The role of feed back diode in ATX power

Status
Not open for further replies.
zereshki said:
2-change in divider resistors network of tl431.
it cause 5vsb increase too. but 5v is useful for digital

3-adding some turns to 12v node in secondary of stand by transformer
I don't test it yet
Click to expand...
In this case, you risk to overload the backup source. How do I know that produces 5V 300mA max standby 1.5 W, that is. But this solution can work.
zereshki said:
some comments on your circuit:
Click to expand...


Microsim provide an additional source for TL494 buffer transformer. In your circuit voltage is limited by Zener diode ZD26, which works only when the spike exceed 24V relative to voltage on pin #12.
 
Last edited:
In this case, you risk to overload the backup source. How do I know that produces 5V 300mA max standby 1.5 W, that is. But this solution can work
Click to expand...

you are right


Microsim provide an additional source for TL494 buffer transformer. In your circuit voltage is limited by Zener diode ZD26, which works only when the spike exceed 24V relative to voltage on pin #12.
Click to expand...


It is a typical atx power circuit,not mine, there is not any zener in my atx. I check out my atx circuit. driver transformer driving, is same as yours but D16 there is not, so I added it while last diode from main rail 12v disconnected, and noise disappeared.

thanks for your comments
 

hello
in the texas instrument doc: Designing Switching Voltage Regulators With the TL494 at page 11:
This illustrates the linear gain characteristics of the amplifiers over the active input range of the PWM comparator (0.5 V to 3.5 V).
what it means? is it's mean that both of set point or feed back output voltage on each of inverting or non-inverting input must be in range of 0.5v to 3.5v?
if yes, so it means that for making a variable switching power supply the set point @pin #2 must be in this range. is it true or not? and what will be happen if set point vary below 0.5v or over 3.5v ?

thanks
 

is it's mean that both of set point or feed back output voltage on each of inverting or non-inverting input must be in range of 0.5v to 3.5v?
Click to expand...

No, you are confusing amplifier output voltage range with input common mode range, it's specified as -0.3 to Vcc-2 V. In many popular TL494 designs, e.g. one input pin is connected to 5V Vref.
 
No, you are confusing amplifier output voltage range with input common mode range, it's specified as -0.3 to Vcc-2 V. In many popular TL494 designs, e.g. one input pin is connected to 5V Vref
Click to expand...

what do you mean output voltage range? do you mean pin#3?
on the other hand,for example if one of the comparator config as a difference amplifier (as picture below) Vout (pin#3) could not rise over 3.5 or under 0.5? is it true?
 

Yes, amplifier output voltage refers to TL494 pin 3, as in figure 12 of the linked application note.

Figure 12 clarifies, that the amplifier output voltage range is not restricted to 0.5 - 3.5 V. It can swing down to 0 and rise above 4 V. But that's not important because voltage levels beyond the PWM ramp voltage swing don't affect the PWM output.

 

After studding some books I thought that the role of diode is cut of spikes in secondary of transformer to 12v standby supplier voltage level. it is in addition to snubber circuit. so after removing diode,I increased snubber cap, noise was somewhat decreased. in all books described about designing RC/RCD snubber in primary side of trans not in secondary,and there is not any mention about this technique to cut spikes to a fix voltage:

1- Why there is spike in secondary side, while there is snubber in primary side?

2- Is it true that diode is for snubbing spikes, if yes what is the necessity of RC snubber in addition to diode, because RC snubber is a dissipative circuit for dissipate spike?

3- How to redesign RC or RCD snubber in secondary, is it same as primary side technique?

4- There is another snubber in primary side in addition to secondary, is it possible remove spikes just by redesign primary snubber?
 

FvM said:
Too bold to answer. (Review forum rules).
Click to expand...

sorry
what do mean?! you mean it's not allowed to bold questions?
I do not find any rule about the format of asking questions :-?
 

I do not find any rule about the format of asking questions
Click to expand...
Of course there are, but they don't mention questions all typed in bold letters. So I just express my private opinion about it. Emphasizing an important question might be useful, but what's the purpose of typing all questions in bold?

Secondly I notice that the snubber topic is completely new to this thread.

Nevertheless a short comment in technical regard:

The main purpose of snubbers in SMPS is to protect the switch transistor, the requirement only exist on the primary. Spikes across the secondary winding aren't a problem as such. Some developers might feel a need to trim waveforms on the secondary or fight EMI.
 

thanks

as I ask at the first post after removing diode the circuit make zzzz noise. now I think (as I said in last post ) it cut some spikes to fixed 12v standby supply (I can not check it, because I do not have scope), so as last post quastions, How to compensate the loss of diode? by redesign which one of snubber of primary or secondary, or by doing something else?
for example this typical circuit:
 

but every tl94 requires startup supply. so their should be a auxillary supply.of 12v . but i cant understand how your smps works without startup
supply
 

pradeep.g.belchada said:
but every tl94 requires startup supply. so their should be a auxillary supply.of 12v . but i cant understand how your smps works without startup
supply
Click to expand...

as I said it's a typical circuit that is not shown standby(start supply) in that. here there are many schematics of pc power supply that in some of them,stand by supply are not shown.I choose one just for show my mean. but all of them have mentioned diode(except a few).
 

that diode adds a additional power to the tl494 vcc. if auxiliary smps of 12v is powerfull then that diode is not necessary
 


Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…