i’ve been contemplating (for about twenty+ years) how having (example purposes only, here simplified) having two water meters on two separate water lines might be a greater risk (increased probability) of leaks compared to exactly the same water line all joined together (fittings and all, same length of pipe etc) on one water meter only.

mostly i’ve been happy that there is no difference, excluding a few things

- that I have to check more than one (drive to them, do the reads, some math, and guess from information available*)
- that a water pipe blowing off near a meter presents less flow restriction (compared to further down the pipe it maybe double)
- meter failure

*like for example each trip I minus the previous total from the most recent one, work out the average flow per minute, and additionally I have the flow measurements when did the totals, and from those quite a lot of information can be extracted.

now, on the subject of more meters increasing risk, rather than thinking of it from a supply perspective I instead saw it from the other direction, being pipes and all present resistance to flow, and that this is ‘communicated’ back to the water meter by the water.

from there I was able to establish there was some ‘communication’ between the separate water lines.

the communication was via me.

generally if I have a leak i’ll do follow up reads after the repair and do the sweep of all of them because I happen to be out that way.

if risk is seen to be the size of a leak and the time it might run for before being detected and fixed, then risk is simply (apparently)lowered by increasing the frequency of reads.

anyway, there apears some subtlety about the ‘assumptions’ made from water meter readings, the confidence level. I can know with some accurancy what has happened, I can’t know what will happen when, though I know of what may happen.

so when I read a meter i’m taking to it (and the associated water line) ‘assumptions’, involving ideas of certainty and uncertainty.

the resistance to water flow of the water line indicated at the meter becomes incorporated into my little computational organ and subject to assumptions.

when I have two meters compared to just one I have twice as many assumptions, if it’s linear.

and every time my attention is drawn to something and I touch it then there is too the added aspect that my ‘interference’ might contribute to that unnoticed/unknowns, even for a competent fixer, double triple checker like me.

I do ~ 700+ meter reads a year.

transition, have you looked into the economics of remote flow meters?

I set the launceston city council up with them on their sporting grounds. When a leak occurs, it sets off alarms in their head office computer, they can monitor water usage at each ground and adjust usage accordingly

>remote flow meters?

designed and built some of them, starting 20+ years back, hence the intimate knowledge of the subject

reading meters is a small aspect of my job, though a lot of information is gotten from, i’m more interested in how human assumptions can effect ‘communication’ between things that appear isolated from each other.

human ‘anticipation’ and ‘expectation’ sort of near fascinates me (not that much gets me real interested these days).

So this isn’t a case of familiarity breeding contempt?

transition said:

>remote flow meters?

designed and built some of them, starting 20+ years back, hence the intimate knowledge of the subject

reading meters is a small aspect of my job, though a lot of information is gotten from, i’m more interested in how human assumptions can effect ‘communication’ between things that appear isolated from each other.

human ‘anticipation’ and ‘expectation’ sort of near fascinates me (not that much gets me real interested these days).

No help from me, I’m only familiar with flow meters used in fluid mechanics experiments, and then only one at a time. Perhaps if you post a diagram, complete with locations of any non-return valves.

Follow on question. What is the mechanics of the flow meters you use? Are they orifice plate, venturi, propeller, pitot-static tube, ultrasonic or other? I assume they’re not hot film.

The essence of dealing with risk is to consider the consequences of failure, and provide redundancy when they are unacceptable.

But try telling the banks that.

The Rev Dodgson said:

The essence of dealing with risk is to consider the consequences of failure, and provide redundancy when they are unacceptable.

But try telling the banks that.

Or the Climate Change Doubter Alarmists for that matter.

The Rev Dodgson said:

The Rev Dodgson said:

The essence of dealing with risk is to consider the consequences of failure, and provide redundancy when they are unacceptable.

But try telling the banks that.

Or the Climate Change Doubter Alarmists for that matter.

The banks have no risk…the gubmint will save them…

are these water meters for irrigation

are you taking readings for a supply authority?

where’s the water being taken from? a river or bore near a river?

are the meters the mechanical types with a turbine? do you have debris that clogs up the turbine ?

>are you taking readings for a supply authority?

The thread’s about if you combine two separate water systems that did have two meters (theoretical Q) and put them on one meter, and that all the plumbing’n pipes’n troughs and all are exactly the same, what factors are at work that do, or seem to reduce the likelihood of leaks (of analysing the probabilities, as it turns out).

It could be water lines, or something else. I used it as an example.

—-

http://en.wikipedia.org/wiki/Segal%27s_law

Segal’s law is an adage that states:

“A man with a watch knows what time it is. A man with two watches is never sure.”

It refers to the potential pitfalls of having too much potentially conflicting information when making a decision.

In reality a man possessing one watch has no idea whether it is the correct time unless he is able to compare it to a known standard,

Glance Fleeting said:

Segal’s law is an adage that states:

“A man with a watch knows what time it is. A man with two watches is never sure.”

It refers to the potential pitfalls of having too much potentially conflicting information when making a decision.

In reality a man possessing one watch has no idea whether it is the correct time unless he is able to compare it to a known standard,

Was thinking of this last week in the context of reading a blood oxygen monitor. The value was fluctuating between 77% (deadly) and 99% (safe), and gave null reading more often than a valid reading. I concluded that a relatively good method was to accept the result if three readings in a row differed by less than 3% or, failing that, taking the median of ten readings.

So, in the context of Segal’s law, take the time from the median value from 10 watches.

Here’s an example from an unusually busy period, quite a way back now, just occasionally i’d put in the quick math to do with comparing present flow with averaged flow (second-to-bottom), generally I don’t put that in, it stays in my head. I’ve taken the pressure readings out to tidy it up some, and left out some in-between reads.

When I do reads i’m doing analysis of risk probabilities. When a meter comes back to zero flow I know for certain there are no leaks, if it comes back to very low flow but hasn’t stopped for a while there is some chance there may be a leak but it’s a small volume so is low risk. That sort of thing. I tend toward doing meter reads when there should be the least or no flow, but that’s not always convenient (late at night/early morning or during or directly after a rain is good).

Some of it might display like a dog’s breakfast, we’ll see. I’ve taken out the square brackets etc.

21/11/13, 11:40pm reads and fix leak, check most feed pad troughs
ball float came off, refitted. Approx 28KL lost, estimate been going for 15-26 hours

east 027, 594/zero/
house 5352, 481/zero/
dump 504, 099/60ml/
bamb 1806, 814/50ml/
hut 946, 693/70ml/
yards 2877, 648/fast leak/ > yards 2877, 811/80ml (15minutes later after fix)
back 2191, 414/160ml/

Tuesday reads only
e 027, 592/zero/
h 5345, 932/zero/
d 502, 383/70ml/
b 1804, 827/60ml/
h 945, 589/40ml/
y 2846, 260/80ml/
b 2188, 912/360ml/

midday 15/11/13
e 027, 588/zero/
h 5343, 969/zero/
d 498, 379/zero/
b 1798, 969/700ml drink/
h 942, 009/70ml/
y 2843, 021/zero/
b 2180, 557/1.7L drink/

Subject: 12/11/13, 2:00pm fix leak + reads
replace bury 5metres line to east kanga trough 20mm poly, 2×20mm end-connectors, 2× 20-25mm bushes. Pump troughs back line, drive line, walk 200metres, dug around a bit, can’t find, is 160ml, have a look some other day is only small.

east 027, 580/zero
house 5338, 399/zero/
dump 495, 595/60ml/
bamb 1795, 587/350ml/
hut 939, 991/70ml/
yards 2841, 248/50ml/
back 2175, 942/400ml/160ml pumped/

Subject: 11/11/13, 9:30pm reads and leak
leak on east meter, trough riser pipe split

east 027, 566/leak/turned off
house 5338, 239/zero/
dump 495, 539/70ml/
bamb 1795, 025/140ml/
hut 939, 468/zero/
yards 2841, 063/80ml/
back 2175, 108/360ml/

Subject: 10/11/13 reads + troughs
5:14pm follow up reads (# = average flow since last read/minute)
east 009,833/zero/——-# zero
house 5337, 889/7L drink likely/#~6H.8L
dump 494, 037/50ml/#~6H1.5L
bamb 1793, 236/2.8L/> later >1793, 271/1.2L slowing/#~6H3.5L
hut 938, 597/70ml/#~6H1.65L
yards 2840, 279/zero/#~6H1L
back 2173, 109/1L/#~6H3.3L

11:00am,10/11/13 clean feed pad troughs + copper, resit’ trough lid closest yards meter.
east 009, 833/zero
house 5337,605/zero/
dump 493, 487/50ml/
bamb 1791, 986/20L fill/
hut 938, 003/zero/
yards 2839, 903/zero/
back 2171, 910/750ml/