Posted by daestrom on April 22, 2006, 3:14 pm
Careful, in the control you need to consider if the greywater inlet
temperature is lower than the tank temperature. Just looking at coil
temperature could result in some adverse operation. Maybe not in your
specific circumstance, but in some common situations. For example, for me
in the winter, the coil (located in the basement) is cooler than the living
space. So flushing the toilet with 70F water from its tank, while the coil
is 55F, would start the pump and circulate 140F water from the tank to the
coil. And that would actually end up *heating* the greywater.
So the controls may need to be a bit 'smarter' and also consider the
hot-water tank's drain temperature. Just so you don't start the pump when
the tank bottom drain is warmer than the greywater.
Well this is certainly true. Energy prices certainly seem to be rising
faster than general inflation, and that means payback on energy conserving
devices of all sorts have better payback.
If we go with the GFX-star setup, and assume we are not drawing off any
water (hot or cold) while dumping this boiling hot water down the drain, we
could *assume* the pump start will circulate about 4 gpm. But how fast does
the water flow when you dump a large pot of pasta through a collander into
the sink drain?
This is the sort of question we could spend quite a long time trying to
figure out. After all, the sink drain is probably a 1 1/2 pipe, that leads
some distance to the main drain. There is perhaps 1 lbm of cold water
sitting in the P-trap under the sink. Your home may have a 'master trap'
that contains another 4 lbm or so of water. Your sink could be metal. The
piping from the house to the tank are probably 70F (same general ground
temperature as your water supply). The factors are many and various.
But looking at the GFX alone, a G4-60 model has about 110 turns of 1/2" Cu
pipe with a mean turn diameter of 4 1/2 inches. That holds about 11 lbm of
freshwater. And the whole thing weighs in at >40 lbm of Cu when dry. So I
don't think you have much to worry about. Take a *large* pot of boiling
water holding 2 1/2 gallons (20.8 lbm) of 212F water. Mix that with 1 lbm
P-trap water at 70F, and 11 lbm of 70F water in the GFX, and 40 lbm of Cu
(equivalent capacitance as 40 lbm *(0.90 / 1.00) = 36 lbm) at 70F and you
get a mixture at...
(11lbm*70F + 1lbm*70F + 36lbm*70F + 20.8lbm*212F) / (11lbm + 1lbm + 36lbm +
20.8lbm) = 112.9F
Now, the 'pasta water' isn't in contact with the GFX long enough to reach
equilibrium with it, but it will lose a lot of its energy just flowing
through a GFX style heat-exchanger. Combine this with the other factors I
mentioned, and I don't think your septic tank will receive any water over
120F. And considering the volume of a septic tank versus the 2 1/2 gallons
of 'no-longer-boiling-hot' water, I wouldn't be concerned unless you making
pasta for an army and dumping pots of water down the drain every ten
Posted by daestrom on April 19, 2006, 9:58 pm
What working fluid would that be? Water has a much higher heat capacity
than most any other fluid you're likely to find/afford. Best to stick with
As regards to increasing fresh-water flow, the GFX folks recommend plumbing
so that the cold water enters the heat-exchanger *before* splitting to go to
the hot-water heater and the cold tap in the shower. So the 'cold' water
side of the shower is pre-heated (need less 'hot' water to stay comfortable)
as well as the inlet to the hot-water heater. This raises the flow on the
fresh-water side to equal the greywater flow.
Putting a pump to circulate between the 'dip tube' into the heater, and the
drain could improve the circulation through the fresh-water side of the gfx.
But you best be sure to insulate the piping. But with this setup would you
still want the heat-exchanger's fresh-water outlet going to the cold shower
I'm a bit skeptical. After all, the greywater coming in is cooler than the
hot-water heater until you've cooled the bottom of the tank. But the bottom
of the tank is receiving fresh-water out of the hx. Increasing the flow
through the hx with a pump is going to put warm water into the hx inlet,
mixing with the cold water from the main. Then sending the mixture of cold
inlet water, and warm water from the heater drain into the hx. Just so you
can put warmer water back into the bottom of the heater via the 'dip tube'.
Net results seem to be higher flow, but the average temperature of the
fresh-water side of the hx is higher and the greywater outlet temperature
will be higher. I would wonder if the increase in heat transfer due to
higher flow on one side only can make up for the lower temperature
difference across the heat-exchanger walls.
Posted by Robert Gammon on April 19, 2006, 10:20 pm
agreed that water is best, but think of Freon as an alternative.
Yes, and this works well in heating dominated climates where ground
water temps average less than 50F year round.
But not here with our ground water already in the high 60s to mid 70s (I
measured mine at 75F)
Yes, as that tempers the water to a reasonable temp. My water will come
from a well, and I need to set my hot water tank to at least 140F to
Foam insulation around piping and the heat exchanger seem to be useful
ideas to conserve energy as does a water heat blanket for the storage tank.
The flow is from the drain plug at the LOW point of the storage tank
thru the pump to mix with fresh water. Then thru the coil, and back in
to the lower third of the storage tank. There will be mixing that
occurs, however, this storage tank should achieve water temps in the
70-90F range depending on entering fresh water temp.
The idea is shown on the web site best under the solar hot water
retrofit. The application data shows efficiency rising as flow rate
thru the coil increases for a given flow rate down the drain.
Grey water temps may NOT increase at all due to the higher flow rate
(same idea as the heat exchanger itself). This higher flow rate comes
at a price of increased pressure loss and expense.
Coil inlet temp will be higher, but then so will coil outlet temp.
Posted by Ecnerwal on April 16, 2006, 1:18 pm
Connecting a GFX to a toilet drain is not going to be very sensible,
however, unless you heat your toilet water - makes the most sense when
connected to drains that might have hot water - putting it in the
blackwater stack just adds one more way to have a toilet flush make your
shower uncomfortable, and reduces the potential efficiency by reducing
the temperature differential across the heat exchanger.
Cats, coffee, chocolate...vices to live by
Posted by Robert Gammon on April 16, 2006, 1:50 pm
Yes, there is no heat to recapture, UNLESS toliet supply is also hooked
to GFX heat exchanger.
The point to adding in the toilets is to make installation simple.
Locate the SINGLE sewer pipe in the basement that collects ALL waste
water, and insert the GFX into the pipe.
Running toilet water to the mix of effluents processed by the system
adds NOTHING to efficiency, but makes installation a BREEZE.