Hello. I have been searching the net trying to find a solution and this site has given me the most info out of all of the rest. I am now at a point where I need a tad bit more info, but I cannot seem to find it for my application or design. I am sure if I knew more about things I could adapt a higher voltage circuit, but I am not that knowledgeable.

Without going into too much unneeded detail, I would like to be able to convert 24vdc(or even 12vdc) to somewhere between 6-12vac. The reason is that I have a device I am building that measures water level of a tank. It needs to generate a 4-20mA analog signal, but also has to use AC power for the contact probes so that electrolysis does not happen. If I knew of a way to keep that from happening with DC power then I would just use it.

I have an idea of how to make it happen using 5 wires, 2 for AC power, and 3 for DC power/signal, but if I could convert the DC power in the relay box then I would just have to run 3 wires from the control cabinet where the analog signal is connected to the PLC.

This is the circuit I am referencing.

Except instead of the LED's, I am going to use a photomos relay. (This is my theory anyways) I am not concerned with the 4-20mA signal part as I have already built and have it functioning correctly. I just need to be able to energize/de-energize the relays based on water conductivity, and using AC seems to be about it unless I want to constantly clean/maintain probes using DC power.

Thanks for the help in advance, I hope I got enough detail in there for you guys.

 

This is a several stage project. Maybe you should change the title to something like, "water level monitor". It would fit right in with a half a dozen other water level monitors we have done here.

ps, Have you tried searching this site for, "water level"?
Your first move was to ask for an AC circuit, when what you need is a water level monitor by the conductivity method.

Anyway...I'll start with a block diagram.
Your low voltage AC is going to be a square wave centered around half (or less than half) of the power supply voltage. A simple oscillator will do it. Drive 3 or 4 or 5 sensors, measure the results, convert to a 4ma-20ma loop driver, and you're done.

Got you started?

 

To keep things simple, at least the way I view it, I was going to divide the mA signal by the number of level points I need. Then, set each relay to conduct a certain resistance.

Like I said, I have the signal portion down so really I am not looking for a water level monitor. I have searched for that before I came to my conclusion and there are none I can find that are applicable in my case. It very specifically has to be 4-20mA signal, but again, I have that part figured out. Just need to figure out how to energize the coils AND have alternating current for the probes with only bringing in 24vdc.

After I posted this, I have done some more searching and think I may have solved my issue in a different way. I am not familiar with the components they are using, but it effectively does what I want. Do I need to start a new thread to ask for help with the components? I guess given enough time with google I could probably figure them out.

It alternates the DC power so that the probes will not corrode. I just am not familiar with the U1A and U1B components. The rest I am sure I can figure out/already know. Am curious though, if I had 24v instead of the listed 12v, would I double the component values?

 

I am not looking for a water level monitor.

Well, I certainly misunderstood.

I think we have a language problem. I say you need a bi-polar oscillator and you say, no, I need it to alternate the DC to the probes.

Right now, your drawings are not visible, so I will have to wait to see what you have.

 

Without going into too much unneeded detail, .

Maybe detail is needed?
You appear to want to detect 5 water levels of a tank, to make it simpler and avoid the electrolysis problem with DC, would it not be simpler to have a tubular float switch that has 5 simple prox detectors in a immersed tube?
The 4-20ma conversion would be ancillary or whatever means you want to transmit the data?
Max.

 

As drawn in post #1 I don't see this working well. You can't just apply 6 VAC to those transistors. Also, when using level sensing probes in that fashion, even AC pulsed, they will still corrode and how fast is a function of how pure the water actually is.

If your goal here is to end up with a 4 to 20 mA loop representing the water level what you have is not the way to go about it. I would leave the 24 VDC alone if you have it. I would get a 0 to 1 or 0 to 2 PSIG pressure gauge. I am in the US so you get PSIG. 1.0 PSIG = 27.7 inches of water column, 2 PSIG = 55.5 inches of wat5er and so forth. Read this overview to understand where I am going with this.

Choose a sensor with a 4 to 20 mA output, they are fairly common. That is how I would likely go about it. Then scale your display or uC accordingly.

Ron

 

Maybe the TS was tried shows the circuit as below.

Avoiding Electrolysis For Water Level Detector Probes Using Alternating Current Detection.

 

Thanks for the replies guys. Very, very valid comments.

Ok, here is the detail. I work at a water treatment plant. Reservoir water source, not waste water source. Near the end of our treatment process we have filter beds. During a wash cycle, there are 6 different level points that determine when certain valves open and close. Two of those points are very specific in that the water needs to be no more than a few inches above or below that point or else we can loose quite a bit of expensive filter media, or not get a good wash on the filter.

We have tried different ways of addressing this process. Submersible pressure transducers and ultrasonic transducers would not work because of the turbulence caused during the wash cycle(the actual part that we need them to work at the most). We have tried different brands, all with the same result. Currently, the only thing we have found to work is a radar unit. Of course, as you could suspect from the name, they are not inexpensive in the least. Float switches were an option, but the fact that this is a addition to an already in place system, made it a no go for adding that many wires into the existing electrical cabinet.

Now, I have thought of using float switches at the various level points to function as noted above. However, there is no way of knowing if they will survive the turbulence of the wash cycle. They may, they may not. The only thing that has been in place since day one and is still going strong is a single contact switch we have at the very bottom of the filter to indicate absolute low level(we always want water to cover the filter media).

So, this is why I would like to build a rod that has various contact points, that will in turn control different relays, which will have different resistance values, giving different mA current, utilizing the already in place wiring. Therefore, while all the above suggestions are nice, they just wont fit into my application. I know they seem like they do/would, which is why the engineers used them, but when it comes to real world function sometimes things just don't work.

I apologize for the last picture not coming through. Here is a link to it if its not against the rules of the forum to post it. I think that it will be exactly what I am looking for, but I am a water technician, not a electronic guru. I can see you all are though, so I am hoping this will not be hard to accomplish. http://freecircuitdiagram.com/wp-content/uploads/2008/08/ac_water_level_detector.gif

 

Maybe the TS was tried shows the circuit as below.

Avoiding Electrolysis For Water Level Detector Probes Using Alternating Current Detection.

I didn't even click on your link till after I posted. yes, this is the exact thing I was looking at. I can reduce the voltage we currently have to 12v, but would like to use 24v as that is what power supplies we currently use for the transducers in place.

I also am deciphering the different components, and while I have built circuit projects in the past, I have never used the U1 4093 type component and am just not sure how it works and all. The rest of the listed components in this circuit I am familiar with and would have no trouble building.

 

It is a square wave generator → full wave rectifier voltage doubler → inverter(buffer) → bjt driver → relay

 

Yea, that goes over my head. My terminology is not correct, nor do I understand the specifics of how things work. Not that I am beyond learning, but it takes a while to sink in.

I have decided to just put in a 12v regulator and use the circuit as is. That way, I will be able to just use the current 3 wire setup (24vdc, 0vdc, and 4-20mA signal) and just regulate the voltage down for the relay circuit in the control box I will mount by the filter beds.

I found the spec sheet for the component I am not familiar with. A "Quad 2-Input NAND Schmitt Trigger". I see how to put it into the circuit via the diagram, but the spec sheet also lists two pin assignments that the circuit diagram does not address. Those would be Vdd and Vss. Are these just standard pins that are connected straight to the power source in order for it to function? Or do they not get put into the circuit at all.

I know that may sound real dumb, but I seriously have not ever messed with these before so I have no clue. Go easy on me please.

 

Now, I have thought of using float switches at the various level points to function as noted above. However, there is no way of knowing if they will survive the turbulence of the wash cycle. They may, they may not.

The type of float switch I had in mind would be totally protected and the switches themselves not exposed to water.
The idea would be a inner/primary tube that carried the prox or reed switches at the various levels, not exposed to water, around this tube would be a single float ring that carries the magnet, this float would be surrounded by a larger tube exposed to the water levels, but protected from turbulence etc.
Max.

 

The type of float switch I had in mind would be totally protected and the switches themselves not exposed to water.
The idea would be a inner/primary tube that carried the prox or reed switches at the various levels, not exposed to water, around this tube would be a single float ring that carries the magnet, this float would be surrounded by a larger tube exposed to the water levels, but protected from turbulence etc.
Max.

I see what your saying. In theory it could work. So long as the float did not get stuck with filter media, or algae, or frozen during the winter months(the filter beds are outside exposed to the elements).

I do like the idea, but versus a known method that works, its hard to get others on board for a test run.

 

OK, then with all things considered I would use the approach Max has suggested. Reed switches with a moving magnet on a float. The outer pipe serves as a baffle so waves or sloshing in the tank won't effect it. Will work fine with the existing 24 volts too.

Ron

 

We are not talking about just sloshing here. It is very turbulent air washing, enough so that it would ruin every industrial type submersible transducer we tried. Which brings up another point. In order for this to work it would have to be free standing away from the filter wall. That would mean the pipe itself would have to be rather large/rigid, then have to figure a proper way to secure it from the top.

Also, the float would have to be heavy enough that the air washing did not falsely lift it up. Then you run into the problem with things that could potentially lodge between it and the pipe, causing it to get stuck at a high level or low level. Also, during the winter, it could freeze to the pipe when the filters are not in use, and have to be thawed when they came back online.

This would be a perfect idea for a standard water tank level, but we are not dealing with a standard water tank level.

 

We are not talking about just sloshing here. It is very turbulent air washing, enough so that it would ruin every industrial type submersible transducer we tried. Which brings up another point. In order for this to work it would have to be free standing away from the filter wall. That would mean the pipe itself would have to be rather large/rigid, then have to figure a proper way to secure it from the top.

Also, the float would have to be heavy enough that the air washing did not falsely lift it up. Then you run into the problem with things that could potentially lodge between it and the pipe, causing it to get stuck at a high level or low level. Also, during the winter, it could freeze to the pipe when the filters are not in use, and have to be thawed when they came back online.

This would be a perfect idea for a standard water tank level, but we are not dealing with a standard water tank level.

I think we understand your desire for the AC, but what is confusing is how the method you show in your first post gets around the turbulance.

 

I think we understand your desire for the AC, but what is confusing is how the method you show in your first post gets around the turbulance.

The way it gets around it is the only thing exposed to the water are contact probes. Air will not disturb them, they would be flush with the surface of the pipe so that the pipe could be attached to the wall in multiple places, and they have no moving parts.

In my first post I was trying to make a different circuit I found work using just DC power from a different location - so I did not have to run another cable carrying AC power.

Now that I found a circuit that will allow me to just use DC power I am good to go.....just as soon as I feel comfortable with how the circuit works component wise.

I really do apologize. I am sure that my level of understanding is WAY below that of most of you all on here. I would venture to guess that you may feel like your talking to a 8 year old kid in trying to explain things. But that is the reason I figured you all could help. I would rather be talked over my head than way under.

I get how simple circuits work, and sometimes I know the answer to a lot of my questions but its stuck in the recesses of my mind and needs help to bring forward.

 

During a wash cycle, there are 6 different level points that determine when certain valves open and close.

I've been working on a probe and circuit design.
How far apart are those levels?
What is the approx peak-to-peak amplitude of the turbulence/ripples?
What is the approx electrical conductivity of the water you're treating?
What are the max and min water depths in the filter?

 

I've been working on a probe and circuit design.
How far apart are those levels?
What is the approx peak-to-peak amplitude of the turbulence/ripples?
What is the approx electrical conductivity of the water you're treating?
What are the max and min water depths in the filter?

The total distance will be around 16'. The distance between the levels will vary from 2 - 6 or more feet.

Peak to peak amplitude is way beyond my pay grade. lol. I would have no idea of where to even start measuring that.

The conductivity I have no idea of, but it will be average. Its not extremely dirty, nor is it super clean. You could say that if you were on a lake and went out a few feet and scooped up some water then that's what you'd have. I have measured resistance before, but I do not remember what it was.

 

Peak to peak amplitude is way beyond my pay grade. lol. I would have no idea of where to even start measuring that.

But you can estimate it? If it's greater than the 2ft interval (hopefully not) it complicates distinguishing one level from another reliably.