# Float switch circuitry without floats?

Discussion in 'Electronic Basics' started by [email protected], Mar 11, 2008.

1. ### Guest

I know very little about electronics though I'm fairly handy and have
experimented with some simple circuitry. My objective is to build a
water level sensor for a 10 gal stainless steel kettle that's too high
for me to see the fill level. My idea is to build a panel using LEDs
to indicate how many gallons are in the kettle as I fill it. My
thought is to place a device(vertical stick) with bare terminals at
specific levels plus a common wire on the bottom. Thus as the level
increases, the water completes each circuit like a switch and lights
an LED indicating each gallon level. My concern is the resistance of
the water. Obviously it would be low voltage so I don't think I need
to be concerned about a shock. Do you think this could work?
Thoughts?

2. ### Jon SlaughterGuest

Something simple is just a tube with one in closed that hooks around out of
the tank for you to see.

The idea is that the water pressure will push water into the tube and
compress the air inside.

Its simple and no electronics needed. You calibrate it by pouring in a known
amount of water, mark on the tube the water level, then repeat with more
water. You can do this several times to get a "ruler" for the water level.
No calculations or anything else is needed. (although the air might be to
compressible and temperature dependent so you would use something else such
as oil)

If your serious about using electronics there are a half million ways to do
it. Another simple method is to use a long resistive rod with a float on one
end. The rod acts as a variable resistor and as the water rises it moves the
rod upward(or I suppose it could be turned upside down) shortening the
length of the rod between the float and a contact. Resistance is
proportional to height which is proportional to amount of water(which by
controlling the resistance you might get it matched well). You can hook up a
cheap ohm meter and use that as your guide. Probably cost just a few dollars
and very simple method. (although I doubt its very clear what I mean)

|
|
| |
| |
| |
|* |*
| |
| -
|
-

So - and * form a circuit and the part of the rod inbetween is a resistive
load - is at the water level and * is fixed(say at the top of the tank). As
the water rises it shortens the distance between - and * (which are
connected to an ohm meter or some other device) reducing the resistance.
Hence smaller resistances means larger height and vice versa(and its
linear).

Anyways, just 2 ideas to consider....

3. ### JamieGuest

Capacitive level probe.
the fluid is the dialectic. the container is one side of the cap
while the probe is the other.. Use it to alter a VFO that gets
converted to an analog voltage derived from the freq.
The probe should be coated with some thing to protect it like
teflon etc..

4. ### John LarkinGuest

Even easier: run a clear tube out of the bottom of the tank and up the
side. The level in the tube is exactly the level in the tank.

John

5. ### bud--Guest

..
It will work fine. You may need a transistor switch at each level (the
current may not be high enough to light an LED). You could use the
kettle as the common electrode. Using electrical conduction for level
detection is common.

6. ### petrus bitbyterGuest

I can't stop thinking
As so often, there are more ways to skin a cat and yours will work though is
has some drawbacks. The resistance of the water will not be constant. Pure
water has a very high resistance, so high it is considered an insulator.
Pure water is rare, so I think your water will have low enough a resistance
but it still can be pretty high. As you have to count for the worse
scenario, you will indeed need an amplifier (transistor) to make sure you
get a reliable signal. Another problem can be caused by the current.
Although the current will be very small, it can solve (part of) your wires.
This does not only depend on the current but also on the minerals solved in
the water. Not to forget the direction of the current. Nevertheless, I've
seen long poles of teflon with conducting rings on the outside and wires on
the inside used to measure waves and tides in the Northsea. I remember from
the explanation that they used special metal and AC-current for the
measuring to achieve an acceptable livetime in the aggressive seawater.

To make a long story even longer, I propose two solutions without
electronics.
1. Take a plastic tube and put reed contacts in it on selected distances.
Make a floater with a hole in it and a magnet glued on it. No need to say
you'd better seal both ends of the tube with some silicone -or other
sanitair kit. As you have some experience with circuitry you can fill in the
use of the construction. yourself.

2. Get some clear flexible tubing with a length over three times the height
of your kettle. Mount one end inside the kettle as close to the bottom as
possible.Gide the tube upward, over the edge, down to litlle below the
bottom of the kettle and then up again to end somewhere above the kettle.
Once you filled this tube fully with water, you can always see the fluid
level on the outside unless the kettle runs fully dry. If that's the case,
you can fool this meter by placing a little tin inside the kettle and make
the inside end of the tube end up in this tin.

success

petrus bitbyter

7. ### John FieldsGuest

---
Excellent idea except for the fact that the enclosed, compressed
volume of air will eventually diffuse into the water column,
rendering the device useless WRT to the initial calibration.

Oh, wait, I forgot...
I'm plonked, so you won't be able to read about how fucking stupid
come up with a better plan.

Oh, well...

8. ### John FieldsGuest

---
Did you miss this?:

"My objective is to build a water level sensor for a 10 gal
stainless steel kettle that's too high for me to see the fill
level."

9. ### John LarkinGuest

I interpreted that statement to mean that the kettle was too high for
him to see over the lip inside, without standing on a ladder or
equivalent. So, is the kettle actually so high that it disappears into
the ceiling?

At any rate, I made a suggestion that might work. If it won't work for
him, it's for him to decide. Your following me around, barking at my
feet like some crazed Chihuahua, is just silly.

John

10. ### John FieldsGuest

---
Oops, sorry... I forgot that:

"I reserve the right to make fun of anybody or anything."

applies only to you.

BTW, did you also miss:

" My idea is to build a panel using LEDs to indicate how many
gallons are in the kettle as I fill it."?

11. ### Bob MastaGuest

Don Lancaster's "CMOS Cookbook" has just what you need
(page 283 in my old 1977 edition). He calls it a "Booster Tank
Level Sensor" for fire engines. The tank is grounded, and there
are sensor electrodes at various depths, insulated from the tank.
Each electrode goes to the input of a CMOS inverters and is pulled
high by a large-valued resistor when there is no water covering it.
When the water hits it, it pulls the input low and the inverter
output goes high. (He shows the inverter driving an LED directly,
without even a series resistor. He might be relying on some
internal current limiting, or this might be a circuit error. You can
use the inverter output to drive an NPN driver through 10K to its
base, ground the emitter, and put your LED and series resistor from
the collector to the positive rail.)

Best regards,

Bob Masta

DAQARTA v3.50
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Scope, Spectrum, Spectrogram, FREE Signal Generator

12. ### John LarkinGuest

Oh, did you say anything funny? I must have missed that.
BTW, did you miss that my response was to JS's liquid-filled tube
suggestion?

OK, post a design that responds to the OP's needs, something a
beginner could put together successfully in a hour maybe. I have one
in mind that takes 12 parts total, and a probe he could assemble in 5
minutes.

All you seem to do lately is whine at other peoples' suggestions. You
probably aren't having fun.

John

13. ### Bob MonsenGuest

I would use a float switch instead. You could have the actual switch outside
of the pot, and have a float connected to a stick, attached to the side of
the pot. When the float gets to the fill line, the other end of the stick
connects to the switch:

Switch
1k
o------\/\/\/\----LED----GND
5V----o--__
o-
|
|
--------------|
| |
| | Float moves, causes switch to close
| |
| |
Inside Pot | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
.---.| Float
'---'|
|
Water Level ^ |
| |
| |
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

FWIW

Regards,
Bob Monsen

14. ### John FieldsGuest

---
Precisely.

When you make fun of someone it's seldom funny for them.

15. ### John LarkinGuest

Geez, you *are* really bummed out.

John

16. ### whit3rdGuest

Common solutions:
sight glass (a glass tube mounted alongside the
kettle, joined at the bottom to the kettle and open at the top).
Breakable, might need cleaning attention.

Float: a calibrated stick visible over the pot will tell you what
the float level is. Loose parts, need to lift it out to clean.

Bubbler: a tube with a slow bleed of gas; when you hear the bubbles,
it's got the end under water. Loose parts, easy to clean. Can use a
pressure

Ultrasonic level sensor: no contact, easy to keep clean. Sonar,
basically.

Capacitive dipstick: two close-spaced electrodes, capacitance goes up
when liquid fills the space. Capacitance is analog to liquid level.
Affected by liquid composition.

Capacitive sensor: two electrodes at a target position, capacitance
shifts
when liquid reaches that position. Capacitance gives over/under level
indication only. Can be combined with the sight tube. I've seen
these offered for chemistry labs, the exact liquid composition
doesn't matter much.

Pressure sensor/switch: depends on density of the liquid times depth.

Force sensor: ten gallons is HEAVY, hang the pot from a spring scale
(or put a scale under the stove).

A simple conductance sensor relies on electrical current in the
liquid,
which is unreliable (the electrodes can crud up even if the liquid is
a great
conductor).