Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
08/07/04 23:03
Read: times


 
#75625 - The advantage is obvious!
Responding to: ???'s previous message
...and if we just have half wave rectification (as previously suggested by Jose and Erik) then I think the dissipation is tolerable. Erik - I'm not quite following what you've suggested with the other opto.

The 'other opto' enables the constant current source for only a short period of time, in order to drastically reduce heat dissipation of whole circuit, as it is shown here:



Assume you want to use your PC900V opto coupler. Then, due to fabrication tolerances, you need a minimum turn-on current for opto LED of 4mA, if ambient temperature differs from 25°C. Means, your constant current source transistor (here DN2540) must at least allow 4mA to flow. But as it is a bad design practice to choose this current equally to minimum level, a certain headroom is needed. Taken into consideration, that constant current will change with temperature of crystal temperature of involved transistor, constant current should be set to at least 6...7mA. Don't forget that common mode noise will also degrade signal integrity of opto link, so that this constant current level seems to be the absolute minimum for relaible operation.

Next assume, that 230V + 10% overvoltage hits your circuit. Makes 358Vp. Heat dissipation of DN2540 then is about 358Vp x 7mA x 2 / pi = 1.6W with full wave rectification, and about 0.8W with half wave rectification. If you now take into consideration that DN2540 has a thermal resistance between junction and ambient of 70K/W, then, assuming half wave rectification, crystal heats up to 96°C, if ambient temperature is 40°C.

The problem now is, that although cooling of this transistor is possible by the help of a heat sink, of course, heat sink must be well isolated from transistor case! And if safety distances have to be provided, additional efforts are needed...

Quite differently with Erik's idea: If full wave rectification is used, constant current source need only to be activated for about 5msec, and this every 100msec. Gives a decrease of heat dissipation down to less than 358Vp x 7mA x 5msec / 100msec = 0.13W! A heat, which can nicely be handled without needing to add a heat sink.

Kai

List of 43 messages in thread
TopicAuthorDate
Wide voltage led (a little O.T)            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
   More information needed            01/01/70 00:00      
      RE: More information needed            01/01/70 00:00      
         RE: More information needed            01/01/70 00:00      
         RE: More information needed            01/01/70 00:00      
            RE: More information needed            01/01/70 00:00      
               RE: More information needed            01/01/70 00:00      
                  RE: More information needed            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
      RE: Wide voltage led - Jose            01/01/70 00:00      
         RE: Wide voltage led - Jose            01/01/70 00:00      
            an idea            01/01/70 00:00      
               An excellent idea!            01/01/70 00:00      
            RE: Wide voltage led - Jose            01/01/70 00:00      
               RE: Wide voltage led - Jose            01/01/70 00:00      
                  RE: Inductor - Jose            01/01/70 00:00      
                     RE: Inductor - Jose            01/01/70 00:00      
                        RE: Inductor - Jose            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
      RE: Wide voltage led (a little O.T)            01/01/70 00:00      
         RE: Wide voltage led (a little O.T)            01/01/70 00:00      
            RE: Wide voltage led (a little O.T)            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
      RE: Wide voltage led (a little O.T)            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
      The advantage is obvious!            01/01/70 00:00      
         RE: The advantage is obvious!            01/01/70 00:00      
            RE: The advantage is obvious!            01/01/70 00:00      
            RE: The advantage is obvious!            01/01/70 00:00      
               RE: The advantage is obvious!            01/01/70 00:00      
            RE: The advantage is obvious!            01/01/70 00:00      
               RE: The advantage is obvious!            01/01/70 00:00      
         RE: The advantage is obvious!            01/01/70 00:00      
            RE: The advantage is obvious!            01/01/70 00:00      
               Help on MOSFETs needed!            01/01/70 00:00      
                  RE: Help on MOSFETs needed!            01/01/70 00:00      
                     RE: Help on MOSFETs needed!            01/01/70 00:00      
   RE: Wide voltage led (a little O.T)            01/01/70 00:00      
      RE: Wide voltage led (a little O.T)            01/01/70 00:00      
         Creation of directory is needed            01/01/70 00:00      
      RE: Wide voltage led (a little O.T)            01/01/70 00:00      

Back to Subject List