SMPS Always Trying to Restart

[dmerry03]

Hello,

I am currently working on a galvanically isolated 15V Switch Mode Power Supply. I am having an issue where the IC I am using is constantly trying to restart after about 1 second. This is demonstrated in the oscilloscope shot below.


Here is an image of my schematic:


It is not shown, but there is a inrush current limiter and a bridge diode to the left, where the "HV" Net Label is.

I am at a loss for what is going on. I do not have extensive experience with these power supplies, but I have tried the following:
1. I removed the D5.b diode and C7.b to determine whether there was a short on the output. The restarting behavior persisted.
2. I used a thermal camera to locate shorts anywhere on the circuit. To no avail.
3. I putting a large Cap (330uF) in parallel with EC11.b, thinking maybe there was a high current draw when the LN9T11 began switching the primary, but this did not help.

Thanks in advance for your help.

[jmelson]

The design is supposed to trickle current into EC11 until the chip has enough voltage to start, and then D9 is supposed to keep it above the undervoltage lockout setpoint.  It sounds like D9 is not able to keep the chip powered up.  When  not powering a load, the current in the transformer may not develop enough voltage in the winding connected to D9.  Possibly, you have the winding connected to D9 backwards.  Flyback supplies are sensitive to having that winding phased correctly.
Jon

[magic]

What are those scope traces showing?

[TimNJ]

Do you have a full datasheet for this IC? Only can find an abridged one. English preferable.

Also, what is the yellow waveform? VDD pin? Quite a low VDD voltage, which is why I'd like to know the VDD under-voltage threshold and start threshold is. Did you build the transformer yourself? If the coupling of the auxiliary winding (to secondary winding) is poor, it may not be able to provide much current once the IC switches over from charging via HV.

[dmerry03]

The design is supposed to trickle current into EC11 until the chip has enough voltage to start, and then D9 is supposed to keep it above the undervoltage lockout setpoint.  It sounds like D9 is not able to keep the chip powered up.  When  not powering a load, the current in the transformer may not develop enough voltage in the winding connected to D9.  Possibly, you have the winding connected to D9 backwards.  Flyback supplies are sensitive to having that winding phased correctly.
Jon

Thanks Jon. I had this SMPS designed by company in China, then I integrated the design into my product and localized the BOM (aside from the custom transformer). They have been telling me to check that winding as well, but I am certain I integrated it into my design correctly. Maybe you could tell me if I am wrong . . .

Their PCB Layout


My PCB Layout


So if I attach a more substantial load, would that feedback winding develop a higher voltage?

[dmerry03]

Do you have a full datasheet for this IC? Only can find an abridged one. English preferable.

Also, what is the yellow waveform? VDD pin? Quite a low VDD voltage, which is why I'd like to know the VDD under-voltage threshold and start threshold is. Did you build the transformer yourself? If the coupling of the auxiliary winding (to secondary winding) is poor, it may not be able to provide much current once the IC switches over from charging via HV.

Hi Tim. Unfortunately no, I will work to get it. I had the whole system designed by a firm in China. You are correct on the yellow waveform from the scope shot

[magic]

What are those scope traces showing?

Nevermind, they are marked on the schematic.

Blue is VDD. As others pointed out, VDD clearly isn't receiving power from the AUX winding and the chip UVLO's at 10V and restarts at 20V.

Yellow is the output and it hints at your problem: the output appears to be regulated at 4.6V for some pathological reason. This is ⅓ of the nominal output and hence AUX is also ⅓ of nominal. That's why VDD isn't being charged. Check the whole regulation feedback path, starting with resistors near TL431. (My bet: R43.b is open circuit.)

Oh, and by the way,

3. I putting a large Cap (330uF) in parallel with EC11.b, thinking maybe there was a high current draw when the LN9T11 began switching the primary, but this did not help.

It did change something, it lengthened the periods the IC stays on.
A useful bit of information to notice

[dmerry03]

Do you have a full datasheet for this IC? Only can find an abridged one. English preferable.

Also, what is the yellow waveform? VDD pin? Quite a low VDD voltage, which is why I'd like to know the VDD under-voltage threshold and start threshold is. Did you build the transformer yourself? If the coupling of the auxiliary winding (to secondary winding) is poor, it may not be able to provide much current once the IC switches over from charging via HV.

Hey Tim, here is the English datasheet for the LN9T11.

[dmerry03]

What are those scope traces showing?

Nevermind, they are marked on the schematic.

Blue is VDD. As others pointed out, VDD clearly isn't receiving power from the AUX winding and the chip UVLO's at 10V and restarts at 20V.

Yellow is the output and it hints at your problem: the output appears to be regulated at 4.6V for some pathological reason. This is ⅓ of the nominal output and hence AUX is also ⅓ of nominal. That's why VDD isn't being charged. Check the whole regulation feedback path, starting with resistors near TL431. (My bet: R43.b is open circuit.)

Oh, and by the way,

3. I putting a large Cap (330uF) in parallel with EC11.b, thinking maybe there was a high current draw when the LN9T11 began switching the primary, but this did not help.

It did change something, it lengthened the periods the IC stays on.
A useful bit of information to notice

Hey Magic, thanks for the insight here. Your last statement makes sense now. I couldn't quite understand what I was seeing by changing that cap.

I did check that feedback path and everything matches the samples the company gave me. I looked for tombstoned resistors and everything, no dice. I guess something is wrong with the auxiliary transformer winding . . . but everything I did matches the samples the designer sent me . . .

[jmelson]

Thanks Jon. I had this SMPS designed by company in China, then I integrated the design into my product and localized the BOM (aside from the custom transformer). They have been telling me to check that winding as well, but I am certain I integrated it into my design correctly. Maybe you could tell me if I am wrong . . .



So if I attach a more substantial load, would that feedback winding develop a higher voltage?

Yes.  I can't comment on the phasing of the auxilliary winding, though.
Jon

[jmelson]

I did check that feedback path and everything matches the samples the company gave me. I looked for tombstoned resistors and everything, no dice. I guess something is wrong with the auxiliary transformer winding . . . but everything I did matches the samples the designer sent me . . .

I'd power the output of the supply with a bench supply, and watch the LED current on the opto.  Then, adjust the supply voltage and see where the TL431 starts pulling current through the LED.  Do this with no mains power coming into the supply.  Then, also verify the phototransistor side of the opto is working.  You couid probably just jump over the 15 V bench supply to EC11 to power up that side for testing.

Jon

[magic]

Hey Magic, thanks for the insight here. Your last statement makes sense now. I couldn't quite understand what I was seeing by changing that cap.

I did check that feedback path and everything matches the samples the company gave me. I looked for tombstoned resistors and everything, no dice. I guess something is wrong with the auxiliary transformer winding . . . but everything I did matches the samples the designer sent me . . .

Maybe the feedback resistors are swapped, then? I'm not sure what else could limit the output voltage to 4.5V for the entire time when the controller appears to be operating (discharging its supply cap). But as long as the secondary voltage is clamped near 5V, the auxilliary will also be way lower than normal.

The idea above is not bad: apply 5V or 12V to the output, verify if TL431 reference input is less than 2.5V, check if its output is high, check if there is any excess current through R32 that doesn't flow through R37. Maybe you missed something, maybe the chip is blown, maybe it's TL432, maybe you screwed something up while transferring their design to your board and still can't see it.

BTW, R37 seems a bit high in value because it only passes 100µA but the TL431 may sink 400µA even if REF is less than 2.5V. You could bodge an extra 1kΩ on top of R37 and see if it helps.

The reason I think something is wrong near TL431 is because a voltage like 3.5~4.5V is the minimum for such TL431+opto circuits to still sink any current through the opto. It looks like as soon as this minimum supply threshold is reached (and no earlier), the opto lights up and throttles the PWM.

Another thing I could theoretically imagine is some semiconductor or insulation breaking down and clamping transformer voltage, but then the PWM would run at full throttle and pump several watts of power into the transformer and all those watts would be getting dissipated somewhere.

[Terry Bites]

Hiccups?

[timenutgoblin]

Looking at the datasheet, it says that the start-up voltage is 21VDC which corresponds to your screenshot when the IC resets/restarts. Try replacing EC7.b with a smaller value such as 68µF or 47µF for minimal load (R23.b 4.99k ohms). Also R33.b is 2 ohms. According to the datasheet it suggests 0.75 ohms (R_CS current sense resistor). Perhaps pin 4 of the switching IC is seeing too high voltage on the sense pin? Can you measure pin 4 and VCC of the switching IC? Try reducing R33.b to 0.75 ohms (2 x 1.5 ohms in parallel). I think the device is detecting OCP due to current sense voltage on pin 4. A phasing issue with the secondary or auxiliary winding is also possible. It's possibly one of the windings, but not both.

Try reducing capacitor EC7.b first and then try reducing resistor R33.b next.

[magic]

R33 limits peak primary current and it should be set to a threshold below core saturation. If the PSU is designed to provide less power than the IC's maximum capability, the transformer may be smaller and R33 larger than in the reference design.

I humbly suggest that neither input power nor output load is the limiting factor, because output voltage rises from 2.5V to 4.5V instantly before it gets stuck.

It also falls down very distinctly when the PWM runs out of power, indicating that during the time when it's stable, the PWM is still operational and apparently in regulation.

[xavier60]

The yellow waveform shows that the PSU is actually going into regulation briefly before the IC's UVLO is triggered.
Although the schematic shows the AUX winding phased correctly, maybe it isn't. As others have mentioned.

[timenutgoblin]

R33 limits peak primary current and it should be set to a threshold below core saturation. If the PSU is designed to provide less power than the IC's maximum capability, the transformer may be smaller and R33 larger than in the reference design.

I humbly suggest that neither input power nor output load is the limiting factor, because output voltage rises from 2.5V to 4.5V instantly before it gets stuck.

It also falls down very distinctly when the PWM runs out of power, indicating that during the time when it's stable, the PWM is still operational and apparently in regulation.

It's interesting to note that the VCC voltage rises up to 20VDC, but then drops below 10VDC before the IC restarts. There isn't any indication that D9.b diode conducts to charge EC11.b capacitor. The capacitor only appears to be charging via the HV resistors. I wonder if R29.b resistor has gone open circuit?

As for the output voltage, I also noticed that it remains steady at 4.6VDC before collapsing to 2.5VDC then restarting.

Theoretically, if there was a 5k ohm resistor in parallel with R35.b resistor then that would explain the output voltage of 4.583VDC. The schematic suggests otherwise, though.

Interesting problem.

[xavier60]

I have just noticed that it should be regulating at 15V. With the output being limited at 4.6V, the AUX winding will not develop enough voltage to sustain the IC.
Check to see what the REF pin on the TL431 comes up to.
If C14.b was shorted, the output would regulate at about what it appears to be.

[xavier60]

If it was just a regulation problem, the output should be holding for longer at the 4.6V.
EC11.b is unusually low in value.