>One question though: Why would you set the mixing valve on the solar
slightly lower than the
>thermostat on the electric tank? My thought was to set it the same or
slightly higher so that
>the electric would only come on when needed.

Maybe I wasn't clear- the mixing valve should be on the output of the
elecric tank, and the feed from the solar store into the electric tank
should be at full temp, whatever that happens to be, undiluted by the cold
water.  If you insulate the electric tank well the standby loss of the
higher temp feed won't be any worse than if that water was in the solar
tank, but it ensures a minimum of power use on the electric tank.  You set
the mixing valve to the lowest temp that actually meets your needs, since
the lower the temp of the water abandoned in the distribution plumbing at
the end of a draw, the lower the net loss.

Typically you'd need 115F water to end up with 110-112F tub-filling water
through an anti-scald mixer with 40F water feeding the cold side of the
mixer.  You need to keep the storage temps above 120F to limit the growth of
Legionella culture- they won't grow at 120F+, but they won't die. It takes
140F over extended periods to actively kill Legionella, but the higher
increases the standby losses by quite a bit, so you don't want to be using
electricity to get there. There will be times when the solar water temps
peak at 140F or higher, so 120-125F for the electric backup setting is just
fine.

If you haven't already, go ahead and swap out the sacrificial anode on the
electric tank (ever 7 years whether it needs it or not- put a tag on it with
both the date it was replaced and the date it should be done again.)  An
electric tank will go pretty much forever if you swap the anodes regularly
as it keeps corrosion at a minimum. it's usually a sub-$35 part.  The only
other thing that's likely to fail is the heating elements, and they too ar
swappable on the cheap. Elements typcially die from thermal cycling from
turning on & off.  Usually the upper fails first, which comes on first
during recovery and is sometimes the only element that fires. With solar
providing most of the water heating energy the elements will see far fewer
cycles than when they're providing 100% of the energy.  Even if the sucker
is already 10-12 years old, if the anode hasn't failed yet you should be
able to get another decade (or even two) out of it, with very little in
maintenance, save maybe an element swap at some point.

As a side note: Something often lost in the tankless discussions is the
standby power and freeze-control power used by gas/propane tankless heaters.
It varies a lot between units, and can sometimes approach the standby power
use of an electric tank heater(!).  So in a low duty-cycle high-standby
application such as solar backup the standby power of a gas-fired tankless
deserves some scrutiny.

Second side note: At the current subsidized cost of photovoltaics there's a
financial argument in this scenario for spending the extra coupla grand you
might have spent on a tankless and putting it into PV.  The unsubsidised
lifecycle cost of PV has already dropped below 10cents/kwh, but with subsidy
we're talking sub-5-cent electricity in some situations.  With reasonably
open shading factors a half-kilowatt array will return over 600kwh/year in
most of New England, and can (depending on subsidy) cost similar money as
tankless HW heater.

>Would it make sense to tap into it for fall and spring radiant floor
heating?

It depends a bit on just how much excess heat you have, and the
type/efficiency of your heating system. Installing radiant (even with a DIY
staple-up) isn't cheap, and it could potentially take a couple of forevers
to break even on fuel savings. Unless you have a very tight and highly
insulated house the money might be better spent elsewhere, from an ROI
perspective.  If it turns out you need heat dumping in the spring/fall to
keep the storage temps bounded, setting up a small supplemental-heating zone
using low temp panel radiators (eg:
http://www.irawoods.com/Myson-SX-60-160G-Select-Panel-Radiator ) would be
cheaper & easier to implement than radiant floor, and still pretty cushy (if
not quite as cushy as toasty-warm floors.)  To meet code it would have to
have a heat exchanger isolating the heating side from the potable side, but
in some instances "open" systems would be allowed, provided that there is a
guaranteed minimum duty cycle on the heating loop to avoid stagnation
issues, which could end up been a net loss.

dana

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
SkiVt-L is brought to you by the University of Vermont.

To unsubscribe, visit http://list.uvm.edu/archives/skivt-l.html