TOP266EG: Low efficiency & Thermal shutdown

Posted by: vijay7000 on 01/07/2017

Hello All,

I have desined a multi output SMPS using PI Expert Software of Ratings 23V/2A(output 1) and 5V/250mA(output 2). After assembly the components on PCB I connect the SMPS to 230VAC and read the output voltage. output voltage was as :

Output 1 : 34V(No Load)

Output 2 : 13V(No Load)

I cheched the regulator (TL431) circuitary which was working fine. I observed the reference voltage 2.6V on reference pin of TL431. But output voltage was too high compare to design specs. I assumed that there is no sufficient current on control pin so I reduced the Resistore value R8(see the attached schematic) which is connected to the anode of optocoupler from 14.3K to 8.6K by adding a 22K resistor parallel to R8.

Result:

Output 1: 23.1V

output 2: 5V

Now Circuit is working fine with "No Load"

Problem:

when I connect it to a load, it works with low efficiency. I checked it with 90VAC - 260VAC drawing 750mA, 1A, 1.5A. 2A from output 1(output 2 is No Load MOde). I record the many readings with different input VI and output VI

result : Efficiency of SMPS vary from 38% to 49%.

A lot of heat generated.

Thermal shut down occurs and circuits is working in On-OFF loop.

I have used the components value and transformer winding as suggested by PI Experts. I do not have the prior experience about SMPS. Can anybody please help me how to resolve this problem. I will be grateful for your kind support.

Thank in advance.

Vijay

Files

TOPSwitch-JX_PI_Sch.pdf (61.23 KB)

TOPSwitch-JX_PIDesign1_Tr.pdf (91.41 KB)

Comments

PLease help !

Hi Vijay,

Thank you for choosing Power Integrations.

Are you using a heatsink on the TOP266? If not you will need to use a heatsink. Could you also send a phot of the power supply unit? Could you also send the PIexpert design file? I would like to help you check it.

Hello PI-Traveler,

Thanks for your reply. I was busy with another project and this project was in pending. But now I want to continue this project and request you to reopen this topic.

current status of the above mentioned power supply is:

*Supply is working with the following output power
output 1. 23V/600mA (Designed specs 23V/2A)
output 2. 5V/250mA (Designed specs 5V/250mA)
but when output current of channel 1 is increased above 600mA, ferrite core and switch is started to heat and after few minutes switch got shutdown. At 2A output current it goes to ON/OFF loop frequently. Looping time depends upon the output current.

It seems like core goes in saturation mode when more than 600mA current drawn.
I have used the same core part no(B66317GX127) which PIExpert has suggested. This core is ungapped core but most of people recommend gapped core for flyback SMPS.

Please see the attached image of PS unit and PI Expert design file.

I am not expert in SMPS technology. I, Therefore, request you please help me to find out the problem so that I can achieve the efficiency and stability of this PS unit as per PI Expert design.

Thanks in advance,

Regards,
Vijay

Attachment	Size
PS_Board.jpg (959.9 KB)	959.9 KB
TOPSwitch-JX_PIDesign1.uds (1.03 MB)	1.03 MB

Hi Vijay,

It's good that you mentioned that you did not gap the core. Your transformer core might be saturating.

Did you measure the primary inductance of your transformer?

If you followed the transformer construction as suggested by PIexpert your inductance value will be much higher than the recommended inductance if the core is not gapped.

This is the possible reason for the heat you are observing.

To verify, you can measure the current flowing through the TOP266 MOSFET using a current probe. The ideal wavshape you should see is a triangle with a linear ramp. A ramp which curves up is an indication that your transformer is saturating.

I suggest that you gap your core until you meet the recommended inductance.

Hope this helps.

Hello PI-Traveler,

Thanks for your valuable reply.
I did not measure the inductance of primary winding. Just I winded the transformer with the parts recommended by PIexpert. PIexpert recommends core part no(B66317GX127) which is ungapped core. So which of the followings would be better to maintain the required inductance?
1. should I keep air gap area manually in this ungapped core part no(part no(B66317GX127))? by this way gap area would be present at all three pole.
or
2. should I used already machined gapped core like B66317G0250X127? with this core, gap area would be at centre pole only.

Unfortunately, I do not have current probe. so I can not observe the waveform. I will try with proper primary inductance.

Thank you very much.

Regards,
Vijay