• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 1 answer
  • Subscribers 61 subscribers
  • Views 370 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.

LM5175: heat problem

gha
Prodigy 100 points
Part Number: LM5175
Other Parts Discussed in Thread: LM5176

Dear Ti Support!

We are working on a power supply with LM5175:

- input: 19-33V

- output: 24V @ 9 A

The output is working fine in buck and boost mode.

The only problem is that the LM5175 draws about 85mA current from 26V input voltage and it gets hot (about 60 C) with no load condition.

Do you have any idea what can cuase the big current consumption?

We tested on 2 borad, guess its not a soldering issue.

Thank you

Best Regards,

Gabor

  • 0
    TI__Guru 58215 points

    Hello Gabor,

    Which mode are you using? CCM? How much current does it draw when you change the input voltage? E.g. to 33V?

    If you use CCM, some current to supply the permanently switching drivers is normal. The drivers are supplied from the input voltage through a linear regulator which regulates to 7.35V which means a power dissipation of (Vin - VCC)*Idriver

    Is the thermal pad of the device soldered on a metal area?

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 100 points

    Dear Brigitte,

    Thank you for your reply.

    Now we are using CCM mode (we tested it with DCM mode in the past and it was a little bit less current draw)

    We have two working prototypes and we did measurement with different input voltages.

    PCB1 :
31V(max) 157mA
30V      148mA
29V 	 122mA
28V      120mA
27V 	 135mA
26V	 161mA		
25V 	 193mA
24V      236mA
23V      278mA
22V      161mA
21V      175mA
20V	 192mA

Vvcc = 7.35V

PCB2:
31V(max) 196mA
30V      272mA
29V 	 217mA
28V      197mA
27V 	 167mA
26V	 162mA		
25V 	 125mA
24V      141mA
23V      189mA
22V      149mA
21V      169mA
20V	 192mA

Vvcc = 7.32V

    Yes, the thermal pad soldered down to GND pad and its conencted via vias to the GND plane

    Can we send you the schematic and layout? Could you send me please your e-mail address?

    Thank you
    Best Regards
    Gabor
  • 0 in reply to gha
    TI__Guru 58215 points

    Hello Gabor,

    I sent you a friendship request for sharing data privately.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    TI__Guru 58215 points

    Hello Gabor,

    The schematic shows LM5176. Are you using LM5176 or LM5175?

    BIAS is supplied from VOUT.

    For current consumption tests, it would be great, if you could try to supply BIAS from an external supply. If you can supply e.g. 9V, it would be interesting to see if the device still gets hot. If not, please check if the device gets hot again when this voltage is increased to 24V.
    If the device gets hot with 9V input on BIAS, there is another root cause for the increased temperature and not the internal LDO.

    As you have as well 12V in the system, you could as well check using this supply for BIAS and check the temperature then.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    TI__Guru 58215 points

    Hello Gabor,

    For the schematic review, which inductor are you using? Please mention inductance and saturation current.

    Please connect DITH to GND at the moment it seems to be open as the capacitor is not populated.

    General comments to the schematic if you plan a new layout:
    - I recommend gate resistors in case they are needed for EMI tests for example.
    - I recommend snubbers next to D4 and D6 in case they are needed.
    - Place R171 parallel to C55 for disabling dithering.

    The rest can only be checked with the knowledge about the inductor.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 100 points

    Dear Brigitte,

    We are using LM5175.

    The BIAS supply will be checked next week and I will come back to you.

    Tested the DITH connection to GND and it do not helped.

    We are using the following inductor: SRP1265A-2R2M

    Maybe we could change to 3.3uH, what do you think?

    For the next PCB version we will add the mentioned EMI components.

    Thank you

    Best Regards

    Gabor

  • 0 in reply to gha
    TI__Guru 58215 points

    Hello Gabor,

    Please check the switching frequency of your system with the switching frequency you want to achieve.

    The inductor looks good. If you want you can increase it to 3.3uH which will reduce the ripple current. But the inductor is not getting hot, right?

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    TI__Guru 58215 points

    Hello Gabor,

    As I did not hear back from you, I expect you have been able to solve the issue. If this is not the case, please post below or create a new thread if this one is already locked.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 100 points

    Hello Brigitte,

    Sorry for the late response, we are working on the new PCB design with your remarks.

    We realized that the 2 pcs TVS diode was missing between BOOT and SW pins.

    Some weeks before we replaced the LM5175 to a new one and it worked for one day with low current consumption, next day it jumped back again to high current. I hope that the TVS diode will solve our high power consumption and heat issue.

    Thank you for the support, I will come back to you after the new PCB proto version.

    We changed to 3.3uH the inductor.

    Best Regards,

    Gabor

  • +1 in reply to gha
    TI__Guru 58215 points

    Hello Gabor,

    Thank you very much for your feedback. Crossing fingers that the problem is gone with the Zehner diodes on the BOOTx pins.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 100 points

    Hello Brigitte!

    New board ready, added zener diodes and now the current consumption looks ok at no load and LM5175 not overheated, so it looks ok, thank you.

    Other topic: the output ripple at no load is about 100-200mV P-P, do you have any suggestion how to reduce, we are using now 3.3uH inductor

    Thank you

    Best Regards,

    Gábor

  • +1 in reply to gha
    TI__Guru 58215 points

    Hello Gabor,

    The amount of output capacitance defines the ripple, so if you want to reduce the ripple, increase the output capacitance.

    Please check again stability if you increase the output capacitance. Normally the reaction is slower when increasing the output capacitance and therefore the stability is better, but it is always necessary to check especially when using capacitors with higher ESR.

    Best regards,
    Brigitte