Field effect tube constant current source heat

非常菜

Field effect tube constant current source heat [Copy link]

 

I have a constant current board based on 3810 field effect tube. After outputting for a period of time, it always gets very hot, and even the inductor of 3810 and the ground end is burned out. The GS waveform tested with an oscilloscope is as shown in the figure. It is not very good, and the cycle is unstable. Can any expert help me find out what is the cause? Thank you very much!

maychang

The picture is incomplete. The load is not shown, and the pulse generation circuit is not shown. It is impossible to guess.

chunyang

Upload the original image, don’t use photos, as they are not clear or complete.

非常菜

maychang posted on 2022-6-15 15:26 The picture is incomplete. The load is not shown, and the pulse generation circuit is not shown. It is impossible to guess.

Sorry, I just found the original picture. Can an expert give me some advice? Thanks.

非常菜

chunyang posted on 2022-6-15 16:14 Upload the original image, don’t use photos, they are not clear or complete.

I just found the picture, sorry, could an expert take a look, thanks

qwqwqw2088

How much load is there at the back?

L1 burnt out?

H12315

Just an off-topic question, is there any special purpose for having the IC3 feedback loop in front of the inductor? Isn't it usually at the back?

非常菜

qwqwqw2088 posted on 2022-6-15 17:00 How big is the load behind it? Is L1 burned out?

When the load is not greater than 2 ohms, L1 and 3810 sometimes burn together, and sometimes only 3810 burns, and the insulating gasket melts.

maychang

Very Green posted on 2022-6-15 16:56 Sorry, I just found the original picture. Can an expert help me with some advice? Thanks

I haven't found the datasheet for XL1509 (XLI509?), so I don't know its function.

But no matter what the function of the chip is, the whole circuit is definitely not the "field effect tube constant current source" as the title says. Because IC2A samples the output voltage from the right end of L3 through R8R12. This is a constant voltage output circuit.

maychang

Very Green posted on 2022-6-15 16:56 Sorry, I just found the original picture. Can an expert help me with some advice? Thanks

The waveform in the first post says that it is the voltage between GS. From the figure, the amplitude is 0~2.4V. 2.4V may be that the MOS tube is not fully turned on. This may be one of the reasons why the MOS tube heats up.

非常菜

maychang published on 2022-6-15 19:34 The waveform in the first post says that it is the voltage between GS. From the figure, the amplitude is 0~2.4V. 2.4V may be that the MOS tube is not fully turned on. This may be due to the heating of the MOS tube...

So the GS voltage needs to be raised to above 4V, right? This is the board diagram that the factory has been using for the constant current source of the electrolytic electrode. Have we been using it wrong? !

qwqwqw2088

How many MOS were burned?

Check whether the welding assembly component parameters are correct

maychang

Very dish published on 2022-6-16 07:13 Then the GS voltage needs to be increased to more than 4V, right? This is the board diagram that the factory has been using for the constant current source of the electrolytic electrode. Have they been using it wrong? !

This circuit is obviously a very old circuit, using self-excited oscillation. Such a self-excited circuit has a great influence on component parameters and has very poor stability. It is recommended to tear it down and redesign it.

Gen_X

It is hoped that the ground of IC6 can be floated to provide a higher driving voltage to ensure that T1 is fully turned on.

不亦心

maychang published on 2022-6-16 09:23 This circuit is obviously a very old circuit, using self-excited oscillation. For such a self-excited circuit, the influence of component parameters is very large, and the stability...

Weird design. This is not self-excitation, it uses a buck chip with built-in power tube to drive the MOS, which is lame. The capacitor is too large.

不亦心

Bu Yixin published on 2022-6-16 13:02 Unbelievable design. This is not self-excited, it uses a buck chip with built-in power tube to push the MOS, which is lame. The capacitor itself is too large

Self = Bootstrap

maychang

Bu Yixin posted on 2022-6-16 13:02 This is an incredible design. This is not self-excited, it uses a buck chip with built-in power tube to push the MOS, which is lame. The capacitor itself is too large

"This is not self-excited, it uses a buck chip with built-in power tube to extrapolate the MOS."

Which chip in the picture is the "Built-in Power Pipe Buck Chip"? Is it the XL1509 marked IC1?

1nnocent

Very dish posted on 2022-6-15 19:22 For loads not exceeding 2 ohms, L1 and 3810 sometimes burn together, and sometimes only 3810 burns, and the insulating gasket melts

The load is so small, how much output current is there?

1nnocent

Very dish posted on 2022-6-15 19:22 For loads not exceeding 2 ohms, L1 and 3810 sometimes burn together, and sometimes only 3810 burns, and the insulating gasket melts

Is this design designed to drive such a small load?

非常菜

maychang published on 2022-6-16 09:23 This circuit is obviously a very old circuit, using self-excited oscillation. For such a self-excited circuit, the influence of component parameters is very large, and the stability...

I cried, I have sold a lot of them, mainly because it is difficult to repair them later