Every now and then I hear a toilet
filling for some number of seconds. There are a couple of
reasons why that might happen. Maybe there's a slow leak and
depending on the amount of hysteriesis between the stop filling
level and the start filling level on the float the toilet is
topping up. Maybe the water company has a large variation in
the pressure that's feeding my house and even after the pressure
regulator (Fig 2) there's still a
consideerable amount of variation in the pressure inside the
So I'd like to monitor the water
pressure. But there's no off the shelf solution for
measuring your water pressure so on this page I'll document what
The first idea is to use the oil
pressure sender for a motor vehicle. It turns out that there
are two kinds. One is a switch that drives the red OIL light
and that will not work for this application. The other is a
sender for an oil pressure guage and that's what I got.
The box is marked:
UPC: 7 07390 84219 6
Ant the sensor is marked:
U.S. Pat 4079351
This sensor (and I suspect most of
them) has 1/8" NPT threads. There are a number of places
that the sensor might be connected. The first thought was on
an outside hose bib, but that would mean everything would be
exposed to rain and large temperature variations. Some sink
faucets have male threads (kitchen) or female threads (bathroom)
so that was a possiblity, but I didn't want to give up the kitchen
sink and getting the correct male threaded adapter for a bathroom
sink may be a problem. Since there is a spare shower that
seemed to be the logical place to install the sensor. The
shower has a male 1/2" NPT fitting so all that's needed is a
Reducing Coupling 1/2" x 1/8" NPT (UPC: 0 48314
Some Teflon tape was used on both joints. See Fig 1.
Using the Fluke
model 87 Digital Multimeter in MIN-MAX Resistance mode will
allow measuring the pressure variation. It's starting out at
At this time I don't know how to convert from Ohms to PSI, but using an analog meter on another location should solve that.
1 June 2012 - Max 89.9, Min 31.4 average 38.3 Ohms.
Bought a new "Watts"
model IWTG made in China pressure gauge from Home Depot
(UPC: 0 98268 15995 0) that fits a hose bib and the
dial is marked 0 to 200 PSI. It measured 100 PSI even when
the pressure regulator screw was all the way out or all the way
in. At first thought the pressure regulator was bad, but
then noticed the the water pressure in the house was very
high. The black needle moves from 0 up to 100 and back
smoothly so the problem seems to be a design defect rather than a
calibration problem. The red tattle tale high pressure
pointer does not have enough friction to hold it's position so is
of no use.
Backing off the regulator screw
lowered the pressure. After doing all that the readings are:
2 Jun 2012: Max 89.9 Ohms, MIN: 28.6 Ohms, Avg: 51.0 Ohms
When there's no pressure the
resistance is 0 Ohms. Note with 0 Ohms for no pressure an
Oil Warning light can easily be turned on in addition to being
used to drive an analog gauge.
So that's one point on the pressure
vs. resistance plot.
If the sensor is linear then one other point will set the scale factor.
NOTE: maximum reading is about 90 PSI even if pressure is higher.
|With kitchen sink running
pressure is about 55 PSI.
||With no water running
pressure is about 110 PSI.
Using a 3/4" x 1/8" Brass Reducer the pressure guage
from the broken air compressor regulator was attached to a
hose bib using Teflon tape. With the kitchen faucet
running full on the house water pressure was adjusted to
50 PSI. With the kitchen faucet still running the
oil pressure resistance was 50.0 Ohms.
Maybe I need to drain the house to re-establish the air traps?
Got this gauge to check the house water system as part of the planning for an external home fire protection sprinkler system. I feel it is much more reliable than the Made in China gauge.
This unit is composed of a 3/4" female GHT to female 1/4" NPT adapter and a pressure gauge with a 1/4 NPT male fitting.
Pressure responsive sender, General Automotive Speciality Co.,
Inc., Mar 14, 1978, 338/36; 73/723; 338/42 - they claim much
better performance and built-in calibration when compared to other
analog pressure sensors (NAPA p/n OP6638).
PRC68 Home Page,
of web pages, Products for Sale,
Page created 31 May 2012