Solution Finder Get Tech Support

Weired issue with over power protection

Posted by: Dhaval Lalani on

Hi all,

I have designed power supply using TOP256EG for 24V-2A output. There is weired issue observed while testing prototype.When I test it with loose circuit (No alluminium housing ) output over power shutdown works perfectly, but testing same circuit with metal enclouser this function doesn't work even till double output power (4A output load)

Value of x pin resistor is set as suggested by PI Expert. In both test case all other features work fine except OPP.

I tried it with grounding metal body to earth too, but no effect.

Is there any issue related to EMI radiation.

Files

Comments

Submitted by PI-KSAN on 02/24/2016

At 4A Output load , is the output voltage is Stable? How did you design your transformer? What's the AC input condition?

Thanks for support,

At 4A output voltage is stable at 24VDC. (Just shut down after some time due to over tempreture)

 

All component values and transformer design are same as suggested by PI-Expert.

 

I have designed it for universal input but tested all at 230VAC. (At 110VAC feature works fine, auto restart at 4A load)

 

But I want to know is there metal housing cause an effect on features, like coupling noise or something simmilar.Then how to solve it.

 

 

Hello Dhaval,

 

Sorry for the late reply. As the schematic you show, i didn't see OPP protection circuit on it. The protection circuit on the schematic should be the OVP circuit.  There is a detailed section about the Implementing Over-Power Protection (OPP) in our Apps note, you can follow that on page 23. below is the link.

https://ac-dc.power.com/sites/default/files/product-docs/an47.pdf

 

Another way to debug it is to capture the waveform at the bias winding voltage to see what's difference if the performance between inside and outside the metal enclouser. it seems that the temperature is really high inside the enclouser. if you can measure the temerature around the bias winding, it will also give us useful information to see if the difference is caused by the high temperature.

 

By the way, the PI EXPERT you run is for TOP-HX. i think you are using TOP-JX. 

Submitted by Dhaval Lalani on 03/08/2016

Thanks to support......

I was unaware of that protection mention in fig 22 of AN47. I will try and check it. But need some direction to selecting component values.

It is difficult to capture waveform when circuit is mounted in encloser, but will try it get some capture.

I dont think tempreture cause any issue, because as I turn on supply and check for protection, its not working. During very small time device not get heated.

I am using HX and JX for different products, but planing to use JX for this project to. Should I do major changes in design while transfering from HX to JX?

I will be back as soon as possible with test result.

Submitted by Dhaval Lalani on 03/09/2016

In reply to by PI-YY

Thanks to support......

I was unaware of that protection mention in fig 22 of AN47. I will try and check it. But need some direction to selecting component values.

It is difficult to capture waveform when circuit is mounted in encloser, but will try it get some capture.

I dont think tempreture cause any issue, because as I turn on supply and check for protection, its not working. During very small time device not get heated.

I am using HX and JX for different products, but planing to use JX for this project to. Should I do major changes in design while transfering from HX to JX?

I will be back as soon as possible with test result.

Submitted by PI-YY on 03/09/2016

In reply to by Dhaval Lalani

Hello Dhaval,

 

Follow the apps note guidline to select the component, The OPP feature relies on the imperfect coupling between the secondary and the bias winding. As the output load increases the bias winding voltage rises due to the effect of leakage inductance. you may need to capture the bias winding volatge waveform to know the value of the components you need.

 

When transfering from HX to JX, no major changes is needed. if you run the PIexpert for JX, you can see samll change of the value , for example the clamp value. i am not sure if you optimize any components value based on your design. if so, just run the JX PIexpert and compare if everything matches.

 

Best Regards

 

Submitted by Dhaval Lalani on 03/16/2016

In reply to by PI-YY

Thanks YY,

I have tried to capture bias waveform but it found clean after rectification. Also I found peice to peice variation in problem that I mention in first question.

But that OPP works well during my trial and error method. I will fixe component values aftersome trials.

Thanks again.........