12 replies to this topic

[MobileMe]

I've been working on a way to monitor whether or not my sump pump is running. I thought about using a CT sensor to monitor the power usage, but turned out to be more hassle than it was worth. One of my challenges is my sump pumps don't have floats and are switched on internally with a digital sensor. I thought about a vibration sensor, but with the amount of water that comes in, I think it will trip when it is not running.

I have a webcontrol board that monitors high water levels, but I have had a problem the sump pump getting stuck and cavitates to death. Also one of the pumps is outside and getting the wiring inside will not be worth the effort.

I'm stuck and don't have any more ideas. I've googled this problem to death. Any thoughts where to go from here? I have a hard time thinking that no one has figured this out yet. Any help pointing me in the right direction would be helpful.

Thanks

[jdale]

With Insteon, you could use a SynchroLinc to see when it is drawing power. Assuming it is plugged in (like mine).

[BraveSirRobbin]

If you simply want a contact closure when the pump is running you can use one of these current sensors. Why did you give up monitoring current with your webcontrol? I would have thought that to be the most reliable method.

[Work2Play]

What could be better than a current sensor to detect when it's running?

[MobileMe]

I was attempting to make this an affordable project. I don't have that much money to sink into a couple of current sensing relays. I thought I could make this current sensor work with the webcontrol board, but failed epically.

[BraveSirRobbin]

I glanced at that and I'm thinking that provides an AC output, which I doubt your board can measure.

There are cheap CT's that produce AC that can then have a circuit to convert it into usable DC.

An example (though I have no experience with this model) is shown below:

http://www.crmagneti...del-19-P81.aspx

http://www.crmagneti...onditioning.pdf

Edited by BraveSirRobbin, 03 April 2012 - 10:19 AM.

[MobileMe]

The PDF is very close to what I was trying to do. I have to say this stuff could be in greek for all I know. I'm a computer and networking guy by trade. This stuff is way over my head. Hooking up some simple sensors with a resistor or two is one thing. Getting this sensor to work the way I want it to is entirely different. Right about now I wish I had a degree in electrical engineering. The math and language is killing me. This is like learning to read another language. Every other word I have to get translated and then translate the translation. It gets very frustrating.

I have found several sites with this basic idea and they all have different wiring diagrams.

Is there a website that I could plug in some numbers and it would tell me what parts I need? That would be awesome! Probably not though.

Thanks for the help.

[Lou Apo]

You can make a current sensor cheap. You simply take some fine lacquered wire and wrap it around the hot wire going to the pump (you need separate hot/common) OUTSIDE of the insulation. In other words, do not strip the insulation of the hot wire. After wrapping it a bunch of times, take a multimeter and turn on the pump and check the voltage on the lacquered wire. You could probably use a single strand of cat5 wire instead.

When the hot wire to the pump is passing current, it will induce a current in the fine wire wrapped around it. With more wraps, you will get more potential. Also it works better the tighter it is wrapped and the thinner the insulation. Just keep adding wraps until you get enough volts for the cai to consistently recognize. You'll need a diode to rectify the AC current since CAI needs DC. The diode will cut the voltage down by about a volt. If you use an analog port on CAI you could probably get away with less than 1 volt as a consistent trigger for the cai. You might be able to build this for less than $10.

[BraveSirRobbin]

I've also wanted to play around with these inexpensive current switches, NPN transistor output. Should be able to trip a cheap relay.

[Lou Apo]

I've also wanted to play around with these inexpensive current switches, NPN transistor output. Should be able to trip a cheap relay.

That is basically a prefab version of what I was talking about making. It is better than what my home made thing would be (depending), but since he already has a cai which is quite good at detecting the smallest amount of induced current, buying anything would be overkill. In fact, I bet he could make my setup for free in about 20 minutes with scraps sitting around the house. Probably a few feet of a single strand of cat5 and a diode is all it would take.

[MobileMe]

I would like to focus on the CT sensor I bought. The good thing about it is that it can clamp over the wire so I don't have to hack and splice. I think I'm getting somewhere with it. I found this site which looks promising. I just have to do some math and figure out what the burden resistor and capacitor size should be. Hopefully I'll get lucky and figure this out.

[Lou Apo]

I would like to focus on the CT sensor I bought. The good thing about it is that it can clamp over the wire so I don't have to hack and splice. I think I'm getting somewhere with it. I found this site which looks promising. I just have to do some math and figure out what the burden resistor and capacitor size should be. Hopefully I'll get lucky and figure this out.

A "clamp on" style sensor, mutch like a multimeter with "clamp on" current sensing, still requires that only one side of the wire go through. In other words, you still have to separate the hot wire from the ground/neutral and only clamp on the hot wire. A home made current sensor is the same way.

There are two styles of wire:

1) The style where hot, neutral, ground are side by side with a little groove between them where you can peel them apart from each other without breaking the insulation on each wire (except ground). In this case you just use a blade to break apart the connection in the groove to free up the hot from the others while still keeping it insulated.

2) The other style is where there is an outer jacket of insulation with the other wires spearately insulated inside it. In this case you need to remove a section of the outer insulation to free up the individual (but still insulated) wires.

If you don't want to mess with the power cord, you can get a short extension cord and do these things to the extension cord.

Whether you use a clamp on or home made, the wire does not need to be cut and spliced, but you do have to separate as above. Of course you could also open up the pump where the power cord enters it. At this point the wires will be separated and you can clamp on there or do your home made wrap there.

Edited by Lou Apo, 04 April 2012 - 12:24 PM.

[fester]

Old tech, but I've had experience with em'...if you could use a simple contact closure to monitor, use a "current relay". To whit here's a link to one on ebay which maybe be just what you need (the relay coil is a heavy winding in series with the load).

http://www.ebay.com/...#ht_1079wt_1321