Sugegestions for monitoring water level in 55 gallon drum

774 views
Skip to first unread message

Tom Gondek

unread,
Apr 30, 2012, 2:42:00 PM4/30/12
to milwaukee...@googlegroups.com
I need to monitor the water / saltwater in different 55 gallon drums.  I have some simple float valves, but that only tells me full of not full. 
 
What I want is to use some type of sensor and arduino to monitor if the drum is 0% to 100% full.  Suggestions on what to “sense” this with????  As mentioned above one drum is saltwater, so anything the comes in contact needs to be saltwater safe.
 
Thanks
Tom

Jerry [HM]

unread,
Apr 30, 2012, 2:44:20 PM4/30/12
to milwaukee...@googlegroups.com

Can you cheat? And measure weight instead?

 

Might give you more options to use with the Arduino?

 

(I just saw something like this posted on the Arduino forum too..) 

Tom Gralewicz

unread,
Apr 30, 2012, 2:45:52 PM4/30/12
to milwaukee...@googlegroups.com
A common solution is ultrasonic range finders, you will need to find a way to keep the salt water from corroding them.

http://www.sparkfun.com/search/results?term=range+finder&what=products

(or you can put each barrel on a scale and read the weight)

Tom
--
Tom Gralewicz
Miller Electronics Recycling
(414) 380-1716
www.deadcomputers.com

Shane

unread,
Apr 30, 2012, 2:47:10 PM4/30/12
to milwaukeemakerspace
I wonder if there's a way to have something that floats on the surface
of the water that is "detected" by something mounted on the rim of the
barrel. Perhaps it could read how far away the float is from the rim
detector?

If the water doesn't move too much, maybe just a floating surface and
use a laser tape-measure hooked up to an arduino to tell you the water
level and/or alert you when it's low.

Pete Prodoehl

unread,
Apr 30, 2012, 2:47:36 PM4/30/12
to milwaukee...@googlegroups.com

PING))) Ultrasonic Distance Sensor?

    http://www.parallax.com/tabid/768/ProductID/92/Default.aspx

Maybe it could be put in a water-tight case...

Pete

Shane

unread,
Apr 30, 2012, 2:51:43 PM4/30/12
to milwaukeemakerspace
You could go ultra simple:

A weighted float that sits on the surface attached to some fishing
line that goes up to the rim, over a little pulley, then down to the
bottom where it connects to an indicator.

As the float drops, the indicator is pulled up a small channel or
groove attached to the outside of the barrel with markings indicating
the water level.

That's about as simple as it gets. You could even have it trigger an
arduino to tweet at you when the level is too low....

Royce Pipkins

unread,
Apr 30, 2012, 2:51:31 PM4/30/12
to milwaukee...@googlegroups.com
The outdoor MaxSonar models might work, but they are $100.

I've seen those magnetic fish tank cleaners that push a submerged
glass cleaner around from the outside. I wonder if you could build a
sealed plastic tube with sonar target inside that is magnetically
attached to a float on the outside. As the water level fluctuates the
the float will drag the exterior magnet and that would move the
interior magnet.
--
The real problem is not whether machines think but whether men do.
B. F. Skinner

Brent Bublitz

unread,
Apr 30, 2012, 2:52:06 PM4/30/12
to milwaukee...@googlegroups.com
Connect a float to a simple pot via swing arm arrangement and have a
arduino or other chip monitor the resistance.

Brent

Brent Bublitz

unread,
Apr 30, 2012, 2:53:31 PM4/30/12
to milwaukee...@googlegroups.com
Shane> even simpler, have a float connected to a rod sticking out the
top with level markings on the rod.

Brent

Tom Gralewicz

unread,
Apr 30, 2012, 2:54:30 PM4/30/12
to milwaukee...@googlegroups.com
Sealed sonar solution:

Use some plastic PVC pipe with a ring float on the outside that has a magnet.  Inside the sealed tube have a plug with another magnet.  As the outer one goes up so does the inner one.  Seal both ends of the PVC and use an ultrasonic sensor or other measuring device inside.

Tom

Tom Gondek

unread,
Apr 30, 2012, 2:58:13 PM4/30/12
to milwaukee...@googlegroups.com
would the sides of the pipe reflect the signal giving incorrect readings or are most of the transmitting cones very narrow?

Tom Gralewicz

unread,
Apr 30, 2012, 2:59:57 PM4/30/12
to milwaukee...@googlegroups.com
Don't know, you would have to give it a try.  You could also build some kind of optical encoder - a larger version of what the electronic calipers use.

Adam Cohen

unread,
Apr 30, 2012, 3:16:01 PM4/30/12
to milwaukee...@googlegroups.com
Put a magnet on a floater and then have a series of hall sensors.  Total cost ~$20

Adam Cohen

unread,
Apr 30, 2012, 3:17:42 PM4/30/12
to milwaukee...@googlegroups.com
even simpler, make it so that when the water reaches a certain level it completes a circuit.

Ron Bean

unread,
Apr 30, 2012, 3:31:47 PM4/30/12
to milwaukee...@googlegroups.com
>Put a magnet on a floater and then have a series of hall sensors. Total
>cost ~$20

I've seen tanks with a series of metal rods of different lengths. The
higher the water, the more circuits get completed. What kind of
resolution do you need? Maybe a few near the top and a few near the
bottom?

[Are you filling the barrel, or emptying it?]


ma...@bryskier.net

unread,
Apr 30, 2012, 5:04:21 PM4/30/12
to milwaukee...@googlegroups.com
Try this. The company that makes it can provide larger sizes I believe.
https://www.adafruit.com/products/464

All the best,
Marty


> -------- Original Message --------
> Subject: Re: [MakerSpace] Sugegestions for monitoring water level in 55
> gallon drum
> From: Adam Cohen <adam.cohe...@gmail.com>
> Date: Mon, April 30, 2012 2:17 pm
> To: milwaukee...@googlegroups.com
>
> even simpler, make it so that when the water reaches a certain level it completes a circuit.
>
> On Monday, April 30, 2012 2:16:01 PM UTC-5, Adam Cohen wrote:Put a magnet on a floater and then have a series of hall sensors.  Total cost ~$20
>
>
> On Monday, April 30, 2012 1:59:57 PM UTC-5, Tom Gralewicz wrote:Don't know, you would have to give it a try.  You could also build some kind of optical encoder - a larger version of what the electronic calipers use.

>
>
> On Mon, Apr 30, 2012 at 1:58 PM, Tom Gondek <tgo...@hotmail.com> wrote:
>
>
>
>
> would the sides of the pipe reflect the signal giving incorrect readings or
> are most of the transmitting cones very narrow?
>
>
>  

>
> From: Tom Gralewicz
> Sent: Monday, April 30, 2012 1:54 PM
> To: milwaukee...@googlegroups.com
>
> Subject: Re: [MakerSpace] Sugegestions for monitoring water level in
> 55 gallon drum
>
>
>  

>
>
> Sealed sonar solution:
>
> Use some plastic PVC pipe
> with a ring float on the outside that has a magnet.  Inside the sealed tube
> have a plug with another magnet.  As the outer one goes up so does the
> inner one.  Seal both ends of the PVC and use an ultrasonic sensor or other
> measuring device inside.
>
> Tom
>
>
> On Mon, Apr 30, 2012 at 1:52 PM, Brent Bublitz <pho...@gmail.com> wrote:
> Connect a float to a simple pot via swing arm arrangement
> and have a
> arduino or other chip monitor the resistance.
>
> Brent
>
>
>
> On Mon, Apr 30, 2012 at 1:47 PM, Pete Prodoehl <ras...@gmail.com> wrote:
> >
> >
> PING))) Ultrasonic Distance Sensor?
> >
> >     http://www.parallax.com/tabid/768/ProductID/92/Default.aspx

> >
> >
> Maybe it could be put in a water-tight case...
> >
> >
> Pete
> >
> >
> >
> > On 4/30/12 1:42 PM, Tom Gondek
> wrote:
> >
> > I need to monitor the water / saltwater in different 55
> gallon drums.  I

> > have some simple float valves, but that only
> tells me full of not full.
> >
> > What I want is to use some type of
> sensor and arduino to monitor if the drum
> > is 0% to 100% full. 
> Suggestions on what to â��senseâ�� this with????  As

Tom Gralewicz

unread,
Apr 30, 2012, 5:07:11 PM4/30/12
to milwaukee...@googlegroups.com
A touch screen for tanks :-)

Jack Driscoll

unread,
Apr 30, 2012, 5:35:44 PM4/30/12
to milwaukeemakerspace
Use a pressure sensor at the end of a tube; you can use stiff plastic
tubing for the saltwater. Depth of water can be calculated from the
pressure. Sensor does not have to be immersed in water, so corrosion
is not an issue.

http://wiki.answers.com/Q/How_do_you_convert_a_pressure_Pa_in_to_head_m

This $9 sensor has an accuracy of ~3Pa, which should get you sub-
millimeter resolution, if I'm understanding the physics properly.
That is probably more than you'd need and the data would need to be
averaged to smooth it out. I'm sure there are other inexpensive
sensors that would be a better fit; I just picked the first one I
found. Don't know about the arduino programming yet, though. I'm
also not certain if or how the diameter of the tube would affect the
pressure.

More equations/background: http://www.engineeringtoolbox.com/static-pressure-head-d_610.html

-Jack

On Apr 30, 1:42 pm, Tom Gondek <tgon...@hotmail.com> wrote:

Kevin Crowley

unread,
Apr 30, 2012, 6:05:17 PM4/30/12
to milwaukee...@googlegroups.com
I almost hate to suggest  this one.  It is almost too simple.
 
Put a tap in near the bottom of the barrel. 
Attach to the tap a clear piece of hose that is longer than the height of the barrel.
Tape hose to side of barrel.  Tape yardstick or similar measuring device to barrel next to hose.
Use Mark I eyeballs to measure periodically.

Mrspeed700

unread,
Apr 30, 2012, 6:25:25 PM4/30/12
to milwaukee...@googlegroups.com
Down and dirty is how accurate dose it half to be. most waist tank monitors are a series of well nuts spaced at measured intervals with resisters between, as water hits sensors resistance changes- just read change. I can get panels that read by thirds.
Rich

Joel Wakefield

unread,
Apr 30, 2012, 8:37:49 PM4/30/12
to milwaukee...@googlegroups.com
It really depends on how accurate you need to want you measurements to be, Tom, and if you are automatically filling/emptying this vessel.  I find that 3 float sensors set at about 10%, 50%, and 90% will indicate 4 distinct levels;  Low, medium, high, and OMG STOP FILLING ME PLEASE BEFORE I OVERFLOW!  This is usually pretty good for automated filling/emptying processes.  You could even calculate the amount of consumption based on the position of the sensor in relation to the volume of the vessel.  If you need really fine accuracy, you could add more float sensors since your accuracy can can be expressed by 55gallons/N sensors = x gallons accuracy, with N being the number of sensors installed in the tank.   Google for GEMS sensors, they are really good, and you may be able to score some samples from Digikey.  They should stand up to a salt water solution really good, however the wire leads may not.  I have 3 of them that you are welcome to try.

Jim Rawson

unread,
Apr 30, 2012, 8:58:02 PM4/30/12
to milwaukee...@googlegroups.com
If you are considering sensors, don't forget to adjust for the thermal expansion coefficient for the salt : water combination you are using.The specific gravity of "water" is around 62 @ 0 deg F and around 59 @ 212 deg f. You could probably get away with the avg, but measure the temp to at least 0.0001 deg f.

Jim Rawson

unread,
Apr 30, 2012, 9:14:08 PM4/30/12
to milwaukee...@googlegroups.com


On Mon, Apr 30, 2012 at 7:58 PM, Jim Rawson <taxi...@gmail.com> wrote:
If you are considering sensors, don't forget to adjust for the thermal expansion coefficient for the salt : water combination you are using.The specific gravity of "water" is around 62 @ 0 deg F and around 59 @ 212 deg f. You could probably get away with the avg, but measure the temp to at least 0.0001 deg f.  Newton's Second Law  

The density can be expressed as

ρ = m / V = 1 / vg         (1)

where

ρ = density (kg/m3)

m = mass (kg)

V = volume (m3)

vg = specific volume (m3/kg)

The SI units for density are kg/m3. The imperial (U.S.) units are lb/ft3 (slugs/ft3). While people often use pounds per cubic foot as a measure of density in the U.S., pounds are really a measure of force, not mass. Slugs are the correct measure of mass. You can multiply slugs by 32.2 for a rough value in pounds. In the SI system the mass unit is the kg and since the weight is a force - the weight unit is the Newton (N). Equation (2) for a body with 1 kg mass can be expressed as:

w = (1 kg) (9.807 m/s2)

    = 9.807 (N) (2b)

where

9.807 m/s2 = standard gravity close to earth in the SI system

As a result:

  • a 9.807 N force acting on a body with 1 kg mass will give the body an acceleration of 9.807 m/s2
  • a body with mass of 1 kg weighs 9.807 N
Or you could just look inside and see if the water is near the top or not.


Jim Rawson

unread,
May 7, 2012, 7:36:21 PM5/7/12
to milwaukee...@googlegroups.com

Shane

unread,
May 8, 2012, 1:01:12 AM5/8/12
to milwaukeemakerspace
I don't know if this has been mentioned, but could you just add a flow-
meter to the outflow tube? You could just reset it everytime you fill
the barrel.

Tom Gondek

unread,
May 9, 2012, 9:12:11 PM5/9/12
to milwaukee...@googlegroups.com
Thanks Jim.  This seems to working great – on a small prototype.  I’ll be trying it soon a the large 55 gallon drums!
 
Tom
 
 
From: Jim Rawson
Sent: Monday, May 07, 2012 6:36 PM
Reply all
Reply to author
Forward
0 new messages