I've spent about 8 hours of my own time trying to figure this out and for some reason I'm not getting it.

We have a customer that has several very large orange groves. The groves irrigation is supplied by a system of canals and pumps. We recently just got this account.

I do most of the control work for the company I work for so I got sent out to figure out why the auto setting on their motor control is not working. The system consists of 2 75hp motors driving two pumps, the system can be controlled manually or automatically. The way it WAS controlled automatically is in my opinion extremly hazardous so I will not put it back that way. The way it was controlled is as follows:

Four wires run from the control panel to the irrigation canal, they are only using three. Two of the wires carry 120 volts and one wire is the control wire going back to the control panel, these wires are placed in a 6" conduit placed vertically in the canal. The control wire has a large split bolt attatched to it, the two other wires are connected to these electrode looking things, they are using the water within the conduit to make the connection from the two hot wires to the signal wire. Obviously all of this stuff was home made, it's been this way for 30 years. The 120v starter is not energized unless both 120v leads are in the water, this threw me off so I put 120v to one wire and tested what I had coming back, it was 57v, due to voltage drop and the resistance of the water, this is why there are two 120v wires out there.

While the system is very hazardous in my opinion I also thought it was pretty ingenius

My problem is I need a way to energize the 120V coil on the starter when the water is low, deenergize when the water hits a certain level and not reenergize until the water is at a level between the low and high water sensors.

It's done this way so when the pump kicks off from hitting the high water it doesn't cycle on and off due to water turbulence ETC. I need to keep this as low tech as possable. I have found small float switches from Grainger but I am having major problems trying to figure out how not to kick the pumps back on until the water reaches a level lower than the high water level.

I also need to do the same thing on the feeder canal but this is to shut the pumps off in case of a low water situation so the pumps don't run dry and burn up.

I keep drawing it out with three floats and three relays but I'm just not getting it.

Try a single float switch and a Time Delay Relay set for however much time that is needed for the water to settle down.

What is the difference in height between low level and high level?

Mount the TDR in the control cabinet.

Larry C

Thanks for the reply Larry. I thought of the timed relay but being it's a canal, I have no way of knowing what the downstream usage is at that particular time or the near future. The low water is a few feet off the bottom of the canal, high water is four feet or so above that and I need the pumps to come back on again between the low and high, two feet or so lower than the high water.

If these are the float switches I am thinking about they have quite a bit of hysteresis. Once tripped on the water level has to drop several inches to get them to turn off or vice versa depending on whether you get the "over" or "under" switch.
I have 3 in my aerator tank on my well water system. You can change the hysteresis by making the cord a little longer or shorter from the pivot point.

Greg, Thanks for the reply, I'm familar with the float switches you are talking about, the problem is as I said above, I only have a 6" conduit so corded float switches are a no go, not enough room for them.

I think I may be able to use latching relays to accomplish what I want, trying to draw it out now.

You could hold the relay through a N/O point that is in series with an intermediate level switch and pick it with the high level switch

Totally lost me on that one, please explain. Are you talking about using a latching relay or regular relay?

Lone Gunman;

The most simplest and "Low Tech" approach to your Control logic is to set up Two (2) Float Switches as a typical "Start / Stop" Motor Control Circuit.

The "Start" Switch would be the Low Water Float Switch.
The "Stop" Switch would be the High Water Float Switch.

These floats would drive the coil of the Motor Starter directly, or indirectly through a Control Relay if needed.

This approach will eliminate the Pump from surging on/off, due to instable high Water levels.

Brief Details of the Control Logic:

The "Low Water" Float Switch is a Normally Closed switch.
When the Water level is at the "Low" set point, this float's contacts will close, and therefore energize the Motor Starter's Coil.

The "Low Water" switch is shunted by the latching contact (as would be found on a typical 3 wire Start / Stop control circuit), so the Coil remains energized as the Water level rises above the "Low" limit point.
This, in turn, opens the contacts on the "Low Water" Float, when the Water rises.

The "High Water" Float Switch is also a Normally Closed Contact switch.
This contact is in series with the "Low Water" contacts + the Latching Contact - as would be normal for the "Stop" Switch on a 3 Wire control circuit.

Until the Water reaches the preset "High Water" level, the float's contacts remain closed.

When the float is activated by the High Water Limit point, the conatcts open, resulting in the Motor Starter's Coil dropping out + opening the Latching Contact; thereby effectively de-energizing the Motor until the Water once again reaches the "Low" level point.

Only Two float switches are required. Each Float only needs one N.C. contact.

If you need a drawing, let me know.

Scott

Assume you want the pump to come on when the water is high.

I am talking about 2 switches N/O make on water rise.

you have a relay with a second N/O contact (might be a side switch on a big contactor)
Wire the upper water switch to the coil
Also wire the lower water switch to the coil but in series with the N/O side switch.
Water rises, bottom switch makes, nothing happens yet because contactor N/O is still open. Then upper switch makes, that makes the contactor. N/O switch is made. Water starts to fall. Top switch opens but the contactor is still held up by lower switch through the N/O. When lower switch opens the contactor drops out.

What about maybe adapting a 3-electrode type mV control that is often used for cooling tower make up water?
They normally have one common electrode, one to open the fill valve and one to close it off after the level is reached. You would need a good ground for this system to work and would also have to either cut off the electrodes at the height you needed or could try using the existing ones.
A contactor to control the pump motor would be needed since these aren’t really designed to handle motor loads. Baltimore Aircoil or Warwick Controls should still have these available.

Try these people: http://www.liquidlevel.com/products_switches_standard_vt_LS-14-180.asp

A customize plastic float switch assembly that turns on at 2 feet and turns off at 6 feet is approximately $300 to $400. The float diameter is about 2" OD.

Larry C

Scott, a drawing would be great if not too much trouble, I'm usually pretty good with control wiring but for some reason this problem is confusing me.

Greg, I need the pump to shut off at high water, I guess high water is not really the right terminology as it's not a high water condition it's where my water level needs to be.

Kjay, I have never seen an electrode system, I bet thats what was here at one time but they are using the electrodes to carry line voltage now.

OK no problem, you just need to use low water "on" switches and wire the bottom one to the relay coil, and N/O side switch through the upper one to turn it off.

http://esteroriverheights.com/electrical/switch.jpg

Years ago I installed a 120 volt electrode system. It used an isolation transformer. Maybe the isolation transformer is missing on your system.

http://www.gemssensors.com/content.aspx?id=2258

LG,
Go to this site and download the instruction bulletin and PDF. I'm used to dealing with them in the differential pump down mode (F2, top row). Your application would require the pump down or up (G2, middle row), or pump up mode (H2, bottom row) mode. I would normally use the H2 for your app but you said there is another pump location that you pump down. So it is the choice of H2 with a spare contact or G2 which goes both ways.

The conduit serves as a still well besides protecting the probes. In your case, the pump contactor(s) would be energized through N.C. contacts. When the water reaches the high level probe, current from the secondary of the integral transformer, flows through the water between terminals 9 & 10. This energizes the relay to open the N.C. contacts and stop the pump(s). Since a jumper is added between terminals 8 & 10, current will flow between 9 & 7, through the Normally Open, Held Closed contact to 8, through the jumper to 10. This will keep the relay energized and the pump(s) off until the water clears the low level probe.

I just offer this description should you decide to stick with the original method. I don't want you to think that it is a dangerous, home brew technique. I have them in 2 locations where an incredibly stupid bubbler/pressure transducer scheme is used as primary control. Some designers have a little too much time on their hands.
Joe

What you really need is a RADAR tank level indicator mounted in a pipe. Sure, it's probably 2 figures more than the other solutions and you don't really need to know the canal level to that high of a precision, but it's also really cool

Joe, yup thats what was there (well I feel like an idiot now) but they were made my magnatech, they were both burned up (power company had a major problem in the area). I showed the thing to 8 different electricians and 3 supply houses, nobody knew what it was except it being a contactor and transformer in one. The existing one had 120v as primary and 240 as a secondary voltage. So these controls are safe when they are putting 120v into water? It just doesn't sound right.

LG,
I'm glad I was able to help. Believe me, it took me several minutes to wrap my brain around how they worked when I first encountered them.

Now let's take a look at the safety angle since it bugged me too. If you have it contained in a grounded conduit / still well with a grounded lowest electrode, you're grounded! So 1 end of your isolated and current limited secondary is now ground referenced. Once the water reaches the highest electrode, you now have a voltage divider of the water/probes and the relay coil. I won't pretend to know the voltage gradient around that conduit and probe. You could short out terminals 9 & 10 with your meter on AC Amps to determine maximum current the system can deliver. Of course, the current through the water and probes would have to decrease as the voltage drop across the probes increased. Your worst case there would be if you jumped in the canal and grabbed the grounded and hi level probes while the level was low. (See posts in other areas pertaining to "Darwin Awards" and "Natural Selection"!) If our typical circuit current is less than what would trip a GFCI, no worries! If we figure there would be enough shunt current available to hurt somebody, post a sign...
DANGER!!! NO SWIMMING IN CANAL!!! ELECTRIC EELS!!!
Joe

Here is the "Start / Stop" Control Schematic:



I included a Control Transformer - to step the Control voltage down to 120V or whatever.

Additionally, I included a HAND-OFF-AUTO (HOA) Switch, just for fun.

Sorry it took so long.

Scott