Forums » General Electronics Discussion

Hi!

I would like to PWM-dim a 230V halogen lamp with the following circuit:

Do you think this layout is correct for this purpose?

The idea for the powersupply is from: http://www.mikrocontroller.net/articles/Datei:Capnt.gif

And for the dimmer the idea is from: http://www.next.gr/circuits/230v-AC-Led-Dimmer-l43256.html

Thank you for your help!

LeFish

@LeFish,

Sorry to go all Nanny State on you but...

Whilst I (and I am sure you) undertsand what you are attempting to achieve, you must understand that your circuit is potentially lethal and, since you have made it public, you have a duty of care to clearly indicate that this is the case.

Also, whilst I apprecitae that your placement of a ground symbol is to assist in the simulation of your circuit, it is extremerly misleading and dangerous for anyone simply reading your schemtic as a simple circuit diagram.

What you have drawn auggests to the uninformed that you can connect the input to the power supply section of your circuit across mains live and earth.

From a practical point of view, your connection to the lamp is on the wrong side of the bridge rectifier, BR1. It may work like that but it is not clear. Please refer to your original source circuit.

I suspect that your nets labelled as GND should be labelled as DC_Out_GND. I am not sure if CL treats a net labelled as GND the same way but if you labelled nets as GND in a spice simulator, they would autmatically be connected to the global GND or 0 net, which in your circuit is the low side of the high voltage AC input and is clearly not what you intend.

Your circuit begs the question of why not just use the dimmer circuit you have already linked to?

Why do you need the additional complexity of a regulated 5V supply?

In view of the possible lethality of your circuit I strongly suggest that you at least revert it to private until such times as you have added adequate safety warnings in the schematic, the description on the circuit page and in any of your postings about this circuit.

Even if you have a good lawyer.

:)

@LeFish,

Thanks for updating your schematic.

For anyone else reading this: my comments above referred to a version of the schematic that was posted on 151123.

The safety warnings, ground symbol placement and some of the netnames in the schematic have been changed since then.

OK, In the abscence of more info on how the circuit is supposed to work... (HInt: a How It Works document or notes is always helpful :) )

I was wrong about placing the lamp is inside the bridge: by connecting it that way you don't short out your own power supply source when the switch is turned on.

I assume the Generic Current Sink is to mimic the current drawn from the 5V rail by the uC that is generating the PWM output that goes to the opto isolator.

10k in series with the MOSFET gate might slow the turn-on and the 100k to ground the turn-off so much that the MOSFET dissipates significant power during those events. At non-zero mains voltages, that could blow the MOSFET.

You'll almost certainly need a choke in series with the mains supply. None of: your wireless hub, your neighbours, the utility company or the Radio Communications Agency/Ofcom/FCC etc. will appreciate you stuffing PWM chopped current back up the mains.

If the total load is over 75W you are probably obliged to apply some form of power factor correction (i.e. smoothing) to limit the harmonic junk such a PWM scheme exports both as conducted and radiated emissions.

Given that the combined voltage drop and consequent power dissipation of the bridge and the MOSFET are probably about the same if not more than that of a TRIAC based dimmer, it rather begs the question: why not just do it that way?

Hi! Sorry for the late reponse.

You are right with the Generic Current Sink: this is supposed to mimic the uC.

Due to your very well appreciated comments I will not build the above posted circuit. I will, instead, use a TRIAC-dimmer circuit.

The above circuit will be deleted.

Thank you for your help!

LeFish