Are the sensors (I presume they're current transformers) given away free with Energy Meters a while back suitable? I'm sure lots of people have got one attached to their meter tail but the novelty of watching the graph go up when the kettle's on has worn off.
Owain
Simple answer is that I don't know. When I designed the circuit for my specific need of controlling the shower extractor fan, I did so knowing that the one I had laying around left over from another project, had a ratio of
2000:1, so the circuit did not need to be especially sensitive to work ok with the fairly low load of the shower pump motor. The ones that were linked to on the RS site are 1000:1, but I reckon that the circuit would still be ok with that as well, if not able to sense quite as low as with the specified current tranny. That said, you would be able to get back to the same sensitivity by passing the wire being sensed through the core of the CT twice.However, not knowing what the spec of the freebie ones is - assuming that they are CTs anyway, and not some form of sensitive Hall device - , it's not really possible for me to say if they would work with this circuit to have enough sensitivity to sense a low load. My best suggestion would be that if you have one to hand, just rat's nest the circuit together, and try it.
Arfa
"Andrew Gabriel" wrote in message news:k2all7$hik$ snipped-for-privacy@dont-email.me...
Agreed in principle - yes. However, bear in mind that I originally designed the circuit with the relatively low current demand of the shower pump in mind, and at those current levels, excess voltage across the secondary winding during the reverse half cycle, is not an issue. I did a number of checks in this regard when John suggested that there may be 're-purposing' uses for the circuit, and I found that up to about 3kW of load, there was not an issue. Above this level (actually simulated by passing the monitored wire through the core a couple of times to double the current density) the waveform became increasingly scruffy, and the voltage was starting to rise a little on the high side. This is why I suggested that if loads above 3kW were to be monitored, R5 should be added to load the CT with something approaching its design impedance. 1k was a reasonable compromise between maintaining sensitivity, matching the transformer impedance, and mitigating another undesirable effect. This was caused by the transformer core starting to saturate at high currents. This is not a problem for the circuit, as it is not looking to measure absolutes. It also has the desirable effect of limiting the peak reverse voltage applied to the diode (or your finger). The undesirable effect was that the core saturation manifested as a squaring off of the top and bottom of the waveform, and this in turn resulted in the winding 'ringing' at the transition points. The addition of R5 stops that, and returns the waveform to a better shape.
The BAT85 is rated for a continuous reverse voltage of 30v. I have not seen any figure quoted for a PIV but it's probably similar. At the sort of levels of monitoring that I originally designed the circuit for, the diode is not subjected to anything like that voltage, so is fine for the job. If the circuit was used for monitoring very high loads, then agreed, it might be prudent to use something a bit meatier. I would reiterate in this regard that the original circuit was designed for a specific purpose, and to use components which came to hand. As I was at pains to point out, most of them are not cast in stone, and to some extent, this must be left in the hands of the constructor if they are repurposing it for some other task specific to them. However, I was also careful to specify the exact CT that had been used for all of the testing at various loads, and which I felt should be used.
No one can really be responsible for what might happen if a different sensor was used, and safety in this regard, must be up to the individual. If they are not competent to be working with mains, then they probably shouldn't be attempting to modify the circuit by using a CT other than the specified one. At the end of the day, almost any item that is considered generally electrically safe, can become dangerous in the wrong hands. A figure 8 lead is dangerous if you stick a paper clip down its holes ...
If I get time at the weekend, I will do the checks again of open circuit output voltage at various loads, and BAT85 reverse applied voltage with and without R5. That should put your mind at rest - or otherwise of course !! I'll report back.
Arfa
In order for a current to flow, the conductor must loop back to the load and power source. It's always at least one turn. A steady current can't flow along a conductor which consists only of a straight wire.
and the transformer generates a tiny voltage drop in the primary, which together with the current flowing is the power it is stealing from the circuit, and delivering to the sense circuit via the secondary. (If it's a perfect transformer and you short out the secondary, then it will in theory become completely passive, but then there's nothing for you to measure anymore.)
What an extraordinary statement.
I know what you are trying to say...
So they've been lying to me all these years, then. ;-)
The early phone and telegraph companies often used earth (as in the ground) return circuits for long distance lines with a single wire. They seemed to work well enough.
A current transformer uses the changing magnetic field round one conductor to sense the current in that single conductor. The location of the rest of the circuit is irrelevant to this.
There ain't no such thing as a free lunch. Even Hall sensors will absorb some energy from the field they are sensing.
You're getting your short and open circuits mixed up, too. If you short the output of a current trnsformer, it will absorb energy according to its impedance, plus any core hysteresis losses. If you leave it open circuit, then it will only absorb enough energy to make up for the hysteresis losses in the core.
Yes, but does he?
In which case the conductor isn't just a straight wire, as it loops back through the earth making the 1-turn coil, and that's fine.
That is the point I was making.
There's no [hysteresis or other] losses in a perfect transformer. If the secondary is shorted, the CT effectively vanishes. If the secondary is open circuit (a very bad thing to do with a CT), then you have an inductance (just as though there was no secondary) which will have an AC impedance which will cause a (tiny) phase shift, but still consume no power.
You can get a bigger response by making a tuned circuit with the secondary.
MBQ
OK. Here's the bottom line. Today, I have set up a number of tests on the circuit, using the SPECIFIED current transformer, and find the following :
Without the 1k termination resistor, R5, the waveform from the CT is 'scruffy' as would be expected. This has no consequential effect on the operation of the circuit tested from 40 watts to 2kW with a single pass thru' of the monitored wire. I then tried the same test with three turns on the CT. Above 1kW load - so approx 3kW simulated - core saturation becomes increasingly visible on the waveform, and limits the positive and negative voltage excursions produced.
A Thurlby 1504 true RMS voltmeter connected across the winding, produces a 'best guess' reading that never exceeds about 14 volts. Not definitive, I know, as the meter is probably calibrated for a sine wave, but never-the-less, a reasonable indication.
A 'scope connected across the BAT85 rectifier shows reverse spikes under all load conditions, that are much higher than the 30v assumed PIV of this device. However, I ran the circuit with a 2kW fan heater as the load, and the three turn configuration, so 6kW simulated, for more than half an hour, and the diode did not fail.
At every test, right up to 2kW and three turns, I applied a mk 1 finger (mine!) straight across the transformer terminals. I even did this with the transformer completely disconnected from *any* load, and at no time was I able to feel even the slightest tingle, let alone any kind of shock.
So here it is. Apart from the possible validity of the comment about the reverse voltage across the BAT85, I can find no fundamental problems or specific dangers with the circuit. If anyone wanting to build it was concerned about possible diode failure from reverse voltage breakdown, then they should feel free to substitute a different diode type. There are loads of Schottky diodes that are dirt cheap, and I'm sure that a bit of delving in the catalogues would find one with a PIV of several hundred volts.
I am not prepared to comment on the dangers of substituting other current transformers, as I have repeatedly said that I would recommend using ONLY the type that I specified, which seems to work ok with loads up to at least
6kW, and almost certainly would be ok at up to 9kW for the most powerful 'instant' shower types, particularly if the terminating resistor, R5 is fitted. If you were trying to monitor loads beyond this, then probably, you shouldn't be trying to do it with a DIY circuit ...I have now spent more time on this than was ever intended, so it's now up to potential builders to decide for themselves, based on their own experience, and my comments on the design. If you have a use for this circuit, and trust what I say in that it won't electrocute you, catch fire, or launch any of its components into your eye, go 'ed and build it. If not - don't. Simples, really.
If anyone does build one, I would appreciate knowing what the use was, and whether it did the job, Also, any issues with getting it to work, and any comments on substitute components or value changes. If anyone dares try a different CT (!) I'd be interested to know the outcome ... I can be mailed direct at any time on the posting address, as can John, I guess. Or if you posted on here, I'm sure one of us would spot it.
Arfa
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.