• State TI Thinks Resolved
  • Locked Locked
  • Replies 17 replies
  • Answers 1 answer
  • Subscribers 52 subscribers
  • Views 1257 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPA548: Output voltage drop for 0,5V when load attached

Vlatko Pistolov
Prodigy 160 points
Part Number: OPA548

HI,

I would like to use the OPA548 as power supply with regulated voltage output from 0 to 5V and current to 3A. I am using the reference schematic from SBOS070C TI document with addition pwm adjustment for the output voltage. When i set the output (for example 4V) voltage and attache the resistor 3,3ohm to simulate load of 1,21A the output voltage drop to 3,5V. My needs is to be able to emulate lipo battery and to privide correct voltage with current support up to 2,5A. Since i get output voltage drop i can't hold output stable. I have set the current limit to 4A and power supply is 12V/5A. 

What could be reason the OPA 548 to not hold the voltage output when we attache the load that require less current then the max current that OPA548 can deliver?

Best Regards,

Vlatko

  • 0
    Guru 140200 points
    Hi Vlatko,

    probably the OPA548 overheats and goes into thermal shutdown.

    With an output voltage of 4V, 12V - 4V = 8V will drop across the OPA548. 8V times 1.21A gives a heat dissipation of 9.7W. Without a heatsink the junction to ambient thermal resistance of 30°C/W must be taken into consideration. So, the die will heat up to 9.7W x 30°C/W = 291°C over the ambient temperature. This demonstrates that you need to use a big heatsink to remove all the heat and prevent the OPA548 from going into thermal shutdown.

    Another remedy is to use a smaller supply voltage. At an output current of 3A the supply voltage must be at least 4V higher than the output voltage. So, you could reduce the supply voltage to 9V.

    Last but not least, you could parallel two OPA548 as shown in figure 36 of datasheet.

    Kai
  • 0 in reply to kai klaas69
    Prodigy 160 points
    Hi Kai,

    Thanks for the fast replay. I also notice in my tests that the input voltage on pin 1 (for adjusting the output voltage) drop when i attache the same load. In my test i have set on pin 1 4.21V and get 4,26V output where my G is 1,01. But when i attache the load with 3,3ohm on output i get 3,77V and on input pin 1 a get 3,97V. This mean the input voltage also drop for the reason which i do not understand. What could be the reason input voltage on pin 1 to drop?
    Is the OPA4548 has some influence on this?

    BR,
    Vlatko
  • 0 in reply to Vlatko Pistolov
    Guru 140200 points
    Hi Vlatko,

    is the OPA548 becoming hot? Do you use a heatsink?

    Kai
  • 0 in reply to kai klaas69
    Prodigy 160 points

    opa548_test.pdfHi Kai,

    Yes i use two heatsink back to back and looks that the OPA548 it is the hot but i can touch them without problem which mean is not to hot.

    I attach my schematic for the power supply part.

    BR,

    Vlatko

  • 0 in reply to Vlatko Pistolov
    Guru 140200 points
    Hi Vlatko,

    have you checked the "thermal shutdown status" on the E/S pin?

    Kai
  • 0 in reply to kai klaas69
    Prodigy 160 points
    Hi Kai,

    I do not have check them. I will check now.

    BR,
    Vlatko
  • 0 in reply to Vlatko Pistolov
    Prodigy 160 points
    Hi Kai,

    I have 4,8V on E/S pin when i enable the OPA548. Looks that i do not have proper control (enable/disable) for output pin. I disable them with the optocoupler with logic "0" but when i enable them with logic "1" i force 5V, which according the document for OPA548 i should left open or pulled to +2,4V .
    Maybe this makes OPA548 to go in termal shutdown when i attach load.

    BR,
    Vlatko
  • 0 in reply to Vlatko Pistolov
    Prodigy 160 points
    Hi Kai,

    Looks that i am wrong about my last post. Thermal shut down status should be 350mV when it is happen but in my case this pin has 4,8V. This mean that OPA548 does not going in to the thermal shut down when i attach load.

    BR,
    Vlatko
  • 0 in reply to Vlatko Pistolov
    Guru 140200 points
    Hi Vlatko,

    please let the E/S pin float during the test. Only connect a DVM to this pin. Which voltage is the pin showing before and after connecting a load to the output of OPA548?

    Kai
  • 0 in reply to kai klaas69
    TI__Expert 6275 points

    Put oscilloscope on output of OPA548 when you connect the load. If output is oscillating immediately upon loading we need ot help you troubleshoot further...

  • 0 in reply to Tim Green1
    Prodigy 160 points

    HI,

    I left to floating the E/S pin and looks that i have the same results. This are the results from my measurements:

    Load: 3,3Ohm

    input pin 1 : 4,21V

    output pin 6: 4.27V

    input pin 1 after attach load: 4,12V

    output pin 6 after attach load: 3.82V

    I will attach oscilloscope on the output and send the result.

    BR,

    Vlatko

  • 0 in reply to Vlatko Pistolov
    TI__Genius 11135 points
    Vlatko

    We haven't heard back from you; were you able to resolve your issue?
    If you need additional help just post another reply below.

    Thanks
    Dennis
  • 0 in reply to Dennis Goeke
    Prodigy 160 points

    Hi Dennis,

    Sorry for my late replay. I did not solve the problem. I did some check with oscilloscope on the output pin and i did not find some instability or oscillations. The test was with fix load (resistors with 3,3ohm) but still i have the problem to hold the voltage level on the output. This project should be like battery emulator to generate stable voltage output in range from 0 to 4.2V, controled wotj pwm so i can do the automated test on the mobile phone. But this output voltage drop change the emulated battery level so i can't trust this solution. 

    What i am training now is to make new PCB with adding the buffer for the pwm so i can have adapted voltage input for the OPA548T. Also i am not sure if the OPA548T can handle gain G=1,01 well without to start oscillate. I think to change the gain to 2 or 4. 

    Do you think that this voltage drop could be related with the gain of the OPA548T?

    BR,

    Vlatko

  • 0 in reply to Vlatko Pistolov
    Guru 140200 points

    Hi Vlatko,

    check this circuit:

    vlatko.TSC

    Provide really good cooling for the OPA548! Take the OPA548 in TO-220 package and mount a 5K/W heatsink. Coat the contact surface with heat transfer compound, but apply it very sparingly!

    The OPA548 must dissipate 6V x 1.177A = 7.1W. The die of OPA548 will heat up 7.5K/W x 7.1W = 53K over ambient temperature. So, when your heatsink sees an ambient temperature of 30°C, the die will heat up to 83°C.

    Kai

  • 0 in reply to kai klaas69
    Prodigy 160 points

    Hi Kai,

    Thanks for the schematics. I will modify my PCB according this schematics and will come with results.

    BR,

    Vlatko

  • 0 in reply to Vlatko Pistolov
    TI__Genius 11135 points
    Vlatko

    Did the circuit above work for your application? If you need more help just post another reply.

    Thanks
    Dennis
  • 0 in reply to Dennis Goeke
    TI__Genius 11135 points
    Vlatko

    We haven't heard back so we are going to assume this resolved your issue. If not, just post another reply.

    Thanks
    Dennis