On 7/10/2014 1:58 PM, Joerg wrote:
>> If you aren't servicing your cooler twice a year (before and after
>> cooler season), chances are, you've got bigger problems!
>
> I service ours about every three weeks. Keeps it nice and clean.
Roof mounted makes that very tedious. Just dragging a garden hose up
there is a challenge!
>> Prior to starting cooler season, we would:
>> - uncover the cooler
>> - reinstall the belt on the fan motor
>> - oil the bushings
>> - verify the quality of the pad (in case we were going to replace it
>> at the end of last cooling season and "forgot")
>> - reinstall overflow tube
>> - turn on water supply
>> - verify water level control valve (float) operating
>> - verify pan never OVERfills
>> - turn on circuit breaker
>> - verify fan, water, and purge pumps operating
>> - verify "dry side" of cooler free of any water infiltration
>> - verify no leaks from bottom of pan
>> - verify cooler pan remains level (no water leaking out side of pan)
>> - remove the manual "gate" and reinstall barometric damper
>
> Similar here, sans the belt because there is none. Sounds more
> complicated than it really is once it becomes routine.
Didn't mean to imply it was complicated. Just another "chore".
I.e., you can't just turn the cooler on (because it "suddenly" got
hot) and expect it to use it.
By contrast, the AC condenser I can just squirt with a garden
hose DRAGGED ACROSS THE YARD to clean off the coils. It's
location protects it from accumulating much debris "inside"
the coil loop. Unlike the cooler, I *can* just turn the unit
on and expect cool air -- even if I forget/delay hosing down the
condenser.
>> By contrast, the ACbrrr just sees "turn circuit breaker on" and
>> "turn circuit off" as it's regular maintenance activities.
>
> Well, you should clean the condenser. I used to do that once at season
> start and then after every major pollen/dust/soot or whatever storm.
Where ours is located, it sees very little plant debris (nothing
to shed leaves onto/into it). It may get a little *dusty* -- but
Monsoon will wash it clean if I forget to.
>>>> ... Other worthwhile
>>>> additions are trivial to implement (e.g., a sacrificial anode in the
>>>> pan).
>>>>
>>>> When I wrote the HVAC controller, here, I watched how the cooler and
>>>> ACbrrr operated (simply by logging indoor/outdoor/operating conditions).
>>>> My goal being *smart* control not just control for the sake of control.
>>>>
>>>> E.g., two speeds is all you need for a cooler -- LOW when the house
>>>> can easily be *maintained* at a desired "comfort setting" and HIGH
>>>> when you need to move *to* a comfort setting (e.g., if house has been
>>>> unoccupied all day; silly to keep it wet and sticky!). ...
>>>
>>> That's why there's Updux :-)
>>
>> And, if you don't have an attic, that's why you have a linear actuator
>> on skylights! :>
>
> The fire marshall might not like that :-)
Upscale homes already have such items. E.g., X-10 remote controlled.
The bigger problem is we don't have skylights in every room. Those
without would not realize the full DIRECT benefits of the cooler's
operation.
E.g., even using a window as a vent/exhaust, it is difficult to keep
the office comfortable when I have several servers running. So,
I try not to *need* more than one at a time! (eventually, I will
extend the ductwork into the storage room off the garage and bring
that into the normal "living space"; but, requires knocking a
doorway through a load bearing wall so I am not in a hurry to
tackle that...)
>>>> ... The same is
>>>> probably true of the furnace (though ours only operates at a single
>>>> speed as it quickly brings the house to temperature)
>>>>
>>>> [Of course, in our case, the cooler is on the roof so we only really
>>>> hear the air flow through the ducts, not the motor noise, etc.]
>>>
>>> VF drive has big advantages. Aside from no sudden air inrush it is also
>>> better for the material. A motor that runs slow and cool most of the
>>> time will live longer than one that runs on-off full bore. Efficiency
>>> can also improve because many coolers suck the pads too dry on hot or
>>> windy days. When it runs 70% all day instead of 100% intermittently that
>>> problem can be avoided.
>>
>> We found that our cooler ran on low most of the day. If it shifted
>> to HIGH, we took that as a sign that outdoor conditions were too
>> much for the cooler to keep up. You *want* the fan to run as it
>> keeps positive pressure inside the house without which, the hot/dry
>> air from outside will blow in through all the open windows!
>
> Sure, but it doesn't even need "low" to keep up that pressure. Also,
> once the cooler is on VF drive that can be thermaostat controlled so it
> would automagically speed up when hot air wafts inside.
You don't JUST want pressure. You also want to move the HUMID air *out*
of the house -- continuously. Of course, it just gets replaced with
*more* humid air but the new air and its motion contribute to the
feeling of comfort.
We did notice the huge swings in humidity (5%RH to 60+%) took a toll
on much of the furniture -- especially the antiques. I guess they
don't like expanding and contracting like that. Worse if you switch
between cooling mechanisms (daily expand/contract vs. "seasonal")
>>>> The biggest win comes from being able to decide *when* to use the
>>>> cooler, when to STOP using it, and when to use the ACbrrrr.
>>>>
>>>> At the very least, you need local environmental data to evaluate the
>>>> effectivity of each at achieving your desired comfort setting. E.g.,
>>>> what's the local wet and dry bulb temperatures.
>>>
>>> Why? It is very easy to sense intake and output temperatures. We are
>>> sensing only output which I find enough. If the air coming out is less
>>> than 74F it's fine, if it's much higher we turn it off.
>>
>> When it's 115 outside, your cooler won't be able to achieve a
>> reasonable "comfort setting"! Even if it can drive the temperature
>> down 30 degrees, the resulting air is *soup* -- 85 and humid sucks!
>
> It doesn't get to 115F much here but it does get to 110F. It worked
> nicely through that because the outside humidity is close to zero during
> those times.
We typically have ~100 days where the peak temperature exceeds 100F
starting in about mid May and ending around 1 Sep (though it's been
that hot in October, sporadically). Unfortunately, humidity
increases in early July through mid August. So, you're facing
100+ with (relatively) high humidity (e.g., currently about 60%RH)
>> If you turn off the cooler when it is no longer comfortable, then
>> you are left sitting in a house with windows closed (else the interior
>> temperature will quickly rise to the outdoor temperature) hoping it
>> doesn't get *much* hotter (and, it is already "humid", inside).
>>
>> Instead, we (in the past) would button up the house, turn off the cooler
>> and turn on the ACbrrr. But, now it has to work to remove all that
>> moisture from the house. ...
>
> Yup, and there goes a lot of the energy savings.
There's no free lunch. Sit in a "muggy" house "relatively inactive"
(lest you drown in perspiration) or use more aggressive means to
make the house more comfortable for a wider variety of activities.
The ACbrrr's biggest appeal, to me, is dehumidification.
I moved 20+ tons of stone into the yard when the outdoor temperature
was above 110F (117 on one day). I was *more* comfortable outside
than I was returning to the swamp cooled house, afterwards -- because
I wasn't soaked in perspiration out in that heat (5%RH) but immediately
was "drenched" when I came indoors -- 30+ degrees cooler but humidity
approaching 70%!
[I was surprised at how easy *heat* is to tolerate -- as long as it is
dry *and* you are not in direct sunlight. You just have to get used
to drinking a quart of water every half hour, NOT perspiring and NEVER
having to take a piss! These tend to have a disconcerting psychological
effect -- "Where the hell is all the water going???"]
>> A better solution is to *know* this is likely to happen and avoid
>> the need to change cooling mechanisms mid-day. E.g., we wouldn't
>> even bother to turn the cooler on once Monsoon started. It was a
>> *known* that it would not be able to keep the house comfortable
>> in the high temps and outdoor humidity that we *expected* each day.
>> Unfortunately, Monsoon isn't a homogenous experience. It is not
>> uncommon for things to "dry out" for extended periods of time
>> *in* Monsoon. Just looking at temperatures isn't enough to tell
>> you how you should control *now* or even *today*. That's why
>> there is a fair bit of "art" to using a dual-cooling system.
>> Watching weather forecasts, actual conditions developing, etc. and
>> using "wetware" to decide which appliance to use on any given day.
>>
>> [you end up playing it safe and using the ACbrrr more often than
>> not as the cooler->ACbrrr transition is usually an admission that
>> you screwed up and are now uncomfortable as a result]
>
> We pretty much play that by the ear. TV has become unreliable as a
> weather forecast since this dreaded switch to digital. But we can gauge
> it pretty well by eye.
Forecasts (even current conditions) are pretty meaningless, here.
Too many microclimates. E.g., our street is probably four blocks
long. But, Spring flowers (same species) will bloom two weeks
earlier on the other end of the street than here. Likewise,
our winter lows and summer highs are several degrees lower/higher
than reported for nearby "weather observations".
Hence the reason for the rooftop weather station -- I don't care
what the humidity is "downtown" -- or, even at the monitoring station
across the was (~1 mile as the crow flies). I can look up and see
rain in one direction, virga in another, and clear skies above! :<
>>>> ... Windspeed also plays
>>>> a role as it can force air through the pads in "gusts" effectively
>>>> allowing hotter, drier air into the system than intended. You also
>>>> need to be able to monitor conditions *inside* to ensure the house
>>>> itself isn't becoming saturated (e.g., by insufficient exhaust
>>>> ventilation)
>>>
>>> That's pretty easy. We make sure the ventilation matches the cooler
>>> setting (high/low). One can also measure the pressure differential.
>>
>> It's not static! Are you going to keep tweaking your VFD every
>> time a gust of wind comes along blowing hot air *in* through
>> your open windows?
>
> Nope. Hint: PID-regulator. A really sloooow one. This is where a uC is
> called for.
Doesn't matter how fast/slow your control loop operates. Gust of
wind is too short of a disturbance -- hot dry air comes *in* the
window. You can't react quick enough to detect and counter it!
By the time you *do* react, it is gone and you've just created
an "artificial gust" inside the house. Overdamp your response
and you've done nothing to counter the incursion.
Given that the outside air is MUCH hotter (20-30F) and much *drier*,
it becomes very noticeable. Your only practical recourse is to
install windbreaks to diffuse these (some office buildings here
do just that -- of course, it's an eyesore to those hoping to
stare out their window...)
>> The cooler's effect relies on open windows in each room -- because
>> the cooler isn't recirculating air but, rather, EXHAUSTING the
>> "hot"/stale air *through* those opened windows. I.e., no matter
>> which direction the wind originates, chances are, you have open
>> windows/vents facing that direction. Stand by a window and feel
>> a blast of 100+ air blowing *in* and you really notice it (windows
>> tend to be located in places where people sit, sleep, work, eat, etc.)
>
> On hot days, if there is any wind, it comes from the south and we don't
> have any windows in that directtion. Well, one slightly in the kitchen
> and then we just open another one instead on those days.
South side, here, is largely consumed by the location of the garage.
Wind comes from every direction, here. Microbursts tend to come
from the North (closest side to the mountains) but storms come in
from the South (Mexico), North, West (CA) and even East (NM, TX).
Wind direction is usually a coarse predictor of our likelihood for
precipitation.
>>> All you need to measure for that is output temperature. 74F is our
>>> limit, works well.
>>
>> And, at that point, you just resign yourself to being uncomfortable?
>> Or, do you resort to some *other* cooling mechanism (gee, maybe AC)?
>
> That's the output temp from the cooler. Yesterday it was 80F in the
> house with the cooler on high. A bit much for my wife but bearable. For
> me, just fine.
It isn't possible to get the house down to 80F with a cooler, here,
during Monsoon. And, in Summer, not possible once the outdoor temps
climb above 110.
By contrast, it'll be 100+ with DP above 60 and the ACbrrr will still
be operating at a low duty cycle.
>> [Forget using one *during* a rain storm!
>
> Wot's a rain storm? :-)
When you get an inch of rain in an hour.
>> Currently, I have a small weather station on the roof so I can
>> watch conditions *here* (which ALWAYS differ from those reported
>> on the "news" due to microclimates around town). I use this
>> for instantaneous climate data for the controller (as well as the
>> irrigation controller -- don't bother to irrigate if it's likely
>> to rain/has rained!) so it knows whether the cooler "makes sense"
>> as a cooling option "now".
>>
>> [I am unhappy with the humidity sensor, though. The constant
>> sun/heat exposure seems to render them inoperable, quickly.
>> And, I'm not keen on the expense and maintenance of a chilled
>> mirror technology!]
>
> Or ... run a MEMS-size mini cooler which gives you that information :-)
Have to come up with solutions that can easily be reproduced without
*me* having to make big investments (that essentially "benefit others"
more than myself). Telling someone to buy any of the many such
"weather monitoring stations" commercially available takes a lot
less of my time and money. And, gives me more information (*is*
it raining? what is the amount of total rainfall? what direction is
the wind blowing? etc.)
Hard to tell what evapotranspiration is likely to be without more
detailed measurements than a "cooler model". Or, whether I should
*rush* to use any harvested rainwater NOW to be able to accommodate
any likely precipitation "soon".
>>> Out here it's easy. You turn on the cooler and occasionally peek out the
>>> office window. If clouds roll in it may make sense to turn it off.
>>> Otherwise we just keep it running and when it gets to cold in the
>>> evening we turn it off earlier than usual.
>>
>> In Summer, it is *usually* 90+ at 11PM. It *may* drop to 80
>> in the early morning hours. (surprisingly, this feels "cool")
>>
>> OTOH, in Monsoon, 80 is icky -- midwestern/new england summer icky!
>>
>> I.e., temperature alone tells a controller nothing. This is why
>> so many cooler controllers are jokes.
>
> You need both the cooler input and output temps, plus the house temp.
Still tells you nothing historical or predictive. Is the barometric
pressure FALLING? If so, chances are a storm may be moving in
(which means humidifying the house will be a losing proposition as
the outdoor conditions will be getting more humid).
These are the sorts of "expert decisions" that a human makes by
"looking out the window". He surely can't tell the *temperature*
by such a visual observation! (unless he is glancing at a thermometer
mounted outside the window! :> )
But, you can only do that if you can make (and record!) lots of
observations *and* actively (intuitively?) correlate between
them. I.e., something to which a machine is ideally suited!
<grin> Doubtful. Had no problems in Boston or Denver. (Boston
I figure was too urban for "natural" allergens; yet Denver/burbs
was nowhere near as much). Only problems "mowing the lawn" in Chicago
(despite being *far* more exposed to natural allergens). No lawns
here but my "seasonal allergies" have become "continual allergies" (as
something is *always* blooming with our year-round growing cycles).
Yet, SWMBO suffers on days that don't bother me at all! :-/
I've seen many folks from "out of town" literally unable to *see*
due to the excessive tearing and sneezing when they encounter our
flora. Don't know whether it's the quantity of airborn allergens,
the variety or just the fact that they are all "foreign" to their
immune systems... :<
Scratch test I think I "hit" on 43 of 45 allergens tested. MD
decided further testing would just reveal "more of the same" :-/
[What was most amusing was the things to which I had *no* reaction!
E.g., negative to "dog" but positive to cat and horse, etc.]
(SWMBO was really annoyed that *dust* wasn't a problem for me as
she was hoping she could use it to motivate me to "dust" the office
more often :> )
Allergies are annoying -- and can seriously impact quality of
life! -- but there are folks with far bigger problems that help
remind us to keep things in perspective!