Conductive liquid level probe for water. I need some assistance
Conductive liquid level probe for water. I need some assistance
(OP)
Hello. I'm having a controller developed for my process, a solar thermal collection tank (a tank full of hot water).
The liquid level is controlled using a high and a low sensor. I am using conductive probes (no moving parts and relatively inexpensive); however, all commercially available probes are usually either switches or really large gague probes with very expensive junctions.
We mount our temperature sensors at the end of a threadded PVC pipe cut to length and submerged into the tank. We would like to do the same with the conductive level probes. Here is a rough schematic:
[img]h ttp://h1.r ipway.com/ stabmaster /conductiv eprobe.JPG[/img]
the probe might be threadded on the dry side. This side may have terminals, a plug, or wires coming out. The submerged side may have probes, connectors, nobs, or anything that will conduct electricity and is connected to the wire side.
Design considerations:
I would like to find something off the shelf. I don't really care what it is intended for: metal on both sides and some pipe thread in the middle should suffice.
I have decided that AC voltage would prevent electrolysis, so the gague of the metal is not as important as it will not be losing mass.
I need a corrosion resistant metal because it will be suspended in moist air much of the time. Is stainless steel the only reasonable choice?
I am going to use a pull-up resistor for the circuit to detect that the probe is on-line when not submerged. I do not know yet how to figure in the resistance due to the unknown "probe constant" ("k value") of this yet to be discovered probe.
Domestic water may have a resistivity of, say, between 5-500 microsiemens per centimeter.
Probe constants vary from 0.01-50 "per centimeter".
Therefore, the expected conductance ranges from .05-25,000 microsiemens.
In general, the higher conductivity of the fluid, the higher probe constant required. Since domestic water is not considered "highly conductive", I'll consider probe constants from 0.05-1.0, for argument sake.
This translates to a resistance range of 2kOhm -4000kOhm.
On a 24VAC circuit, this translates to 0.006mA to 12mA. Does this make sense? I am rusty on my electrical circuits.
Power requirements would be 0.024 mW to 24 mW.
So the proper pull-up resistor must be on the order of, say 40MOhm? Say, an order of magnitude greater than the greatest resistance expected from the conductive probe.
Will there be a problem with EM interference? Should the pull-up resistor be mounted in the pipe (in the tank), or in the controller? I don't know enough about this to make the call myself. The tank is mostly rubber and foam with very little aluminum on the outside. A copper pipe may be used if EM interference is a problem?
Now this leads to design considerations of the probe. The probe constant may be on the larger end (say near 50) so that the resistance is 2-20kOhm thus decreasing the sensitivity to interference. Is this true? I don't generally have any clue as to the design parameters of a sensor. I'm just using the knowledge i've gained from a course in electrical circuits and no intuition. Please help me gain some intuition on the matter.
The liquid level is controlled using a high and a low sensor. I am using conductive probes (no moving parts and relatively inexpensive); however, all commercially available probes are usually either switches or really large gague probes with very expensive junctions.
We mount our temperature sensors at the end of a threadded PVC pipe cut to length and submerged into the tank. We would like to do the same with the conductive level probes. Here is a rough schematic:
[img]h
the probe might be threadded on the dry side. This side may have terminals, a plug, or wires coming out. The submerged side may have probes, connectors, nobs, or anything that will conduct electricity and is connected to the wire side.
Design considerations:
I would like to find something off the shelf. I don't really care what it is intended for: metal on both sides and some pipe thread in the middle should suffice.
I have decided that AC voltage would prevent electrolysis, so the gague of the metal is not as important as it will not be losing mass.
I need a corrosion resistant metal because it will be suspended in moist air much of the time. Is stainless steel the only reasonable choice?
I am going to use a pull-up resistor for the circuit to detect that the probe is on-line when not submerged. I do not know yet how to figure in the resistance due to the unknown "probe constant" ("k value") of this yet to be discovered probe.
Domestic water may have a resistivity of, say, between 5-500 microsiemens per centimeter.
Probe constants vary from 0.01-50 "per centimeter".
Therefore, the expected conductance ranges from .05-25,000 microsiemens.
In general, the higher conductivity of the fluid, the higher probe constant required. Since domestic water is not considered "highly conductive", I'll consider probe constants from 0.05-1.0, for argument sake.
This translates to a resistance range of 2kOhm -4000kOhm.
On a 24VAC circuit, this translates to 0.006mA to 12mA. Does this make sense? I am rusty on my electrical circuits.
Power requirements would be 0.024 mW to 24 mW.
So the proper pull-up resistor must be on the order of, say 40MOhm? Say, an order of magnitude greater than the greatest resistance expected from the conductive probe.
Will there be a problem with EM interference? Should the pull-up resistor be mounted in the pipe (in the tank), or in the controller? I don't know enough about this to make the call myself. The tank is mostly rubber and foam with very little aluminum on the outside. A copper pipe may be used if EM interference is a problem?
Now this leads to design considerations of the probe. The probe constant may be on the larger end (say near 50) so that the resistance is 2-20kOhm thus decreasing the sensitivity to interference. Is this true? I don't generally have any clue as to the design parameters of a sensor. I'm just using the knowledge i've gained from a course in electrical circuits and no intuition. Please help me gain some intuition on the matter.





RE: Conductive liquid level probe for water. I need some assistance
RE: Conductive liquid level probe for water. I need some assistance
RE: Conductive liquid level probe for water. I need some assistance
They are used for level control with deep well submersible pumps and for water storage tanks. The control circuit is an adaptation of the old three wire motor control circuit, so with two probes and a return path you can set a high level and a low level. Turn on and turn off.
The return path may be a metalic pipe supporting the submersible pump, a metalic well casing, a metalic tank or another electrode.
respectfully
RE: Conductive liquid level probe for water. I need some assistance
RE: Conductive liquid level probe for water. I need some assistance
Dan - Owner

http://www.Hi-TecDesigns.com
RE: Conductive liquid level probe for water. I need some assistance
What do you think of this circuit? Unfortunately 120V is not going to be a solution for us because we decided 24VAC max on tank side for UL and building codes.
So my question is if my theory is sound. I can easily average out values; else I can use a switch. I can't find a commercial switch that will thread into a pipe as i've described.
I've found a switch for $14 but it's apparently prone to failure.
So should i go with a switch or a (passive) sensor? If i do go with the latter would it require the pull-up resistor in the pipe or on board? Is it a bad idea due to EMC effects?
I could not find any cheap water detectors. I only found the circuit linked to above. Any ideas? Thanks so much.
RE: Conductive liquid level probe for water. I need some assistance
http://www.eng-tips.com/viewthread.cfm?qid=58310
Another one:
http://www.eng-tips.com/viewthread.cfm?qid=174309
Whatever you decide to do, I think you should provide a backup circuit.
RE: Conductive liquid level probe for water. I need some assistance
RE: Conductive liquid level probe for water. I need some assistance
http://
These are common manufactured items for doing what you want to do. Note that they accept wires or rods on the wet side.
I have used spark plugs with NPT threads and entered thru the side of the tank, I have also used stainless steel bolts thru the side of the tank, with a rubber washer on the inside. In these aplications the levels were not adjustable and the wiring was exposed.
Actually "weighing" the water is rather cheap. Many times I have dropped a presser sensor into a well (just yesterday) or tank and have gotten the water depth. Or you could tap the side or bottom and install a threaded or hose connected pressure sensor. Some real cheap PCB mounted devices are availablr for sensing air pressure. These could be mounted above the water level and connected to the bottom of the tank with tubing, or mounted to a probe pipe inserted from the top of the tank. When the tank fills, the air in the connection/pipe is compressed. Water level is calculated as 0.433 PSI per foot of water depth above the sensor, in water connected applications and 0.433 PSI/ foot +/- above the air connection.
Steve
RE: Conductive liquid level probe for water. I need some assistance
There's one of these in my living room. It works exceptionally well.
Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com
RE: Conductive liquid level probe for water. I need some assistance
The Rols royce used to use a pair of gold plated 3mm wires about 70 mm long protruding into the tank, a box of transistors swithched on the lamp, it seemed to be a DC system.
The Merc washer fluid sensor uses a reed swith and a float with an actuating magnet attached inside the float.
RE: Conductive liquid level probe for water. I need some assistance
I still have yet to find a solution that costs less than, say, $100 or so. This one:
h
is perfect except (1) it has been discontinued and (2) the threads are on the wet side. At $14 per unit, a switch with a visible LED works perfectly.
Also, I was hoping to utilize these as flow switches as well, in order to detect if flow is at a point in the return pipe. If all the sensors are the same, than it is easier to develop a controller and all of the design considerations reduce by one degree of freedom.
The search continues. Thanks so far! Is it safe to say that a switch is a smarter way to go than a sensor for the extra few bucks?
RE: Conductive liquid level probe for water. I need some assistance
http://
$3.72
http://
This one seems like the right price! Unfortunately, the temperature limit is 80C and I need around 95C
RE: Conductive liquid level probe for water. I need some assistance
$143
too bad. This looks perfect. I expected a price somewhere closer to maybe 10% of that.
RE: Conductive liquid level probe for water. I need some assistance
Do you have the model and serial number of the pressure transducer you used for under $100 as i am having difficulty finding one that cheap.
Thanks
Justint
RE: Conductive liquid level probe for water. I need some assistance
Cheap packaged low-pressure sensors for level sensing. Should be well below $100 in quantity.
RE: Conductive liquid level probe for water. I need some assistance
The best low cost and reliable solution is a mercury tilt switch. It requires no electronics, so it wins on simplicity.
The trick is to get it to tilt at the correct water level, and that can be arranged several ways. For an enclosed tank you need a float on the end of a descending cable.
If the tank is low, the float hangs vertically on the end of its electrical connection cable.
If the tank is full the cable goes slack and the float floats. The trick is to arrange the float so it inverts when buoyant. At the float tipping point, the mercury in the glass bubble makes or breaks the electrical contact.
By feeding in more or less cable through a top gland, the water sensing level can be very easily changed. This has worked for me, and is simple, versatile, cost effective, and reliable.
RE: Conductive liquid level probe for water. I need some assistance
Did not know the budget was under $100
Give me break! I thought this was an engineering forum. Go to a handyman forum and let them tell you to get a sump pump switch.
RE: Conductive liquid level probe for water. I need some assistance
http://
RE: Conductive liquid level probe for water. I need some assistance
Humidity will be 100% in there, and there is bound to be plenty of condensation. While touch sensitive capacitive switches work fine, try dipping one into a bucket of water, and see how well it works afterwards.
Not saying the idea won't work, only that it may have some significant long term reliability problems.
A float with some sort of dependable electrical detection is still the cheapest and most reliable. A magnet and reed switch is one way, another is a mercury tilt switch. Cheap, simple and reliable.
RE: Conductive liquid level probe for water. I need some assistance
http:/
On page 13 is a chart that will give you an idea of resistivity of common fluids.
The advantages of using a packaged relay would be simple sensitivity adjustment in the field and a built-in adjustable time delay for wave compensation.
The level control relay will cost you about $75 for one-off, the probes around $50. A lot less in qty.