Ice detector

[Engineer1138]

Since there's a Water Boil Detector, I figured I'd post a thread here about a subject near to my heart: an Ice detector

Winter's almost here in MN and soon the outside water trough will have a floating heater in it to keep the surface from freezing. Occasionally it gets unplugged and ices up before anyone notices and then it takes an axe to get the ice out. Ever been hit by flying chunks of ice when it's 10 below zero? Hurts!

Has anyone built an ice detector? I researched a few ideas used for detecting bridge and aircraft icing, but I want something SIMPLE. It's not enough to just detect water temperature, I need to detect actual ice. Right now I'm thinking of an optical method that uses the difference in clarity or refractive index to tell water from ice, but outside in bright sun = difficulties in detection, so before I go there I'd like to hear of better ideas if you guys have any.

[zotdoc]

My truck has an outdoor thermometer and when conditions are right, it reads "ICE" instead of the temperature. There must me some chip out there that will make the circuit you want easy. Try the Maxim site.

[MrAl]

Hi there,

My suggestion is very simple.
Since your heater becomes unplugged you can use a current transformer
to monitor the current getting to it. If the current decreases by too much,
it sounds an alarm. Would be easy to program a chip to do this or
just use a comparator chip like LM339 for example.

I have done this with several things already, including an electric
coffee pot that shuts off automatically but doesnt have an indicator...
when the water is hot the coffee pot shuts off and the current stops
and the LED connected to the current transformer goes out.
Simple and reliable and it doesnt have to be bothered by the
weather outdoors as it can all be mounted inside the house.

BTW, this would also detect if the element stopped working because
the element opens up and that also reduces the current. Use a buzzer
instead of an LED and have it buzz when current goes to some low
value below the normal low line level current.

[jwax]

A real project would be to try the ultrasonic route. Should be possible to differentiate the pulse echo of water vs ice. Or, maybe even the speed of propagation.
For real simple, imagine a closed metal cylinder on the water surface, with a rubber diaphram closing off one end. Mount a touchy microswitch against the diaphram so that expanding water, as it freezes, closes the switch.

[MrAl]

Hi again,

jwax:
I was originally thinking something similar with the pressure increase
of the ice when the water freezes, ie use a pressure sensor. Then
i realized when he has the problem the heater doesnt draw current
so i suggested the current sensor.

[philba]

maybe I don't understand the problem but why isn't reading the water at a specific temperature, like say 0C, a good indication that ice is forming. your heater must be keeping the water above the freezing point.

Or maybe tracking the temperature and when the water falls below a predetermined value, raise the alarm.

[MrAl]

Hi again,

philba:
Yeah now that you mention it, that makes a lot of sense too
If the water temp falls below 2 deg C then it going to be freezing
soon (hee hee). If it's above 2 deg C it's not going to freeze.
Dont know how big the area is though, as sometimes the wind
evaporation could freeze some parts before other parts, so
maybe multiple thermistors? I dont see how this could be easier
than the current sensor however, unless he wants to save energy
and only have the heater come on when it's needed, which would
be very very nice i think...less energy wasted is always good.

[Robert Reed]

Hello 1138
What is the trough for? feeding live stock, roof gutters or water runoff? For starters, why not use a more tamper proof plug connector. Or as others suggested a near freezing point alarm?
But since you posed the question, I assume there must be other variables involved here. One of my hobbies is woodworking and I have a large dust/chip collection system in my shop. A problem occurs when the dust accumalator drawer nears fill up and efficiency starts to drop off. I have had a design started (in my mind only at this point) that a small Inverted DC motor with a vane mounted on its shaft, and placed near the top of the drawer would alert me when the collection drawer is near full. This would be a constant speed motor and of the impedance protected type. Normal motor running current would be monitored, but as soon as wood chips filled the drawer, the vane would cease to turn and stall the motor. The motors current sensor would in turn relay this info to the control circuitry to operate an alarm.
Would something like this work in your situation, even possibly indicating slushy water before it turned to ice.
Of course what you need would be operating in much harsher conditions (temperature,moisture,etc.) but I thought this might be food for thought for your problem.

[jwax]

Robert- If you're constantly stirring the water, the temperature that it will freeze goes way down. The surface could freeze except for the place the stirring is occurring.
PS Hope your motor in the sawdust collector is explosion proof.

[Robert Reed]

J-wax
The motor I had in mind was very slow speed or gear reduced shaft speed. Don't know how much minimal surface turbulence would effect freezing point as this would in no way compare to "bubbler" sysstems that are specifically designed for that purpose. But you did bring up a good point and I guess only experimentation with various low speeds would determine its effectiveness.
As to the dust collector, only the shafted vane would be mounted in the drawer and the motor mounted externally in a sealed container and well grounded. The real danger in terms of high speed dust collections is the rapid rush of particulate flowing thru PVC duct work systems as static can build up here very quickly and consequent discharge cause explosions internal to the duct. The simple solution to this is a continuous static wire from end to end and thoroughly grounded at at least one point. My system has worked flawlessly for years. The only problems I have to consider is that any grinding of ferrous metals could pass short lived hot embers through the system and ignite trapped saw dust. No explosions here but the potential for fire is high.

[Engineer1138]

You know, oddly enough, detecting heater disconnect had never occurred to me. Must be getting old or something

It's for a livestock trough and I think that sensing loss of current in the heater will fix 90% of the occurrences.

I don't think merely sensing temperature will be sufficient because the air temperature will be well below freezing and on very cold days there is a skin of ice on the outer edge of the water, indicating that the surface is just barely above freezing. Not enough margin to measure reliably.

Funny thing is I was discussing this with my wife and her solution was even simpler: move the trough closer to the outlet so the horses can't snag the cord. I had tried this before, but there was a problem; have to revisit the idea.

[Engineer1138]

Submitted too soon!

Robert, the reason for avoiding a locking plug is that it's safer for the horses to unplug the heater than for them to possibly trip over the cord if it's icy out there.

[MrAl]

Hi there,

If you really need it i would look into installing an underground cable.
Would save many headaches it sounds like
Get the good cable made for underground use.

[jwax]

Horses? Locking plugs? Sounds like I missed something. Or everything.
Thought we were working on an ice detector.
Hope the solution is at hand Engineer!

[Engineer1138]

Nope, you didn't miss anything
The ice I want to detect forms on a trough used for watering horses. Sometimes they unplug the heater that keeps it from icing up.

The solution is to move the trough closer to the outlet that the heater plugs into. If that's not possible then I'll try detecting the current loss in the cable.

Horses? Locking plugs? Sounds like I missed something. Or everything.
Thought we were working on an ice detector.
Hope the solution is at hand Engineer!