Triacs control board - Is this circuit correct?

[atferrari]

Triacs control board

From an old issue of Elektor, every time I try to understand this circuit which seems to be drawn in a somewhat weird way , I get confused. 

Could anyone with experience in AC control tell me if it is OK?

I will appreciate that.

[mij59]

Hi,

The circuit works, to understand what's going on take a look at this http://en.wikipedia.org/wiki/TRIAC .
The triac is controlled using quadrant 2 and 3.

When the opto coupler is turned of, T1 will conduct, and there can flow no gate current OUT of Tri1.
Note that the positive terminal is D.

[T3sl4co1l]

It's OK.



More specifically:

For whatever reason, they decided to draw it "upside down".  This was common back in the days when electron current flow and PNP transistors reigned, but rare and ugly by today's standards.

If you can imagine everything upside down and negated, it works thusly:

A, C are like VCC = -8V, and B, D are like GND.  All the transistors are common emitter, and they either allow a current (about 7-8mA DC) out of the TRIAC gate, or shunt it to ground, keeping it off.

The optoisolators are opposite polarity (NPN), but this doesn't matter much because they're really just being used as switches, too.  Namely, the 22k pull-up resistors are keeping the transistor bases forward-biased, so they sit at -0.7V, unless the respective opto turns on, shunting it to ground, turning off the transistor and turning on the TRIAC.

The TIC206D says +/-5mA gate current for quadrants 1-3, and 10mA for quadrant 4.  This circuit is using quadrants 2 and 3, so the ~7mA bias should be sufficient, at least for casual line switching applications (and probably still fine anyway, for zero-crossing or phase control applications, up to 100-200Hz).

Note that a TRIAC is like a BJT, in that positive input current (G-MT1 looks like a diode or Vbe junction) allows positive output current, but it also allows negative output current: when forward biased, the whole damn thing kind of turns into a mushy pile of conductive silicon, and doesn't stop until you let go from all sides (gate AND load current at zero for >> 10us).  The gate looks like parallel back-to-back diodes, so it can be "forward" biased with either polarity: it just so happens, one combination (positive gate, negative load -- 4th quadrant) doesn't work so well, and requires a little bit more drive to ensure conduction.

Tim

[eneuro]

Sometimes it is better make own 25A triac using two thyristors in parallel

Here before final assembly-5mm width tracks covered with added soldered double copper wires 1.7mm in diameter each, so more than 4mm2 of copper per track 


Additionally I use AC mosfets switch to triger this thing with galvanic isolated mosfets gate driver