New wiki article: Current activated switch

You did not read the comment after the photo then ;-)

ok, fixed the link - the site did not like deep linking. If you clicked on the address after the 403 and hit enter it then worked since the referring site was them the right one! Done it as a search now.

OK for some applications, but its not a general solution. Say I want to build a dust collector controller - I now need a solution that will run for a wide range of tool powers from a 150W to 3kW.

For an app like sensing when a electric shower is on, I sure don't want to be inserting things inline with its 45A load. In fact I don't even want too many extra connections.

Feel free to do a version that you think is better, and we can post that as well. ;-)

No. A CT requires you break the circuit to thread the conductor through. That is invasive.

MBQ

This is a shower pump that is live all the time and draws current when the water flows. The neon would need to be in series with the existing wiring to detect the current flow. A mains neon needs mains voltage to illuminate. What's then left for the shower pump?

MBQ

Doing it that way carries additional risk of an open circuit on the wires from the CT and failure of the insulation on those wires though overvoltage. The connection from the CT to the detection circuit needs to be as short as possible, with as few connections as possible and preferably with something like a spark gap or MOV to suppress overvoltage if something does go wrong as open circuiting CT's on load is a *very* bad thing to do.

I accept it was expedient at the time, but felt a wiki article could have a simpler circuit. It would nearly half the number of solder joints that had to be made!

You still need to strip the outer insulation from the cable to get at the individual conductor. That is still invasive.

MBQ

plenty of CTs which open and then close round the conductor.

You're both arguing over nothing.

Any attempt of a measurement will in some way or form affect the circuit being measured. Fact!!

No...

Still broken. No links shown on the "403 Forbidden" page that I see. The full HTML of which is:

403 Forbidden nginx

But I expect half the news reader out there will render that rather than display it as raw HTML...

Dug about the uxcell site and found it at:

The last bit can be mitigated by putting the burden resistor across the terminals of the CT at the CT rather than remotely.

Stop squirming.

The cabling will be split into individual insulated wires inside the junction box that joins the fixed wiring to the appliance flex anyway.

not if you are doing it inside the CU where the conductors are separated.

ok try now... linking to :

You seem to be discussing very subtle effects on the thing being measured brought about by the process of measurement. No one is disputing these happen, but it is not what is meant my "non invasive" in this context.

For this application, non invasive means without modifications of the thing being sensed, and without direct connection to its circuit.

Well, you just go right ahead and think what you like. Clearly, you haven't ever had much to do with high ratio current transformers.

I have no need to defend or justify the circuit to you. I had a problem. I designed this. It works. I don't really care whether you like it, hate it, think it's utter bollocks, or possibly just don't understand it. It does what I needed it to. John felt it was elegant enough to do something else with and if it can help others to solve a problem, then fine.

End of.

Arfa

Oh, pulleees ...

Arfa

Well, possibly true. But we're only talking a few joints total anyway, so no real biggy to anyone with enough soldering iron savvy to attempt to build something like this, I would have thought ? It was not me that put it on Wiki. John thought that it warranted putting up there, so he decided that it was ok as it was.

Arfa