Skip to main content
Topic: Twin triac AC switches on DP5050 (Read 4152 times) previous topic - next topic

Re: Twin triac AC switches on DP5050

Reply #1
This forum is not called Dangerous Prototypes for nothing! Niklas pointed out the the creepage and clearance are inadequate in this design, and the snubber capacitors are specified in too small a package. With limited space on the board I have removed the snubber circuits entirely. Snubberless triacs (e.g. BTA16-600W) are therefore a necessity for any kind of inductive load.

I have respun the layout splaying the pins of the TO220 package triac to increase the clearance between the hot traces to at least 2mm.

Re: Twin triac AC switches on DP5050

Reply #2
Prototype boards ordered. Where are people getting their chunky TO-220 heatsinks? The last ones I got were from Futurlec, but I can't be waiting 6 weeks for some to turn up.

Re: Twin triac AC switches on DP5050

Reply #3
I don't think you need any chunky heatsink.  The triac is not in linear mode like a mosfet in a constant load - it's either on or off and have rather moderate losses.

Look at the calculation & design examples in this document. http://www.nxp.com/documents/applicatio ... N10384.pdf

Re: Twin triac AC switches on DP5050

Reply #4
Right you are. I was aware of the general rule of thumb that the voltage drop is about 1V so the triac dissipates about 1W for each 1A of current. I just wasn't sure what the thermal resistance of these heatsinks were.

I once pulled the speed control out of a 1200W / 240V vacuum cleaner and the heatsink was just a square of thin aluminium sheet about 3cm wide. The triac was something like a BT138.

 

Re: Twin triac AC switches on DP5050

Reply #5
Board populated with 2x BT136.