Solution Finder Get Tech Support

how to make a highly accurate constant current DIMMABLE, ISOLATED driver

Posted by: stalindm on

Sir,

     we made der296 highly accurate constant current dimmable non isolated driver and it is working absolute  fine. Now we need a same driver with isolation. We are working on it and trying to implement the isolation with optocoupler. But how dimming can be incorporated with Optocoupler? I understand that D7 (in der296 ) is reverse biased and disengages U2 during dimming.  Please explain how this function can be implemented in an isolated circuit with optocoupler. Kindly guide us.

Regards

Stalin.

Comments

Submitted by PI-Dunwell on 08/10/2012

U2 will be at the secondary side together with its compensation network(C11,R30,C12, R31, and R18).

The diode  side of opto is connected between R18 and cathode of U2, you may need to adjust value ofR18 in this configuration. Then on the primary, the cathode of D7 will be connected to the collector of the opto and the emitter side of opto connected to primary return.

 

The operating principle is the same with the non-isolated version. In this case when the out put current falls below the full conduction angle value, the collector of opto will rise until it reaches the bias until the point where D7 stops conducting and natural dimming commences.

Submitted by stalindm on 08/16/2012

Sir,

     thanks for your guidance. We made it and it is working fine. During dimming there was small  flickering  observed at 115v and flickering is predominant at 230v. We connected a 27v zener across U2 anode to cathode similar to der296, then dimming became very smooth at 115v. How it is happening? Also at 230v dimming is not working correctly. What could be wrong? One more question ; Addition of this U2 and optocoupler ckts will affect the power factor of the original driver? 

thanks

Stalin.

Submitted by PI-Dunwell on 08/17/2012

For TRIAC dimming, you either need to increase your bleeding capacitor ar damping resistors. You need more of this at high line because of the small input current compared to low line. THe PF should not be affected if the loop is slow enough.