Driving a led with mains. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Driving a led with mains.

first previous 4 5 6 7 next last

Move Thread

LAN_403

Steve Conner

Tue Sept 24 2013, 10:52AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

They certainly can be used for powering a number of things, but the whole circuit has to be separated from the user by Class II double insulation. No problem for something like a LED night light, but a non-starter if the user wants to connect the circuit to anything else.

The reason is that the current supplied by the capacitive dropper comes directly from the mains with no isolation, so you have to treat it as if it were the mains itself.

You can arrange it so that the output voltage is referenced to the neutral, but this is still forbidden as many countries covered by the IEC use 2-pin plugs that can be inserted either way round.

Ash Small

Tue Sept 24 2013, 01:41PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Steve Conner wrote ...

They certainly can be used for powering a number of things, but the whole circuit has to be separated from the user by Class II double insulation. No problem for something like a LED night light, but a non-starter if the user wants to connect the circuit to anything else.

The reason is that the current supplied by the capacitive dropper comes directly from the mains with no isolation, so you have to treat it as if it were the mains itself.

You can arrange it so that the output voltage is referenced to the neutral, but this is still forbidden as many countries covered by the IEC use 2-pin plugs that can be inserted either way round.

What about LED lighting systems?......Lighting syatems using 'Swan-Edison' bulbs or 240V halogen bulbs have no mains isolation, and other types of lighting systems have capacitors connected directly to the mains.

BigBad

Tue Sept 24 2013, 03:08PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

You shouldn't be able to touch any part of the circuit in those cases either. For a phone charger, you could plug virtually anything in, some of which items may have an earthed case.

Incidentally, powering LEDs off the mains is becoming much more common; the cost of LED lighting is now below the cost of filament bulbs (including the cost of the electricity).

Ash Small

Tue Sept 24 2013, 07:13PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I was under the impression, from reading the Epcos application notes, that Y1 capacitors don't require insulation:

,property=Data__en.pdf;/ PDF_EMIGeneralStandard.pdf

Anyway, back to the simulations, and it looks like this circuit is immune to the ringing present across the LED's in my earlier attempts:

The first trace is 'normal operation'...the second trace is when it is switched on at 240V (mid-way through a half-cycle)....no ringing or spikes present across the LED at all, and everything within tolerance.

Also, no resistors on the input side at all. None until the output stage.

EDIT: This would still need a bleeder resistor, maybe two....and a suitable fuse.

BigBad

Tue Sept 24 2013, 08:13PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

Not bad, it's pleasingly symmetric.

Can you combine the inductors and put them in series with the diode or does something blow up?

Ash Small

Tue Sept 24 2013, 08:29PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

BigBad wrote ...

Not bad, it's pleasingly symmetric.

Can you combine the inductors and put them in series with the diode or does something blow up?

Thanks

I threw the inductors in to simulate the inductances in the capacitors, that's the main reason there are two, of different values. I tried various values, but most gave similar results, just a different frequency and amplitude for the spikes/ringing. All were way to high, so I( had to 'isolate' the LED from it. The capacitor across the LED has a low resistance (from the list of availavle models), and presumably only drops to the foreward breakdown voltage of the LED (1.8V). I think it's the diodes that block the spikes, I think they may be too slow to start conducting, and strong enough to withstand the voltage spikes....or there may be another mechanism, I've not fully investigated yet.

EDIT: I've just rectified a mistake. I forgot to change the settings for the LED. I've done it now:

I'll post a tidier schematic later. I'll also measure the self-inductance of the capacitors.

BigBad

Tue Sept 24 2013, 10:10PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

That's much, much, much too much inductance isn't it? 1 microhenry for a 1 microfarad capacitor?

Ash Small

Tue Sept 24 2013, 10:19PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

BigBad wrote ...

That's much, much, much too much inductance isn't it? 1 microhenry for a 1 microfarad capacitor?

I agree, but the 10uF cap is the biggest I've ever seen. I did try much lower values, as well as higher ones, and the results were pretty much 'of a 'muchness'', just differences in frequency and amplitude, this circuit eliminates all of it. I'll try to dig out the 'scope and sig. gen. tomorrow. I'm pretty certain I can make a few more tweaks, and 'maybe' eliminate the resistors completely, but I think I've already proved a point.

EDIT: I can always add a couple of inductors if I need to

BigBad

Tue Sept 24 2013, 11:44PM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

You're off by 3 orders of magnitude; the inductance of a microfarad capacitor is about a nanohenry.

Steve Conner

Wed Sept 25 2013, 08:46AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

As a very rough guide, the stray inductance of all types of circuitry is a few nanohenries per inch. This holds for a 1 inch long capacitor just the same as a 1 inch piece of wire.

Some capacitors have a lot more inductance than this because they are made out of a long roll of metal and dielectric with the connections just wedged into it at a single point. The current might have to flow 10 feet inside the capacitor. Microwave oven capacitors are one example.

Inductance per inch of wire can only be a very rough guide because it is really loops of wire enclosing magnetic flux that have inductance. If you take 30 inches of wire and lay it out in a circular loop, it'll have a lot more inductance than if you scrunched it up into a little ball. This is why we twist gate drive wires together rather than allowing them to flop around at random.

first previous 4 5 6 7 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob