On Wed, 2 Jul 2014 08:55:03 -0400, "Flying Pig"
wrote:

I'll leave the original below for reference, as it's an old thread. It's so
old you might want to skip down to the attribute-marks section ([ "" in
front of each line] to get a basis point, or if you'd not seen it, to get a
clue; the masochists will read the entire referenced thread in CF) before
reading further...

So, in the end, there were 3 11x16 SS clad evaporator plates, in series,
through a constant-pressure control (functioning like an expansion valve),
running through a large evaporator-dryer, with sight glass, to a BD85 air
and water cooled compressor, all from Sea Frost.

The BD85, of course, has a higher capacity than the 50 it replaced, but at a
large cost in electrons. Low speed (there's a controller similar to the SSC
of Frigoboat, with, perhaps, some more smarts, as I don't know the specs of
how Frigoboat's unit manages the speeds, but the specs look impressive)
pulls 6A, the same as high on the 50. Ergo, at low, it should be removing
as much heat as the previous system did at high (watts = amps = BTU
removed).

The Frigoboat system did an admirable job while it was working. I could
take it to an indicated 0° in the freezer (control used box temp), far
colder than I needed, and still have it not running at full speed once the
temp had been reached.

Unfortunately for me, including the addition of the water cooling (another
complexity made unnecessary by the keel cooler in my Frigoboat system) by my
installer, Clay Hansen of Hansen Marine in St. Augustine, the system still
isn't right. As the initial installation had several defective components
(two temperature probes and the constant pressure valve) Clay gave it some
serious tweaking for about a month, while we were off the boat, and declared
it fit. On Christmas Eve 2013, when it wasn't hot, ever, it was working,
and we headed south. But...

It's voracious for electrons, and even at low speed, runs about 80% of the
time to achieve a box temperature in the 10-15° range, controlled by plate
temperature, the probe being at the bottom of the last plate. The
refrigerator, as it's fed from a spillover fan, continues to maintain its
temperatures satisfactorily, but at the cost of the freezer continually
having to run. Sea Frost's computer, if you let it run it (automatic
setting), looks for a 52% run time. Without a great deal more than the 6.5°
hysteresis recommended, there's no way possible to achieve that ratio on the
computer controlled version - and a higher gap would mean much more box temp
variation, something I'd become accustomed to not even thinking about, with
my previous (box temp) hysteresis of 2° - and to do so would be unacceptable
to me in any event.

Worse, after a decommissioning during a month-long trip ashore (I was on a
mooring and didn't dare leave it running), the third plate was only about
half frosted, with the compressor running non-stop, whereas before, it had
been complete, with the frost line ending an inch or so into the plate
before the return line to the constant pressure valve.

The Sea Frost owner, Cleave Horton, in my followups recently, had me fiddle
with the CPV, which resulted in an immediate frosting of not only the plate
but the return line (too much), and subsequent fiddlings have it back to
where it was, with the entire 3rd plate frosted, but the return line not
only very cold but almost frosted. Probably a pretty good place for it.

But that's with my having - at Cleave's strong suggestion - increased the
plate temp for shutoff by 2.5°, to 5°. So, I now have a warmer box, and
still about an 80% run time at low. I can jack up the speed and have it run
for shorter periods, but at a larger amp consumption. My expectation is that
running essentially full time, assuming the temp stays at the level you want
it to be (the function of the smart controller, rather than leaving it on a
given speed), is the most efficient. However, Cleave suggests leaving it at
low full time for the least amps used.

At that rate, based on no-wind nights, and next to nothing else on (2 0.1A
fans and breaker panel overhead), we're averaging 8 amps or higher just for
refrigeration - at night, when it's cooler. I can't support 200AH daily
loads, with added daytime (more heat, more stuff running, boxes being opened
occasionally) loads with my wind and solar unless conditions were absolutely
perfect - all bright sunny days and consistent winds of 15 or higher. The
amp draw is pretty consistent, whether on "automatic" - the equivalent of
the SSC in Frigoboat systems or "low"; we had one night of unexplainable 3A
average, and a couple of 12A average, but otherwise it's been in the 8A or
fractionally higher range overnight; this over more than 3 weeks of
observation.

After several emails back and forth, and several phone conversations, the
best Cleave could come up with was to play with the CPV, and "good luck" for
recommendations. This, despite his own manuals cautioning against lowering
the temps into the double-digit negative zone instead of merely to zero, was
accompanied by the assertion that 8° was way too cold, and that something
closer to the low to mid 20s was ample for freezing. Maybe if you're going
to rotate your stock on a daily basis, but that doesn't ring true for
anything resembling longer storage, let alone whether whatever it was you
wanted hard frozen would melt 5 minutes out of the freezer. As this system
cost, all-in, half more than the comparable (air cooled with keel cooler,
added filter-dryer and full wrap-around evaporator) Frigoboat system in
cost, this is most distressing.

Better yet, the metal screen filter in the water cooling installed (a Groco
WSB-500) started to disintegrate almost immediately, never mind its being
clogged and requiring frequent cleaning. So, for a small time, we ran it
without the water cooling. There was no chance. Temps in the freezer
remained in the mid to high 20s despite the system running full time.
Likely I'll replace the metal one with plastic, assuming I can source it
(how many times does a filter media fail?? - not much in the way of stocking
dealers), but it's just another annoyance. In the meantime, we're running
the pump, as we have a dome filter on the exterior, left over from when this
boat had air conditioning, and Cleave assures me that if it can't get
through that, the pump will happily move it along. Update since I drafted
this; I got new plastic filter media from Groco - a much larger mesh - and
sure enough, the flow improved greatly. Without the water cooling, our
system doesn't stand a chance in S. FL; it MIGHT be ok with air alone in
cold water and cool temps. However, here in Vero Beach, this last time we
took it off (only a few days' worth of running) to swap out the media, there
were 3 small barnacles on the housing. If those get into the cooling pipes,
we're dead. Obviously, whatever it is that barnacles start out as can fit
through the dome filter outside; I'm not sure that the plastic media is much
smaller. There isn't much dead time (no pump running) for these guys to get
a foothold, so I'm not optimistic. Does that mean I'll soon see a
degradation in my cooling water output even with a clear filter media???

I have no idea, at this point, whether the Carel thermostat I took out (his
electronic control has a Carel incorporated) was accurate, but I'd had it
set at 8° with a 2-degree hysteresis; it maintained it easily. This box
shows temps radically above that - the probe is near where the other was -
but shooting it with an infrared thermometer shows it to actually be about
10° at the spillover near the fan, and other temps much lower. E.g., bread
closest to the spillover 5°, hamburger package top near the plate 0°,
chicken package vertical to the plate -3°, and so on, with all three plates
well under 0°. So, regardless of the calibration, including that I've now
upped the plate shutoff to 6°, it's pretty cold in there, and the reefer
does (a good thing; it's been cold enough in the past to do it solely by
convection) occasionally have to use the spillover fan. I have yet to put
ice cream in there, but I suspect it would be OK. But still, it's an energy
hog.

If there had been any way short of pulling the boat (for a new keel cooler)
and destroying the galley (to get a new evaporator into the freezer), I
think I would have been much happier with a Frigoboat keel and air cooled
(air for when on shore, per Rob, Frigoboat's distributor in the US)
evaporator system, but also to move more air over the compressor),
protected - by the addition of a filter-dryer - from years-away refrigerant
oil contamination, allowing the capillary tube system to do its work. As it
is, the solution seems to be that I MUST run our Honda genset, every day,
for a full tank, about another ~$150/month, to keep up with the load. It
wouldn't take very long before even ditching and replacing the system every
5 years or so, including the necessary haulout, would be more cost efficient
than what I have now - if it were possible to replace the evaporator without
having to start over in the galley.

Finally, to forestall questions about the box itself, it's 6" of extruded
polystyrene, encased in epoxy, with radiation and conduction barriers
outside (aluminum foil and doorskin furring strips to give an air gap).
Both doors are double-gasketed, and, after much fiddling, I'm confident that
the gaskets are efficient. The boxes are, respectively - 16.25"D, 24.5"H
and 14.75/28.5"W - 3.4 and 6.56 CF respectively.


====

We've been through several iterations of freezers on our trawler,
starting with a water cooled holding plate system, progressing to a
pair of Engel dual voltage air cooled units, and most recently an
Igloo home style 110 volt air cooled unit that we got from Home Depot.
The Igloo home style freezer easily maintains zero degrees with fairly
reasonable current draw from the inverter. It has a large capacity
of about 7 cubic feet and only cost about $300 brand new, actually
free since I got it with Amex reward points. The big problem on a
sailboat would be placement since it is counter top high, 2 feet deep
and about 3 feet long. We installed ours on the flybridge with a
custom canvas cover that blends in fairly well with the rest of the
decor.