Re: Emergency Water Turn-Off?

["Robert Green" <ROBERT_GREEN1963@xxxxxxxxx>]

"Matt" <mattmorgan64@xxxxxxx> wrote in message
news:1115672290.514935.234300@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> How about measuring flow rate over a period of time, instead of
> acoustics?

Probably a better method, but I'll bet it's much hard to implement.  IIRC,
the flow rate on my water meter is pretty coarse.  By the time it indicated
a leak, you could have some serious damage.  I assume you can get ones that
measure to the tenth of a gallon, but the expense might start climbing into
the prohibitive range.

> An icemaker, or a toilet refilling should be a pretty low rate of flow,
> and also shouldn't go for more than 1 minute.
>
> A burst washing machine hose, pipe, etc, on the other hand, would have
> a high rate of flow and go on forever.

Agreed that there are profiles to various uses, but IMHO there are SO MANY
potential water use profiles that there are going to be an awful lot of
false shutdowns.  Consider a guest (worst case, a MIL)  brushing her teeth
and letting the water trickle slowly from the faucet.  Bang!  Off goes the
water.  That will get tired *very* quickly.  Maybe if you had a flowmeter on
both the water main and the waste line you could really differentiate leaks
from a MIL brushing her teeth.

> If you could get a measurement of flow rate, (which must be
> possible as most water meaters are now read remotely),

Not mine!  Some guy comes around with a plate puller.  The company rounds
out usage to the nearest 1000 gallons.  I'm not sure you can get the kind of
readings from existing meter that would be useful in detecting small leaks.

> combined with time of day/day of week and occupancy should
> result in the computer being able to make a pretty good guess
> about whether or not to shut down the main.

I agree with the theory, but in practice, I think you could spend an awful
lot of time futzing with such a system only to eventually abandon it because
the exceptions became too bothersome.

> Some paramaters would probably have to be tweaked over
> time for each home, and providing an override function to turn
> the main on or off, or tell it not to shut off, regardless of what it
> thinks should be done would be nice too.

Yeah, you'd have to have a MIL cutoff that killed the function completely.
I don't think an override would be very effective for guests unless you
mounted some sort of controller over each faucet.  In my experience it's the
"guest test" that really makes or breaks a lot of neat HA ideas.

> I don't care much for the detectors on the market today, for the
> reasons mentioned by Robert. I can say I'd rather have one valve
> on the main that occasionally gave me a nuisance lock out, if it
> also one day shut the house down when a pipe burst vs individual
> gizmos for each toilet/sink/etc.

The problem I see is that if a leak matches a valid usage profile, you've
got a flooded basement.  Unless you go about putting pinhole leaks in pipes,
cracking old solder joints, rupturing washer hoses and letting tubs overflow
you're only profiling half of the equation.  I don't think it's possible to
build a system with a high confidence of correct operation if you only model
the proper behavior and not the actual leak cases that you're hoping to
find.

It reminds me of the gentleman here a while back who built the cat door that
used imaging to process "go/no go" rules.  The camera picked up intrusions
by racoons, birds and unknown other creatures and allowed the neighbor's cat
to go in and out freely as well, since it, too, had a "cat" profile.  The
problem I see is that house leaks happen maybe 5 times in a lifetime.  Too
few to accurately model or to spend lots of time or money to prevent.

> To counter the loss of power problems, after the program was
> written, it could be burnt onto an eprom.... hell there are entire
> PC's now the size of a cell phone; from there it's just a matter of
> providing an API

Agreed that there are many ways to counter the power problem, but it's one
of many that need to be dealt with it.  I'm just not sure it's worth it.  If
I decide to equip my next house with automatic sprinklers, that will even
further complicate the issue for me.

> to interface with via a PC. Keep the logic on the eprom, and let
> the user tweak paramaters via an API.

Unfortunately, my experience is that the simpler the system the more likely
it is to be used and not end up in some extended "forever testing" mode.  I
suppose that's why WaterBug and WaterCop are as popular as they are.
There's a lot of engineering smarts embedded into the units that aren't
apparent at first glance.

--
Bobby G.