Closed loop detection without voltage

[Scarr]

Hi all,

I have a 3m cable loop with a reed relay in it, I want to detect the status of the relay ideally without using applied voltage at all Can this be done?

If we must use a voltage what sort of voltage could we get away with and get a reliable reading

Thanks Steve

 

[MikeMl]

Do you mean "reed switch" instead of "reed relay"? A reed switch is actuated by a magnetic field, while a reed relay is actuated by a current (i.e. applied voltage).

The loop resistance with a closed contact will be only a few mΩ, so if you apply a small voltage, say 10mV, and detect the resulting current. (you are effectively building an Ohmmeter).

 

[Sceadwian]

No you can't, there is no way to sense it's state without applying at least a basic voltage (and getting a current) for continuity testing, why do you want to avoid any voltage application? You can do continuity testing quiet easily with VERY low voltages, but your reasons seem a little vague so it's hard to suggest a solution.

If you for some obscure reason actually required electrical isolation you could use a seperate isolated mechanical system (inside the switch itself) to detect it's state, but that's getting obscenely complex with no rationality for it that you've given.

 

[Scarr]

Yes it is a reed switch

Yes it is a "reed switch" and it's in a gas environment hence trying to avoid any voltage, iso we are saying 10mv is the lowest voltage to get a good reading?

Thanks

Steve

 

[Gary B]

So, you are really asking how to detect the presence of a magnetic field in an explosive environment, correct? Also, you want to detect a static magnetic field and not just a changing field. Since the reed switch is currently your magnetic sensor and it is in a glass envelope, the only issue is the possibility of heat dissipation or sparking at the other end of the wires. Limit the power to a few milliwatts and put the rest of the circuit in a gas-tight explosion proof box.

 

[Scarr]

No not really

Hi Gary B,

In answer to your question, No, I have is a gas meter with a magnetically operated switch inside and flying leads to this switch, I want to apply the least amount of voltage to this switch but still be able to reliably detect the switch operation.

I have been looking at the 10mv or even 1v options and it now poses another question, if I supply 1v to the switch input wire and it is closed, I can see 1v at the switch output wire I want to detect this on my Atmega32 using just an I/O Pin ideally, can I just attach the switch output to a pin input along with a pulldown resistor?

P.S. The meter switch voltage range is 0v-24v so I now I'm being anal but it's more of a "how low can we go" thing


Thanks all

 

[MikeMl]

Here is how I would do it. The minimum voltage across the switch must be higher than the input offset voltage of the opamp. The current through the switch is purposely set to ~1mA to make the switching more reliable. The PORT output is compatible with your uC. Use a rail-to-rail output opamp. Any of the modern CMOS opamps will work fine

 

[JimB]

A few questions:

What kind of "gas meter" ?
Is this a domestic gas meter or some industrial device?
Is the reed the type in a sealed glass envelope?
Does the gas meter have any EEx rating information on its rating plate?

A few comments:
In industrial environments where flammable atmospheres may be present, it is usual to limit the amount of energy available in the circuit.
There are various graphs published which show the maximum permissible voltage and current in a circuit for various gas groups and circuit conditions.

As a horrible simplification, if the circuit is resistive, voltages less than 30v and currents less than 100mA will not ignite gases in group 2C (hydrogen and acetylene), these being the most explosive gases.

If you use a 5v supply, and current limiting resistor of 1k ohm or more, my best guess is that you will have no problem.


Caution This information may be "out of date" and should not be construed as authoritative on the subject of electricity in explosive atmospheres.

JimB

 

[Sceadwian]

Umm, why don't you use an encapsulate reed switch? If it's hermetically sealed it's not IN the explosive environment so it can be ignored, there are many such devices available.