EEVblog Electronics Community Forum

Electronics => Beginners => Topic started by: 5282 on February 22, 2016, 06:46:05 pm

Title: 3842 problem
Post by: 5282 on February 22, 2016, 06:46:05 pm

i have problem with 3842 ic
 
(http://4.bp.blogspot.com/-IunbZnTzD8Q/TWNCf4qxGII/AAAAAAAAAHY/BUW6NbBbKmY/s1600/IC3842.gif)

i use 150k resistor for r1 the voltage on c3 is only=2.8v 

i don't connect the mosfet to 3842

Title: Re: 3842 problem
Post by: T3sl4co1l on February 22, 2016, 07:15:04 pm

...Yes?

Do you understand the datasheet? Have you read the datasheet?

This circuit has problems. I have never seen R3 used before. There is usually a capacitor from pin 3 to ground, pin 8 to ground, and an R+C across the '431.  As shown, it will oscillate horribly (once the MOSFET is connected and all, that is).

Tim

Title: Re: 3842 problem
Post by: Andy Watson on February 22, 2016, 07:18:09 pm

What have you done with pin 5 on the 3842 ?

Title: Re: 3842 problem
Post by: prawncrackers on February 23, 2016, 08:39:37 am

i burn 3842 ICs when i dont connect them to the mosfet AND the current sense resistor(R6 in your schematic) so definitely check that out.

you can try a series incandescent bulb(around 100w mains voltage) to limit the current to your circuit if youre afraid something might pop.

you also forgot your pin 5 in the schematic; this goes to ground right?
also you might want to try and omit R3. experiment only when your very basic circuit functions then work your way up for the modifications.

Title: Re: 3842 problem
Post by: 5282 on February 23, 2016, 02:24:41 pm

thank all
i not use  R3 in my board
yes i already connect pin 5 to GND
there is low votlage in 3842 vcc=2.8v with 150kohm
now the mosfet is useless because 3842 not work with 2.8v

Title: Re: 3842 problem
Post by: T3sl4co1l on February 23, 2016, 08:12:51 pm

...I guess that's a no, then...

The startup circuitry will keep turning itself on and off, holding VCC at a low average voltage.  It's still peaking high enough to turn on the MOSFET.  If you do not have an oscilloscope, you will not be able to see this.

Without a MOSFET connected, there is no power from the AUX winding, and the controller will never remain on.

I don't recommend building SMPS without an oscilloscope, anyway, because you will never have a way to tell whether it's working well or not.

Tim

Title: Re: 3842 problem
Post by: Seekonk on February 23, 2016, 10:54:07 pm

Two things to try:  Greatly increase the value of C3.  Reverse terminals 3&4.

Title: Re: 3842 problem
Post by: prawncrackers on February 24, 2016, 12:57:53 am

also did you check the orientation of transformer windings? im thinking this is a flyback converter.

in any case you might want to try using an auxiliary power supply to power the 3842. i sometimes use one when making converters, just a 12-15v wallwart/charger to pin7.

Title: Re: 3842 problem
Post by: Seekonk on February 24, 2016, 02:35:44 pm

A LED indicator on the output may give an indication if the oscillator ever starts working when you don't have a scope. As said before, powering the chip independently is the best step