I can't say that I've seen anything on this topic in this group, so I thought
I'd post a quick note.
William Kemp's book 'Te Renewable Energy Handbook' is a great read. In it he
talks about Drainwater Heat Recovery and references a web site. Check out the
link below for more info.
In short, you put a heat exchanger on the shower drain. The cold supply for the
shower picks up transferred heat on its way to the shower. You'll therefore turn
down the 'hot tap' since some of the heat is coming from the cold line.
A friend built a home-made heat exchanger. Checking with a laser thermometer,
his heat recovery rings in at about 30%. Or put another way, would you rather
buy gas at $.90 or $.60 ?
A simple system might preheat fresh cold water with outgoing greywater
and run it into the cold inlet of a conventional tank water heater.
The cost needs to be consistent with the savings, say $00-500/year.
Rerouting plumbing to separate grey and black water could be expensive.
And if the height difference between the greywater inlet and the sewer
outlet is too small, a GFX won't work without a pump. And a GFX doesn't
store greywater, so it won't help much with a bathtub.
Backflush into the sewer line, which might be automatically indicated when
greywater starts to flow through an overflow bypass? A filter could also
help, eg a nylon stocking inside a vertical 4" PVC pipe.
There's lots of room for improvement. Very efficient heat recovery could
make solar hot water obsolete, if it only tops up 100 F water to 110.
Setting the water heater to the shower temp and fully-enclosing a shower
can also help. See
Here's a simulation with 7 $1 100'x1/2" pipes inside a $3 100'x3" 70 psi
layflat ag irrigation pipe, which might hang in an 8' flat spiral under
a basement ceiling. The inner pipes might be tied together in a hexagonal
cross section, with 6 pipes kissing a central pipe.
20 DIM IPT(100),OPT(100)'pipe temps
30 L0'pipe length (feet)
40 IPID=.622'inner pipe ID (inches)
50 IPOD=.742'inner pipe OD (inches)
60 IPAD=(IPID+IPOD)/2'average pipe diam (inches)
70 A=PI*IPAD/12*L'pipe area (ft^2)
80 IPVOL=L*PI*(IPID/24)^2'inner pipe volume (ft^3)
90 IPCAP=IPVOL*62.33/8.33'inner pipe capacity (gallons)
100 IPDIS=L*PI*(IPOD/24)^2*62.33/8.33'inner pipe displacement (gallons)
110 OPCAP=L*PI*(3/24)^2*62.33/8.33'3"outer pipe capacity (gallons)
120 U0'pipe conductance (Btu/h-F-ft^2)
130 CMIN*8.33'heat capacity rate (Btu/h-F)
140 N=7'number of inner pipes
160 EFF=NTU/(NTU+1)'counterflow hx effectiveness
170 IPCAP=N*IPCAP'total inner pipe cap (gallons)
180 OPCAP=OPCAP-N*IPDIS'net outer pipe capacity (gallons)
190 PRINT N,EFF,OPCAP,IPCAP
200 GRATE=3'greywater flow rate (gpm)
210 GSIZE'greywater burst size (gallons)
220 NPSP'#pipe segments
230 DL=L/NPS'pipe segment length (feet)
240 UAS=N*U*A/NPS'pipe segment conductance (Btu/h-F)
250 DIC=8.33*DL*IPCAP/L'inner pipe segment cap (Btu/F)
260 DOC=8.33*DL*OPCAP/L'outer pipe segment cap (Btu/F)
270 GTIME=GSIZE/GRATE/60'burst time (hours)
280 DT /3600'time step (h)
290 GAVG=DT*8.33*GSIZE'average gw flow rate (lb/h)
300 DGC=8.33*DT*GSIZE'greywater cap step (Btu/F)
310 IPT(NPS+1)`'freshwater inlet temp (F)
320 OPT(0)0'greywater inlet temp (F)
330 FOR BURST=0 TO 20'greywater bursts
340 FOR T=0 TO 1-DT+.000001 STEP DT'simulate 1 hour
350 IF T<GTIME GOTO 390'skip burst fill
360 FOR S=1 TO NPS'pipe segment (1<->gw in)
370 OPT(S)=(DGC*OPT(S-1)+(DOC-DGC)*OPT(S))/DOC'gw seg temps after move (F)
380 NEXT S
390 FOR S=NPS TO 1 STEP -1'pipe segments (1<->gw in)
400 IPT(S)=(GAVG*IPT(S+1)+(DIC-GAVG)*IPT(S))/DIC'fw seg temps after move (F)
410 NEXT S
420 FOR S=1 TO NPS'pipe segment (1<->gw in)
430 HEATFLOW=DT*(OPT(S)-IPT(S))*UAS'heatflow (Btu)
440 OPT(S)=OPT(S)-HEATFLOW/DOC'gw temps after heatflow (F)
450 IPT(S)=IPT(S)+HEATFLOW/DIC'fw temps after heatflow (F)
460 NEXT S
470 IF BURST THEN EOUT=EOUT+GAVG*(IPT(1)-IPT(51))'accumulate gain (Btu)
480 NEXT T
490 NEXT BURST
500 FOR I=0 TO 9'freshwater temps
510 FOR J=1 TO 4
520 PRINT TAB(15*(J-1));IPT(5*I+J);
530 NEXT J
540 PRINT TAB(60);IPT(5*I+5)
550 NEXT I
560 EIN=8.33*GSIZE*(OPT(0)-IPT(NPS+1))'total water heating energy (Btu)
570 EFF=EOUT/EIN'simulated effectiveness
580 PRINT EFF
# inner pipes theo eff outer cap inner cap
7 .9782665 21.00167 11.05245
The theoretical effectiveness is 98%, if we store 10 gallons of greywater in
one reservoir and store 10 gallons of freshwater in another and slowly feed
them into a counterflow heat exchanger and feed the output into another
reservoir. The inner and outer pipes hold 11 and 21 gallons.
fw outlet (F)
93.78576 91.04849 88.49059 86.10591 83.88806
81.83024 79.92535 78.16596 76.54438 75.05275
73.68311 72.42754 71.27815 70.2273 69.26755
68.39181 67.59336 66.86586 66.20341 65.6005
65.05206 64.5534 64.10021 63.68856 63.31481
62.97566 62.66806 62.38924 62.13663 61.9079
61.70092 61.51371 61.34447 61.19156 61.05344
60.92875 60.81619 60.71462 60.62297 60.54024
60.46555 60.39807 60.33703 60.28177 60.23164
60.18606 60.14449 60.10645 60.07149 60.03921 fw inlet (F)
simulated effectiveness: 0.8250335
The simulated effectiveness is a lot lower (it was 0.646 with 4 pipes), but
it looks like the last 50' vs the 100' pipes don't do much, with gw and fw
temps both close to 60 F. So we might not lose much if we used a 50 vs 100'
length. We'd still have a 10.5 gallon gw capacity, consistent with DOE and
SRCC test protocols.
Perhaps we can do better. I got 91% in a previous simulation. This time,
I fed 10 gallons/hour of fresh water through the greywater at a constant
rate, back into the cold side of a tank water heater. Last time, a diff
thermostat ran a pump whenever the gw inlet temp was greater than the fw
outlet temp by more than a predetermined amount. I'll try that again.
Got any more promising schemes?
PS: There's also an ICC plumbing code issue, which probably needs
a hose bibb vacuum breaker and some PE probability calcs to fix.