This is a flyback switching power supply circuit built with UC2844 chip. After removing the 1000uF filter capacitor of the 5V main feedback output

西里古1992

This is a flyback switching power supply circuit built with UC2844 chip. After removing the 1000uF filter capacitor of the 5V main feedback output [Copy link]

 

 

After I powered it on again, it stopped working and kept on hiccuping. I guess there was a short circuit somewhere in the circuit? However, I used a multimeter to measure the circuit and found that there was no short circuit. I also replaced all the main components, but it still couldn't power on and kept on hiccuping. Can you provide some ideas?

This is the test waveform. The blue one is the 5V main feedback output voltage, and the green one is the driving voltage Vgs of the switch tube. Both of them are hiccuping.

Here is the circuit diagram

maychang

[The flyback switching power supply circuit built with UC2844 chip, after removing the filter capacitor 1000uF of the 5V main feedback output, it stopped working after powering on again and kept hiccuping]

This capacitor is a basic component of the flyback switching power supply and must not be removed.

Why don't you remove the power switch Q1? Or remove the 5.1 ohm resistor R107 (short circuit)?

maychang

It seems that you have not figured out how the flyback switching power supply works. In the flyback circuit, when the power switch tube is turned on, the energy provided by the DC power supply is stored in the transformer core (magnetic energy). When the power switch tube is turned off, the energy in the transformer core is converted into electrical energy and stored in the filter capacitor (C15, C6, etc.). Now that you have removed C15, the output voltage rises extremely quickly, and its rising speed far exceeds the ability of the feedback control circuit, so the chip enters protection (hiccups).

西里古1992

maychang posted on 2024-8-28 09:26 It seems that you have not figured out how the flyback switching power supply works. In the flyback circuit, when the power switch tube is turned on, the energy provided by the DC power supply is stored in...

No, the description here is incomplete. This circuit has been successfully tested before and only needs a 1000uF capacitor to work, so removing one to reduce costs, in fact, it is calculated that only 560uF is the critical point, this is 2000uF, and all hardware circuits are tested without short circuits. The hiccups still occur after reinstalling the capacitor

maychang

Sirigu1992 posted on 2024-8-28 09:42 No, the description here is incomplete. This circuit has actually been successfully tested before and only needs a 1000uF to work, so removing one reduces the cost...

[This circuit has been successfully tested before and only needs a 1000uF to work, so removing one to reduce costs, in fact, it is calculated that only 560uF is the critical point, and this is 2000uF]

You mean, there are two 1000uF capacitors, and it works even if one is removed? There are no two 1000uF capacitors in parallel in the picture.

Which capacitor did you remove this time? Please state the number, not the capacity.

maychang

Sirigu1992 posted on 2024-8-28 09:42 No, the description here is incomplete. This circuit has actually been successfully tested before and only needs a 1000uF to work, so removing one reduces the cost...

[Reinstall the capacitor and the hiccups still occur]

There are several possibilities:

Your capacitor was damaged during disassembly.

Other components on the board were damaged during the removal and installation of the capacitor.

………………

Why are you removing this capacitor?

西里古1992

maychang posted on 2024-8-28 09:52 [This circuit has actually been successfully tested before and only needs a 1000uF to work, so removing one to reduce the cost, in fact, it is calculated that only 560u ...

And C17 is also 1000uF

maychang

Sirigu 1992 posted on 2024-8-28 14:03 And C17 is also 1000uF

C17 and C15 are different. C15 does not pass through the inductor L1, but C17 does. Your control circuit gets the feedback signal from both ends of R20.

Which capacitor did you remove? C15 or C17?

西里古1992

maychang posted on 2024-8-28 14:30 C17 and C15 are different. C15 does not pass through the inductor L1, while C17 does. And your control circuit obtains feedback signals from both ends of R20. ...

The removed C15, isn't the L1 inductor used for EMI treatment?

maychang

Sirigu1992 posted on 2024-8-28 16:28 The removed C15, isn't the L1 inductor used for EMI treatment?

The L1 inductor is marked as 4.7uH in the diagram. The LC resonant circuit composed of 4.7uH inductor and C17 (1000uF) has a resonant frequency of 2322Hz, which is much lower than the operating frequency of your flyback switching power supply. The phase shift caused by this LC circuit cannot be ignored and will definitely affect the stability of the feedback loop.

If you remove C15, the frequency characteristics of your feedback loop will definitely change, which may be the reason why your switching power supply "hiccups". But you also said "it still hiccups after reinstalling the capacitor", I don't know about this, and I only mentioned two possibilities on the 6th floor.

maychang

Sirigu1992 posted on 2024-8-28 16:28 The removed C15, isn't the L1 inductor used for EMI treatment?

From the waveform in the first post, the output of your flyback switching power supply rises sharply from zero, then enters protection mode and the output is zero. It restarts after being zero for quite a while, which is called "hiccups". The reason for the "hiccups" is probably the change of the frequency characteristics of the feedback loop, which has become unstable.

不亦心

The optocoupler is powered by C15, and the 431 feedback is powered by C17.

beyond_笑谈

"After removing the 1000uF filter capacitor of the 5V main feedback output, it stopped working after powering on again and kept hiccuping"

This capacitor is simply understood as energy storage, not filtering.

西里古1992

beyond_笑谈Published on 2024-8-29 08:22 "After removing the 1000uF filter capacitor of the 5V main feedback output, it stopped working after powering on again and kept hiccuping." This capacitor is simple...

Yes, the one behind it is LC filtering

西里古1992

maychang posted on 2024-8-28 17:17 The L1 inductance is marked as 4.7uH in the figure. The resonant frequency of the LC resonant circuit composed of 4.7uH inductor and C17 (1000uF) is 2322Hz, which is much lower than your flyback switch...

Yes, now the capacitor appears and the hiccups still occur after adding it. It feels like a component is damaged, but the MOS tube, TL431 and power management chip have all been replaced.

西里古1992

maychang posted on 2024-8-28 17:20 From the waveform in the first post, the output of your flyback switching power supply rises sharply from zero, then enters protection and the output is zero. After being zero for quite a while...

So although both are 1000uF, C15 and C17 have different functions, right? C15 is the energy storage capacitor of a typical flyback switching power supply, and C17 is an LC circuit used for filtering with L1.

maychang

Sirigu1992 posted on 2024-8-30 16:33 That is to say, although both are 1000uF, C15 and C17 have different functions, right? C15 is the energy storage capacitor of a typical flyback switching power supply, and C17 is paired with L1...

[Although both are 1000uF, do C15 and C17 have different functions? 】

It is not easy to say that. C15 and C17 both have the function of energy storage and filtering.

However, the two capacitors have different effects on the feedback loop characteristics. The reason for the difference lies in the inductor L1.

I guess the original author of the circuit wanted to add a stage of LC filtering after C15 to achieve lower ripple.