Electronics > Projects, Designs, and Technical Stuff

Using AC current metering to control a heater.

(1/2) > >>

soldar:
I have a mains circuit where I want to keep current consumption close to, say, 10 A. For example, I have a space heater but I have to disconnect it when I use the microwave. So the idea is that this control circuit would shut off, or even modulate, the heater and keep the total current as close to the setpoint as possible.  Let us say the target is to keep it at 10 A, then the control circuit would control the current to the heater so the total is as close to 10A as possible. If the other demand goes up then the heater gets less power.  Proportional control would be good although I guess two or three steps would also work.

So I already have a volt & current panel meter and I think I could get a current signal from there although it would probably have to be amplified. Alternatively I could just measure using a shunt.

At any rate, suppose we have a usable DC signal proportional to the AC current. Now what? I am struggling to find a simple way of doing it.

I suppose a really simple way would be when current signal goes over a certain limit then disconnect heater for x minutes. Really simple. It just tries again automatically every x minutes and cuts again if current is still too high.

It would be more elegant and useful if I could have a triac control the heater current so that the total remains close to the setpoint. I could use optically isolated triacs but I cannot think of how to achieve this.

A midway solution would be to have two or three steps with relays. It would be relatively easy to switch triac resistors with relays.

Well, I hope this description is enough to get the ball rolling. It might be simpler than I am making it.

David Hess:
It is certainly possible to do with feedback in the analog or digital domain using phase control or cycle skipping; I would go with cycle skipping for lower noise but in practice I doubt it matters.  The current measurement made with an isolated current transformer is compared with a reference using a simple PID control loop perhaps with an operational amplifier to drive the phase control or cycle skipping power element.  Some thyristor preregulators for linear power supplies work in a similar way.

You might be able to bodge a general purpose oven temperature controller into doing it for you by replacing the temperature input with an average current input.  Or a process controller that can control the heater load could be used.

soldar:
Thanks but the more I think about it, the more complicated it seems. I don't want this to become a complicated project because it is just to save me the trouble of keeping an eye on the meter and switching things off and it is not worth a huge effort.

So I am thinking it might be worth going the simplest way of detecting overcurrent and switching off the heater for a few minutes. That seems really simple compared to the proportional control unless I can see a very simple way of doing proportional control.

Another issue is that the heater(s) is not located at the panel where the sensor transformer would go so either I provide a long cord from there, which is impractical, or I provide some sort of remote control.

So my thought right now would be to get a current transformer like this one, which gives me a low level ac signal, amplify, rectify and when it exceeds a certain level triggers whatever...

Then it activates a circuit which injects in the AC line a high frequency tone which causes the receiver at the heater to switch off for a set time.

After this time the heater connects again and if the current goes too high it will be disconnected again.

I could probably find this type of control over powerline if I knew what terms to look for. 

Any ideas welcome.

I just had a thought for a crude and simple over-current sensor which would be a reed switch with probably only one or two turns of ac wire around it needed to activate it. It would chatter at mains frequency but that is good as it would stop as soon as the current fell below the threshold. All I have to do is get this to work acceptably and find a circuit for signaling over the powerline... or even over the air.

David Hess:
If you want something simple, then detect the microwave load with a current sensor and switch a relay shutting off the heater outlet while the microwave is operating.

soldar:
Well, the microwave was just an example. I have some heater(s) which I would like to switch off when other loads come on; it could be microwave, toaster, stove or anything else.

For now I just keep visually monitoring the meter and I have a fair idea of what each thing uses anyway but it would be nice to automate it. 

I will do some tests with reed switches and see if that system can be used. I also need some remote switching capability.

After some consideration I have lowered my expectations a lot :)

ETA: After a brief test with a reed switch it looks very promising. It switches on/off 100 times/second very cleanly and with no problem. Roughly at 7 amp*turn so it would take even less than one turn to switch. I think I could just make one turn in the wire and then adjust the switching threshold by moving the reed switch closer or farther.  Seems like a really simple way to detect excess current.

Searching online for remote switch over power-line I cannot find anything useful as mostly it shows Ethernet switches.  I remember building decades ago a similar thing. It just injected a 100Khz signal into the power-line, through a capacitor, very simple. The receiver was also quite simple. A high pass filter and little more. I think this is simple enough to get started in this direction unless I can find ready-made power-line communications remote switches. It seems this is the kind of thing you would find cheaply on ebay but I cannot find it.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod