-4
\$\begingroup\$

Is there any device which allows the electricity to flow only after a certain wattage and not below it ? Something reverse of circuit breaker

\$\endgroup\$
15
  • 5
    \$\begingroup\$ @user104591: so buy an always off relay. If its off, there will be no flow of electricity and since it will never reach your "on" threshold it will never switch on... \$\endgroup\$ – PlasmaHH Feb 22 '16 at 10:08
  • 3
    \$\begingroup\$ A wattage implies a voltage and current, both non-zero. This means it cannot be sensed without a flow already happening. If your device is blocking the flow, then the wattage will be zero. \$\endgroup\$ – Neil_UK Feb 22 '16 at 10:09
  • 3
    \$\begingroup\$ You could explain what is the primary problem to solve, why you would need a such device, then we could find some workarround with existing methods. \$\endgroup\$ – Marko Buršič Feb 22 '16 at 10:35
  • 4
    \$\begingroup\$ The problem you really need to solve is: how do you know what the load power consumption ('wattage') will be until you actually apply power to it? Many loads are non-linear, so applying a low voltage/current, measuring the effect and inferring what would happen with the proper supply just won't work. Even something as 'simple' as an incandescent light bulb doesn't 'look the same' electrically when its off because its resistance changes when it heats up. \$\endgroup\$ – brhans Feb 22 '16 at 12:31
  • 3
    \$\begingroup\$ Is there a practical / real world problem that you are trying to solve? Maybe adding some details about that will lead to a solution. \$\endgroup\$ – Tyler Feb 22 '16 at 15:06
6
\$\begingroup\$

What you ask for can't work. You have a fundamental chicken/egg problem.

You only want to switch on a load if it would take a certain minimum power. However, you don't know how much power the load would take without switching it on first.

This simply doesn't work in the general case without having special knowledge about the load.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ "This simply doesn't work in the general case without having special knowledge about the load." is true enough, but it's just a gobbledygook platitude. And, It certainly can work; just consider the MCU fan in a PC. Until the MCU heats up to a certain point because it's dissipating power, the fan won't turn ON, but when the MCU gets hot enough it causes electricity to be sent to the fan, which is precisely what the OP was asking about. \$\endgroup\$ – EM Fields Feb 22 '16 at 16:38
  • \$\begingroup\$ If the OP was asking about that, the OP should have asked that, in the question. The question remains nonsensical as presently stated. If the question is not the question as stated, the question needs editing. @EMFields - your assumption does not match the question asked. \$\endgroup\$ – Ecnerwal Feb 22 '16 at 17:13
  • 1
    \$\begingroup\$ @EMFi: That's not what the OP is asking about. The MCU fan is a totally different case. One thing is being switched depending on a different thing. The OP asked for the inverse of a circuit breaker. The thing being switch is the same as the thing being sensed. \$\endgroup\$ – Olin Lathrop Feb 22 '16 at 17:17
  • \$\begingroup\$ @Ecnerwal: Since English doesn't seem to be the OP's first language, I think what he originally stated shouldn't be taken literally, but with a grain of salt. A much more sensible (to me, anyway) exposition of what he wants is in his comment which followed his original post by four hours, namely: " I want a relay which can allow the flow only after certain load –" which I take to mean "activated after certain load conditions are met" a typical thermostat scenario. \$\endgroup\$ – EM Fields Feb 22 '16 at 17:49
2
\$\begingroup\$

Is there any device which allows the electricity to flow only after a certain wattage and not below it?

Strictly speaking, no. Because say you had this:

schematic

simulate this circuit – Schematic created using CircuitLab

So the idea is to close SW1 when the power (wattage as you put it) reaches some value.

But the power in this circuit is always zero, no matter the voltage of V1, or the load. The power is always zero because the current is always zero, because the switch is open.


More possible is something that allows current to flow when voltage reaches some point. You asked for something "reverse of a circuit breaker", and this thing is called an antifuse. It is a device that is open (allowing no current to flow) until the voltage across it reaches some value. Then it becomes closed.

Antifuses are what make your Christmas lights keep working when one burns out.

\$\endgroup\$
1
\$\begingroup\$

Yes, there is.

Since power being dissipated by a resistor will cause the resistor's temperature to rise, using a thermistor to detect that temperature rise and, through the use of proper circuitry, to send current (the flow of charge/electricity) through a relay coil once a particular temperature/power level has been detected, will meet the criterion described in your subject line.

\$\endgroup\$
7
  • 2
    \$\begingroup\$ This doesn't work. With the relay off, there will be no power dissipated and no temperature rise. \$\endgroup\$ – Olin Lathrop Feb 22 '16 at 14:30
  • \$\begingroup\$ @OlinLathrop: Part 1. Of course it works. Configured with The COM and NC contacts made when the thermistor's cold, they'll supply power to the resistor, heating it up. Then,when the resistor heats the thermistor up enough, the thermistor will energize the coil, causing the COM and NC contacts to break, which will cause the resistor to cool. When it cools enough, the thermistor will cut the current to the coil, the COM and NC contacts will make, and the cycle will begin anew. \$\endgroup\$ – EM Fields Feb 22 '16 at 15:59
  • 1
    \$\begingroup\$ But that's not what the OP asked about, and you're turning on the load before you know it should be turned on. \$\endgroup\$ – Olin Lathrop Feb 22 '16 at 16:04
  • \$\begingroup\$ @OlinLathrop: Part 2. The opposite scenario also works; that's the one where a cold thermistor makes the COM and NO contacts, supplying power to the resistor until it gets hot, then the thermistor will cut the current to the coil, breaking the contacts until the resistor cools, then the cycle wil begin anew. \$\endgroup\$ – EM Fields Feb 22 '16 at 16:06
  • 1
    \$\begingroup\$ A thermostat is not what the OP asked for. He wants the reverse of a circuit breaker. The load should be off when it draws less than a threshold, and turned on when it draws more than the threshold. Your answer doesn't address this at all. \$\endgroup\$ – Olin Lathrop Feb 22 '16 at 17:14
0
\$\begingroup\$

The closest thing I can come up with is a triac or SCR, with logic on the gate that looks at the voltage drop across a fake load. You'd have current going through the fake load, but it wouldn't be going through your real load.

\$\endgroup\$
0
\$\begingroup\$

I suppose you could rig a load to operate via spark-gap, but they are not exactly consistent devices, as a rule. And that's not really sensing "potential wattage use of the load," just voltage across the spark gap.

\$\endgroup\$
0
\$\begingroup\$

If the load is purely resistive, a smart device (e.g. microcontroller plus surrounding power supply control and sensing circuitry) could be built which periodically probes its resistance by generating a small voltage pulse and sensing the current. If the resistance is too great for the load to dissipate the minimum wattage, then this gentle probing process simply continues. If the resistance is in the right ballpark, then the power supply is operated "full throttle": the full voltage is applied to the load. While the voltage is applied to the load, the device continues to measure the current; if that drops below the required value, it shuts off again. (This could happen, for instance, with filament which has a low resistance which rises when it gets hot). The software running the device could contain some algorithmic safeguards to prevent oscillating on/off behavior, like exponential backoff in situations when the device doesn't stay on long enough, and the application of a hysteresis (a higher value of estimated wattage is required to turn on, a lower one to turn off). With a reasonable sample rate, device could also detect that the load is disconnected and reconnected, by noting the brief moment during which the current in response to the sensing pulse is zero. When a "misbehaving" load is detected, whose wattage drops below the threshold as soon after power is applied, the device could go into a mode in which it probes for the load to be removed. Only when it detects that a load is disconnected and reconnected does it then try the procedure again with the new load.

Obviously, we are not meeting the letter of the requirement ("no electricity flows"). Rather, we have a trickle of electricity to try to estimate how the load will behave when full power is applied.

There is no telling what the load will do (and consequently, whether it should be powered or not) without passing some electricity through it.

\$\endgroup\$
0
\$\begingroup\$

You can try a motor starting (current sensing) relay. These are used for switching in the starting circuit when a single phase motor starts up, and switching it back out once the motor is up to speed. You spec them based on the current you want them to switch at. These come as both solid state, and electromechanical devices. you will have to do some math, and maybe add another relay to get the type of contacts you want (NO/NC).

Klixon is one manufacturer.

Begin edit:

schematic

simulate this circuit – Schematic created using CircuitLab

(Assume the pickup and drop out current is the same, you wont find one but this simplifies the example)

The wattage of the load is proportional to voltage squared time resistance. in my example above the load will draw 100W at 100 volts. I sized R1 and the current sensing relay so that the relay would energize when circuit voltage is >= 100V.

Obviously if you are working with AC the nature of you load will change (R->Z) but the idea is similar.

yes this will always have parasitic losses, and there is a better way via a transistor, but this is close to what you want and not too difficult to understand.

\$\endgroup\$
2
  • \$\begingroup\$ But how do you prevent it from coming on at first if the load is only drawing a little? Once off, how does this detect that the load is drawing more? You are completely missing or ignoring the "catch 22" in what the OP is asking for. \$\endgroup\$ – Olin Lathrop Feb 22 '16 at 21:55
  • \$\begingroup\$ Instead of telling the OP it's impossible (as has already been pointed out), I chose to say how it might be possible with a little compromise. \$\endgroup\$ – Jrican Feb 22 '16 at 23:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.