Reverse phase dimming...

Status
Not open for further replies.

Bazza175

New Member
Hi,

I have this circuit found on the web, its used for reverse phase dimming.
I have worked quite a lot with trailing edge dimming and know exactly what im doing with the mains etc.
However this circuit has me a bit stumped.

I can see how the Mosfet is switched ona nd off and i can see how the mains will create the zero crossing signal using the opto. The 2 parts i am puzzled with is why the rectifier is used (how does it work in this insance) and what is the circuitry at the top doing ?

Also one last question, in a circuit such as this is it adviseable to isolate the mosfet using an opto in much the same way as you would using a triac?
Can anyone explain for me please?

Here is the schematic.

View attachment 68709

I have just looked at this picture and all the values are missing from it. Here is a link to the site if you need it. I am looking at option 2 schematic.

http://www.dz863.com/circuits-6-832...-Phase-Dimmer-Option-2-Application-Schematic/
 
Last edited:
The rectifier (bridge) is to provide a single-polarity supply for the FET and the IC.
The circuit at the top (I presume the transistor and its associated resistors and zener) form a voltage-dropper.
 
The rectifier is sort of working the other way round, if you put a bridge in line with an ac load then apply a resistance to the dc output of the bridge a similar approximate resistance appears across the ac side of the bridge (- the diode drops), so you can use a dc resistive element such a mosfet or transistor to control an ac load.
And as alec says the bits at the top left are a power supply, the tranny and the resistors connected to the collector/emitter would get very warm with any sort of load on them, an xcap/resistor/zener arrangement might be better.
 
Last edited:
The rectifier (bridge) is to provide a single-polarity supply for the FET and the IC.
The circuit at the top (I presume the transistor and its associated resistors and zener) form a voltage-dropper.

So the mains supply is powering the the IC and FET?
If i were to have a real power supply for a PIC driving the FET i.e 3.3v or 5v supply how would i control the mains? Will i still need the diode bridge or can a FET switch AC like that of a triac?
 

Ahhhh i see so if i remove the voltage dropper circuitry and provide the PIC and FET with regulated voltages then i can leave the diode bridge in place and use it to simply switch the mains on and off?
 
Nope you still need the bridge, the fet can only handle voltages of one polarity.
You can power the micro (and the o/p of the opto) from a seperate dc supply yes, be carefull with that though as the low voltage dc supply will be at mains potential.
 
Im not too keen on the idea of the DC voltages being at mains potential. The low voltage DC would need to be completely isolated from the 240v lines, this is easily achieved using a triac, is it not that easy using this method instead?
 
Even with a triac you dont have mains isolation, unless you mean its easier to provide isolation.
In my opinion I'd use a triac with an opto triac for isolation, sorted and not much complexity.
If you had to go with the bridge/mosfet idea then a good way to isolate that would be to use a pulse transformer, which would add a little overhead to the micro code and would use an extra pin as you'd need to drive the transformer above and below ground to get the fet to switch proper.
 
Currently i am using a triac with opto triac for isolation,this way my DC side is completely isolated from the mains.
The reason for wanting to use a FET, (actually i want to use a IGBT to be honest), Is so that i can use reverse phase dimming. It is quieter and easier to control in the even of short circuit etc.
 
I see, opto isolation makes things a lot safer, stick with it if you can, and it beats trying to wind pulse trannys.
You could use an opto tranny to fire an igbt of fet, but you'd still need a load of comps to provide dc to switch the device with.
If you dont want high resolution in brightness steps you can modulate brightness on a wave basis, switch the triac at zero cross and the on time is a full no of half cycles, it restricts the number of brightness levels but maintains zero crosses and is very quiet.
 
I think i would like to look into this FET switching a little bit more, i think it could be of more benefit than a triac, although simpler they are very hard to protect against short circuit. I have had instances where when a bulb blows it takes the triac out with it.
 
Some might say fets are worse, with a rds on of milliohms it doesnt ake long to get silly currents.
That said with carefull design a harder circuit design ought to be possible.
I might have a link to mosfet gate driving and the use of pulse trannys, if I find it I'll post it.

edit: the link I had is dead, but I found this, its an interesting article on fet gate driving, its aimed more at smps design, but the same applies to your circuit, as it could be classed as a switching reg, the circuit of fig2 would probably work for your app, drive the second inverter from your micro, however you'd need to make a trans you knew had good isolation.

https://www.electro-tech-online.com/custompdfs/2012/11/transformer20isolated20gate20drive.pdf
 
Last edited:
I have just found a few examples of AC switching with mosfets, but they use 2 mosfets to do it in series. Do you have any knowledge on this sort of setup dr pepper?
 
Hmm got to say not really, the only thing similar to that I've played with was a zero v drop diode circuit using fets, cant remember what that was for now.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…