           2'          |
                 carbo
             poly  s  nate 
        flat       p       clear
  clear            a            flat
                   c                 poly      nate...
                   e                     carbo
         epdm licl r lake epdm  heater  dry bed            epdm
    epdm 2x4 epdmepdmepdm 2x4 epdmepdmepdmepdmepdmepdmepdm 2x4 epdm
--------------------------------------------- top of sip ------

(what's a good lake to heater area ratio?)

how many square feet of 80 f licl solution (precooked to 160 f) are needed
to remove 20 pounds of water from 80 f house air with w = 0.012 in 12 hours? 
here's a 9-pound calc for miami, based on some crude assumptions:

1) the licl still operates at a constant temp for 12 hours per day.
2) the solar energy that enters the r1 glazing with 90% transmission
equals the sensible and latent heat energy needed for concentration.
3) the solution cools to 25 c at night. 
4) the solution gains heat like an ashrae pool loses heat. 

the next step might be a simple tmy2 simulation. 

10 a1=12.7409'licl vapor pressure constants from the 1993 hawlader paper
20 a2=-.065536
30 a3=-8.2416e-04
40 b1=-4675.4
50 b2=+29.31
60 b3=+.66911
70 c1=372690!
80 c2=-1689.8
90 c3=-187.1
100 ta=82.8'average ambient august temperature in miami (f)
110 sg=1770'average august sun on ground in miami (btu/ft^2-day)
120 h=12'distillation day length (hours)
130 w=.0176'average ambient august humidity ratio in miami
140 pv=25.4*29.921/(1+.62198/w)'ambient vapor pressure (mmhg)
150 p=9'dehumidification load (lb h2o/day)
160 for tc=60 to 90 step 10'solution temp (c)
170 tk=273.1+tc'solution temp (k)
180 c=a1+b1/tk+c1/tk^2-log(pv)/log(10)
190 b=a2+b2/tk+c2/tk^2
200 a=a3+b3/tk+c3/tk^2
210 conc=(-b-sqr(b^2-4*a*c)/(2*a))'equilibrium soln conc (wt%)
220 tf=1.8*tc+32'solution temp (f)
230 concsurf=1000*p/(.9*sg-h*(tf-ta))'licl surf needed for conc (ft^2)
240 tk=298.1'solution temp (25 c)
250 ap=a1+a2*conc+a3*conc^2
260 bp=b1+b2*conc+b3*conc^2
270 cp=c1+c2*conc+c3*conc^2
280 pvc=10^(ap+bp/tk+cp/(tk^2))'vapor pressure at 25 c (mmhg)
290 pvi=29.921/(1+.62198/.012)'indoor vapor pressure ("hg)
300 pvl=pvc/25.4'licl vapor pressure ("hg)
310 dryrate=.1*(pvi-pvl)'lb/h/ft^2 h2o (like an ashrae pool)
320 drysurf=p/(12*dryrate)'licl surface needed to dry p lb h2o in 12 h (ft^2)
330 print tc,conc,pvc,drysurf,concsurf
340 next

still         solution      licl pv       drying        concentrating
temp (c)      conc (wt%)    (mmhg)        surf (ft^2)   surf (ft^2)

60            39.15389      5.493444      21.42437      9.927201
70            45.89019      2.75522       16.3801       13.03215
80            52.33653      1.244954      14.49746      18.96333
90            58.57794      .5091767      13.72873      34.80277

if the still temp is too low, it looks like we need lots of drying surface.
if it's too high, we need lots of concentrating surface.
a 70 or 80 c still temp seems good...

nick