DHW heater for college Rec Fitness center
Steve Shantz
Solar DHW system to heat water for the showers in our Rec-Fitness
center. The State of Indiana has a $25,000 matching grant program for
this type of project, but it may not be used to heat pools. To the
best of my knowledge, they have not yet given out a grant yet because
nobody has built a system yet.
The facility is used year round. We have metered our DHW consumption,
and typically use about 1200 gallons per day (gpd), with a maximum of
about 1500 gpd. The maximum sustained load is about 6 gpm for a
maximum time of about 20 minutes.
About a year ago we had 3 'contractors' spec out their systems, and
they were basically scaled up versions of residential systems, and
outrageously priced. They tried to store 500 - 1000 gallons of hot
pressurized water, and we found out that commercial tanks of that size
cost about $10 / gallon. This was one of the things that killed the
initial 3 designs.
So I stewed on this project for a long time and came up with a design
that I think would work really well. I would appreciate a critique
please...
My design is a drain back system. Arrays of collectors (24 to 30 4' x
8' or 4' x 10') would be mounted on the roof of the pool. The roof
happens to be pitched to the west just slightly, which would
facilitate complete drainage of the collectors in shutdown mode.
Piping runs would need to be pitched appropriately for complete
drainage. The roof is standing seam construction. Any comments on
how to fasten the arrays down so they don't become expensive kites
would be appreciated. There are heavy steel beams supporting the
roof.
I plan on building a cistern in the ground right beside the building,
well insulated, and containing somewhere between 3000 and 5000 gallons
of water. The cistern would be 3 - 4 times deeper than the diameter to
help with stratification. We would pump from the bottom of the
cistern, through the collectors, and return the heated water to the
top of the cistern with some type of baffle to prevent mixing and to
preserve stratification in the cistern. This heating loop would have
either an air gap where the water returns to the cistern or suction
relief valves at the top of the collectors so that water can drain
completely when the pump stops. No shutoff valves are allowed in the
collector piping loop. Obviously the drain back system must work
robustly with no failure modes due to accidentally closed valves.
Cold water entering the existing heating system would first go through
a 120 gallon insulated storage tank before going into the existing
heating loop. Water from the bottom of the 120 gallon storage tank
would be pumped by a very small pump though a brazed plate HX,
returning to the top of the tank. Hot water from the top of the
cistern would be pumped through the other side of the HX, returning to
the bottom of the cistern. The 120 gallon tank becomes an inexpensive
buffer tank to meet the peak flow requirements. The pumps on the HX
could be switched on and off as required to meet demand requirements.
The system would be monitored and controlled by our existing energy
management system, using a VFD on the collector loop to minimize
energy consumption and maximize thermal energy collection. Our energy
management system has an outdoor light meter which can be used to
determine when the sun is shining so that pump start up and shutdown
is automatic. Or, would it be better to put a temperature probe at
the top of a collector and start the system when it gets warm?
The overall design is very simple, and rather inexpensive to build.
Critical review of the design would be appreciated.
Thank you for your comments,
Steve
Malcolm "Mal" Reynolds
1f69f9...@r36g2000prf.googlegroups.com:
You haven't mentioned how much it currently costs to supply the hot water.
Does your college have a central plant providing hot water to any
buildings on campus?
With 6gpm as your highest load, I'd be inclined to suggest one large or
two medium tankless water heaters.
Robert Scott
You have obviously put a lot of thought into this design, and it goods basically
sound to me. Using a HX and a separate storage/collection loop from the DHW is
good for more reasons than just avoiding the cost of a large pressurized storage
vessel. It is also easier on the collectors because you can reuse the same
water over and over and avoid infusion of new dissolved minerals and corrosive
agents that attack collector metals. But check with your local codes. You may
need a double HX to keep the collector water from entering the DHW in case of a
single-point HX leak. I know it would mandatory if the collector water were
poisonous antifreeze. You cannot rely on the fact that the DHW is pressurized
and the collector fluid is not. There might be a pressure failure. But I'm
not sure if it is required if the collector fluid is just plain water that gets
reused. I assume that the code requirements for shower water is the same as for
potable drinking water. Maybe your brazed plate HX is intrinsically safe
anyway.
Is it possible to integrate the HX into the cistern? That would save the
expense of the circulating pumps and the 120 gallon tank. The incoming cold
water would go through the cistern (from bottom to top) as a pre-heater to your
existing system.
As for collector pump controls, temperature is a much better indicator than
light level in deciding when to turn them on.
Robert Scott
Ypsilanti, Michigan
daestrom
> I work for a small college in Indiana, and we have been considering a
> Solar DHW system to heat water for the showers in our Rec-Fitness
> center. The State of Indiana has a $25,000 matching grant program for
> this type of project, but it may not be used to heat pools. To the
> best of my knowledge, they have not yet given out a grant yet because
> nobody has built a system yet.
>
Steve, you've obviously been thinking about this for a while now. But have
you given any thought to what happens to that hot water when it runs down
the drain? I don't know the specifics of your set up, but if you have
access to the drain water, and it has a *vertical* drop somewhere underneath
the showers, you might look into GFX technology.
It provides a high heat recovery rate but requires a *vertical*
installation. It is ideal for situations where the water runs in/out
continuously such as a shower.
I installed one in my home and have measured it's performance several times
and found it to recover quite a bit of heat.
Recovering some of this heat before it just goes 'down the drain' can save
money and reduce the size of DHW heating plant.
daestrom
http://www.gfxtechnology.com/
P.S. I have no affiliation with them whatsoever, just a satisfied user.
Steve Shantz
> you given any thought to what happens to that hot water when it runs down
> the drain? I don't know the specifics of your set up, but if you have
> access to the drain water, and it has a *vertical* drop somewhere underneath
> the showers, you might look into GFX technology.
daestrom,
I've looked at the GFX technology quite closely. The problem in this
application is that the showers are on the ground floor, with no
basement or access to the drain system from below. I have wondered
about this for some of the other dorms and for the dishwasher in the
cafeteria. In both cases, I don't have quite enough height, and the
waste water would need to be pumped back up to the level of the sewer
lines.
Steve
Steve Shantz
> Is it possible to integrate the HX into the cistern? That would save the
> expense of the circulating pumps and the 120 gallon tank. The incoming cold
> water would go through the cistern (from bottom to top) as a pre-heater to your
> existing system.
Part of the problem is that there is about 50' of run between the
closest place the cistern can be located and the inlet to the existing
DHW system. In low flow rates we may loose a substantial amount of
heat before it gets to the existing system. However, insulation is
cheap compared to the cost of the 120 gallon tank, pumps and
controls. Definitely worth thinking about. Even if we don't get
complete heating at the highest flow rate... so what. My objective
isn't to always heat all of the water, but rather to heat as much as
possible at the lowest cost.
However, I have submitted a proposal to do something like this, but
everybody else I'm working with thinks I'm crazy. What I have in mind
is a 2' x 2' air to water HX, like those used to heat furnace air when
using an outdoor wood fired water heater. I would submerge the HX
just below the top of the cistern, laying flat (horizontally). As
cold water moves through the internal tubing, it would be heated, with
the cistern water falling through the aluminum fined tubing as it is
cooled, and naturally falling to the bottom of the cistern. This cold
water convection could even be aided somewhat by building a small
chimney for it to fall down in. Several units could be stacked one on
top of the other for even better performance.
I have bought one HX, and I'm hoping to test it out this winter using
our central boiler blowdown tank as the source of heat. I'm not sure
how long a HX such as this would hold up to these operating
conditions. This configuration would be intrinsically save (I think),
as any leak in the HX would result in the water leaking into the
cistern, which can NEVER be pressurized.
Additionally, given that the cistern is in the ground, and the 120
gallon preheating storage tank is on the second floor, I've wondered
if this DHW preheat circulation loop could run without a pump, being
powered by the different densities of the cold water going down to the
cistern vs the hot water floating back up. The pressure gradient is
small, especially in marginal heating situations, so I'm not sure I
would get enough flow. I'll have to see if I can run the number on
flow rates vs pipe size, number of fittings, height, and temperature
delta.
Thanks for the feedback!
Steve
David Williams
-> application is that the showers are on the ground floor, with no
-> basement or access to the drain system from below. I have wondered
-> about this for some of the other dorms and for the dishwasher in the
-> cafeteria. In both cases, I don't have quite enough height, and the
-> waste water would need to be pumped back up to the level of the sewer
-> lines.
-> Steve
I have a home-brewed heat exchanger in my shower drain. It warms the
cold water that is destined for the shower's "cold" tap and for the
water heater. Basically, it consists of two copper pipes, one
concentrically inside the other. The drain water flows in the central
pipe, and the incoming water goes in the opposite direction in the
annular space between the pipes. The thing is about 3 metres long, and
is installed horizontally under the bathroom floor. It's not
super-efficient, but it does recycle some heat. I did some measurements
while I could easily get to it. (Now, with the floor down, it's pretty
inaccessible.) The drain water started at about 40C and was cooled to
35C, while the incoming water was warmed from 5C to 10C. That's a lot
better than nothing! I reckon that in the 15 years or so I've had it,
it had paid for itself many times over.
However, it doesn't conform to building codes in some jurisdictions.
Some places do not allow a single copper wall to separate drain water
from incoming water.
dow
Morris Dovey
> I work for a small college in Indiana, and we have been considering a
> Solar DHW system to heat water for the showers in our Rec-Fitness
> center.
I've been trying to visualize how the building is laid out. It looked as
if the pool area might be at the west end of the building. Is that
correct and if so, where are the showers relative to that? (I can't
really tell much from the satellite view or the campus tour views.)
--
Morris Dovey
DeSoto Solar
DeSoto, Iowa USA
http://www.iedu.com/DeSoto/
dhruv
Consider using 4 10 ft satellite dishes with solar tracking. The ones
you use for CU band stations. 4 of these will
be able to heat 1500 gallons of water a day. The good part is you get
maximum temperature in about 2-3 minutes.
So even if you have 2-3 hours of clear sunlight you will be able to
heat the water .
The dishes can be installed with the help of students. It should cost
you approx $4000 per dish if you put the labour yourself. You could
start with 2 and then add more.
Let me know if you want more details. I have a working prototype for a
6ft dish in MA. If you put
solar dish collector as the key word in you tube you can see my
video. My user name is mdhruv.
Dhruv
Robert Scott
>Consider using 4 10 ft satellite dishes with solar tracking. The ones
>you use for CU band stations. 4 of these will
>be able to heat 1500 gallons of water a day. The good part is you get
>maximum temperature in about 2-3 minutes.
>So even if you have 2-3 hours of clear sunlight you will be able to
>heat the water .
>
>The dishes can be installed with the help of students. It should cost
>you approx $4000 per dish if you put the labour yourself. You could
>start with 2 and then add more.
Concentrating collectors are great for generating very high temperatures, but
are not very cost-effective for generating typical DHW temperatures, especially
in the midwest where diffuse light though some cloud cover is so common. In
that case a fixed flat-plate collector will give you much more heat per dollar
invested. If they are low-mass, they will also be generating heat in 2-3
minutes.
Robert Scott
Ypsilanti, Michigan
Steve Shantz
> I've been trying to visualize how the building is laid out. It looked as
> if the pool area might be at the west end of the building. Is that
> correct and if so, where are the showers relative to that? (I can't
> really tell much from the satellite view or the campus tour views.)
For those of you who have looked up Goshen College on Google Earth,
the Rec Fitness Center is on the south end of the campus, directly to
the east of the railroad tracks and north of the baseball fields. The
pool is the north-west segment of the building, facing north and west.
The south end of the pool roof is mechanical space, above which we
would mount the collectors. The roof is standing seam, pitched just
slightly to the west for drainage, which is just perfect for draining
the collector manifolds in a drainback system.
My biggest questions right now are how to mount the arrays to the
roof. Our facilities people don't want the collectors and a whole
bunch of roofing to turn into an expensive kite when the first big
wind storm comes up. I've seen the aluminum blocks that clamp to the
seams, but what would prevent the whole section of roof from being
pulled up in a high wind? One option is to install some beams above
the roof, with penetrations through the roof, and welded or bolted
onto the large supporting beams directly under the roof. Another
option is beams on the roof clamped to the seams, but extending out
about 10 - 15 feet north and south of the collectors. Maybe with a few
bags of sand on the ends to prevent them from lifting.
How about the back supports? 1 1/4" or 1 1/2" square aluminum or steel
tube?
Does anybody know who can design such a structure? How have other
people fastened large arrays to roofs? Something tells me this might
be a job for engineer$$$$.
Any help here would be appreciated.
Steve
Morris Dovey
>> I've been trying to visualize how the building is laid out. It looked as
>> if the pool area might be at the west end of the building. Is that
>> correct and if so, where are the showers relative to that? (I can't
>> really tell much from the satellite view or the campus tour views.)
>
> For those of you who have looked up Goshen College on Google Earth,
> the Rec Fitness Center is on the south end of the campus, directly to
> the east of the railroad tracks and north of the baseball fields. The
> pool is the north-west segment of the building, facing north and west.
> The south end of the pool roof is mechanical space, above which we
> would mount the collectors. The roof is standing seam, pitched just
> slightly to the west for drainage, which is just perfect for draining
> the collector manifolds in a drainback system.
Good info - and it looks as if what you want to do has been well
thought-out and should be workable.
> My biggest questions right now are how to mount the arrays to the
> roof. Our facilities people don't want the collectors and a whole
> bunch of roofing to turn into an expensive kite when the first big
> wind storm comes up. I've seen the aluminum blocks that clamp to the
> seams, but what would prevent the whole section of roof from being
> pulled up in a high wind? One option is to install some beams above
> the roof, with penetrations through the roof, and welded or bolted
> onto the large supporting beams directly under the roof. Another
> option is beams on the roof clamped to the seams, but extending out
> about 10 - 15 feet north and south of the collectors. Maybe with a few
> bags of sand on the ends to prevent them from lifting.
>
> How about the back supports? 1 1/4" or 1 1/2" square aluminum or steel
> tube?
>
> Does anybody know who can design such a structure? How have other
> people fastened large arrays to roofs? Something tells me this might
> be a job for engineer$$$$.
>
> Any help here would be appreciated.
Even if you had the perfect design, I think due diligence considerations
would require some kind of engineering approval.
Without knowing anything at all about Goshen's administrative culture,
I'll suggest that if you bait your hooks properly, you might be able to
get some first class free engineering help from Purdue and Rose-Hulman
ME and Civil Engineering departments in return for well-earned bragging
rights. Then let your expensive engineer select the better plan (and
modify it if he deems necessary).
You might even consider allowing the students and faculty advisor of the
winning engineering team to do some of the work and photograph/video
record the process for posterity (and the winner's admissions department).
ga...@builditsolar.com
You might have a look at this system by Alan R:
http://www.builditsolar.com/Projects/WaterHeating/LargeDB/LargeDrainBack.htm
And a bit more on one of his more recent systems:
http://www.builditsolar.com/Projects/SpaceHeating/AlanTank.htm
These are nice simple, cheap, long lived, non-pressurized tank systems
with simple heat exchangers.
The systems serve multiunit apartment buildings with large hot water
demands -- similar to your needs.
This should lend some credence and track record to your proposal for a
simpler system -- which I think is 100% on track.
You can also use large pipe coil heat exchangers of PEX, as I do:
www.BuildItSolar.com/Experimental/PEXColDHW/Overview.htm
Gary