Regular readers may recall Arfa Daily mentioning an electronic circuit that he knocked up to enable a extractor fan to be run automatically when a shower pump kicked in, but without having it slaved to the light switch, or run from a (potentially temperamental) humidistat.
I enquired about this circuit, and Arfa has very kindly provided me with detailed schematics for it, and also conducted a number of experiments with different prototypes to prove what looks like being a rather nifty little gizmo with a number of applications.
I have had a bash at wikifying it all here:
formatting link
applications listed here:
formatting link
please add any more novel uses that you can think of.
Seems a bit overkill. The current into a shower pump could easily be enough if passed through a LV AC relay coil- (say 6 or 12v) be enough to do the same thing.
Either squeeze two wires from shower and rumble grinder through the current transformer, or if they both run on the same circuit (and nothing else does) then fit the CT upstream of them both.
Failing that, you could lose the LED and re-purpose the NAND gate driving it with a 2nd CT and sense circuit, or splash out on a 2nd 4011, or use a 4009 hex inverter chip instead of the 4011 quad NAND ...
transformer. Bigger junction box and stick it inside remembering to additionally sleeve the output connections of the CT until they are outside the mains enclosure. B-)
But a neat little circuit. I've yet to buy/build the mains monitoring I want for the heating system here. I was going to use a designed for the job opto isolator chips but that would have several resistors at mains voltage disipating a watt or so and would have mains on the PCB... This appears to be much more elegant.
How much are the CT's? The "eBay" link isn't and returns a "403 Forbidden" error.
Why bother with any of the gates? What's wrong with a simple 2N7000, or other logic level FET after the crude rectifier. Also the LED indicator could be across the relay output.
Like I said. Use a low voltage relay in series with the load - put it in a bridge rectifier if you must and limit its voltage with a power zener if you like.
> Seems a bit overkill. The current into a shower pump could easily be
Although in principle that seems like a simple way of doing it, in reality, it's not quite that straightforward. Any relay chosen to do the job in that way, must have a coil capable of taking the current drawn by the pump (or other monitored load) . As well as this, it means connecting the relay directly into the pump circuit and, where the relay is being used to switch a mains feed, as in the case of controlling my shower fan, no matter which end the relay was located, it would need a cable carrying mains voltage to run between the pump area and the extractor fan control circuit. Done in the way that I chose to, the relay goes up at the fan, and is permanently connected into the mains circuitry inside the fan controller, and only a low voltage bell wire link is needed to carry the drive up to the relay.
Absolutely true, but I didn't have a little FET to hand. As I said in the description, it was a solution to a Saturday afternoon problem, using stuff that came to hand. At the end of the day, a 4011 is a dirt cheap chip with FET inputs, so is little different from a single FET (except it takes a tiny bit longer to solder in). As to using a gate to drive the LED and the relay driver transistor, I only did this to 'use up' the spare gates. They are there anyway, so for the sake of a couple of blobs of solder and a couple of bits of etch revision wire, you might as well use them. They also serve to keep each bit of the circuit independent from the other bits. A win all round, really ...
I think I paid about 6 USD post free for them, so about 4 quid. If you just stick the type number into the FleaBay search box, it will come up with a few Far East suppliers.
Ok then, use a mains neon, and optodetector a battery and a relay. neon comes on. optoresistors drops its resistance, current flows into relay..
Heck even a LED isolator can run off mains voltage with suitable capacitors diodes and resistors
Often you do not NEED to detect current. And even when you do its not hard to insert something other than.a transformer in the loop - which suffers from the same constraints as a relay coil anyway.
All current detectors
- must not disturb what ther are detecting too much
- must be able to cope with overload
Whether you use a resistor or an electromagnetic detector like a transformer or relay coil or even a hall effect sensor near the wire is no different
What don't you understand about "non-invasive to monitored circuit"?
A simple series relay isn't that simple when the load could be 0.1A or
20A... so to make the series relay work sensibly you start adding bridges and zeners. Not simple any longer.
Simple would be a a mains coil relay and across the load but that could be considered "invasive". Hum, forgot I have half a dozen plugin, DIN rail mount, mains coil relays in the old heating control box that are now redundant... I don't think I'll worry about the 1.2W per relay extra load.
Never rush into things, a simpler/better solution will appear in the fullness of time. B-)
One of the problems with having been a professional circuit designer is that instead of exultingg that you have found A solution, you simply compare it with all the OTHER possible solutions you have implemented or encountered over the years and consider that its just another solution. The trick is to find the one that matches the exact problem the best.
It would have to be a pretty good optodetector to respond to the (weak) AC light from a mains neon. The relay would almost certainly need a driver transistor as well. There is little point that I can see to using a battery to power it.
But neither solution is less complex or more reliable at doing the job under all conditions, than my circuit. The downside to your solutions are that they are invasive to the circuit being monitored. Apart from this, resistors for making an opto run from mains = heat, and if you use a capacitor as a 'wattless dropper', it needs to be an X rated type for safety reasons.
Well, not if it's a current transformer ... These are completely non-invasive, and have no care about the load current ... In the case of monitoring my shower pump, which after all is what I designed the circuit to do, what else would I have monitored besides its current draw, to determine that it was running ?
As indeed is the case with non-invasive current transformers. 'Overload' is only an issue if you are trying to keep the output linear for absolute measurement purposes. Otherwise, it just results in core saturation occurring which, from my checks on this very condition, appears to have no other effect than to screw up the waveform from the transformer, which is of no consequence to the operation of this circuit.
Hall effect sensors are actually quite poor at detecting small current flows in AC circuits. A relay coil needs to be appropriate in resistance and current carrying capability to correctly respond to just one single design value of current to be monitored. In contrast, a current transformer does the job extremely well for a very large range of monitored current values, and does it non-invasively, reliably, and without generating heat. As someone else commented, in its simplest form, my circuit would only need a small FET and the rectification components in addition to the current tranny, in order to function at least adequately. I only used the 4011 as it came to hand, and seemed an elegant enough way to stick what amounts to a FET input voltage to current converter on the end of the rectifier. As I said, the whole object was to find a reliable solution to a real-life problem, using parts that were lying around ...
No such animal. The closes you can get to non invasive is a hall effect sensor. Or a scope current probe or clamp on meter. But even these draw some power.
Not that invading a power shower is any problem at all. even a 12v drop is within mains variability.
However, in the case of my shower pump, I can't detect voltage without getting inside it. The mains supply is constant, and goes to a speed control circuit board inside the pump body. The (speed controlled) supply to the pump motor is only switched on as a result of one or both flow switches in the pump outlets closing, which they will do under static pressure when the shower valve is turned on.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.