• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 2348 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LM5085: Output voltage and current drops after some time

Tadeusz Tatery
Intellectual 310 points
Part Number: LM5085


Hello,

I have a big issue with step-down DC-DC converter based on LM5085.
I used design shown as typical application in datasheet, however with different components since my working assumption is a little bit different.

Assumptions for calculations:
Vout=5V
Vin=19.5V
Vin_max=30V
Vin_min=10V
Iout_min=0.5A
Iout=4.5A
Fs=300kHz
FET turn-on=11ns
FET turn-off=17ns
FET td=6ns
td=0.000000056    
Vos=5mVp-p

First issue, with calculated R1 (Radj) that is 1k91 I'm not able to achieve expected max current (4,5A). I had to change it to 6k2 to be able to get max current. But now, after few seconds of stable work (Vout=~5V, Iout=~4,4A) voltage drops to 4,7V and current decrease to 4A. Additionally L1 is getting hot.

Second issue is voltage ripple at the output which is about 80-100mVp-p.

I've already tried:
- removed C3
- removed C8
- changed R3 to 68k, 45k
- R1 replaced by variable resistor (10k)

When R1 was replaced by variable resistor connected with wires to the PCB then experimentally I was able to achieve expected values at the output but only for about 120-200s then voltage dropped to 4,7V and current to 4A. When I replaced variable resistor mounted as R1 by solid one with value 6k2 (with this value I was able to achieve expected output current) then output voltage and current were OK for 20-30s. After this time output voltage dropped to 4,7V and current to 4A. Oscilloscope plots for both variable and solid resistors are attached below.

Questions that I'm not able to answer myself and I'd like to ask you:

1. Why it is needed to use R1 as big as 6k2? I calculated it few times using equations from datasheet and it should be 1k91 (if I used all input values as in datasheet then I get the same values as in datasheet example)

2. Why output voltage and current dropped after some time?

3. What should I do to decrease ripple voltage at the output?

Designator Comment Footprint Mfr. No
U1 LM5085 MSOP8 LM5085MY/NOPB
Q1 FDMC86139P Power-33-8 FDMC86139P
L1 15uH/5A SRR1210 SRR1210-150M
C1 1000p C0603 710-885012206083
C2, C3 100u/6V C1206 T55A107M6R3C0045
C4 22u/6V C1206 JMK316AB7226KL-TR
C5 3300pF C0603 C0603X332K5RECAUTO7411
C6 0u47/35V C0603 CGA3E3X5R1H474K080AB
C7, C8 100u/35V C7361-38 (V) 80-T495X476M35ATE230
C10 1u C0805 CC0805ZKY5V9BB105
C11 0u1 C0603 710-885012206071
RSEN1 0R01 R0805 RCWH080510L0JQEA
R1 1k91 R0603 RK73H1JTTDD1911F
R_FB2 3k4 R0603 RK73H1JTTDD3401F
R_FB1 10k R0603

RK73H1JTTDD1002F

Designator Comment Footprint Mfr. No
U1 LM5085 MSOP8 LM5085MY/NOPB
Q1 FDMC86139P Power-33-8 FDMC86139P
L1 15uH/5A SRR1210 SRR1210-150M
C1 1000p C0603 710-885012206083
C2, C3 100u/6V C1206 T55A107M6R3C0045
C4 22u/6V C1206 JMK316AB7226KL-TR
C5 3300pF C0603 C0603X332K5RECAUTO7411
C6 0u47/35V C0603 CGA3E3X5R1H474K080AB
C7, C8 100u/35V C7361-38 (V) 80-T495X476M35ATE230
C10 1u C0805 CC0805ZKY5V9BB105
C11 0u1 C0603 710-885012206071
RSEN1 0R01 R0805 RCWH080510L0JQEA
R1 1k91 R0603 RK73H1JTTDD1911F
R_FB2 3k4 R0603 RK73H1JTTDD3401F
R_FB1 10k R0603 RK73H1JTTDD1002F
  • 0
    TI__Genius 11040 points
    Hello,

    Thank you for your description of your problem as well as sharing the BOM, layout and schematic. I will forward your questions to the product expert for his feedback.

    Best Regards,
    Katelyn
  • 0
    TI__Guru* 76852 points
    Hi Tadeusz,

    Please forward a completed LM5085 quickstart file for you design. I noted that the SBRD10200 10A/200V diode is really high voltage based on your Vin-max of 30V. You will achieve higher efficiency and lower temperature rise with a 40V diode.

    More important, the sat current rating of the Bourns SRR1210-150M inductor is only 5.3A (and this will be lower at hot). Please check with a higher sat current rating inductor.

    Regards,
    Tim
  • 0 in reply to Timothy Hegarty
    Intellectual 310 points

    Hi Tim,

    OK you are right that too low sat current of L1 could be the reason why output voltage and current drop after some time. But, why I have to use R1 as big as 6k2 instead of calculated 1k91? And what about ripple voltage?

    What do mean by "completed LM5085 quickstart file"?

    BR,
    Tadeusz

  • 0 in reply to Tadeusz Tatery
    TI__Guru* 76852 points

    Hi Tadeusz,

    If the inductor saturates, the peak current will be much higher than expected for a given DC output current, hence why you had to increase the ILIM resistor.

    The quickstart file is available by download from the product folder. I recommend filling it out for your design as it provides assistance with component selection.

    Regards,
    Tim

  • 0 in reply to Timothy Hegarty
    Intellectual 310 points
    Hi Tim,

    If "quickstart file" is also called "LM(2)5085 Quick Start Calculator" then it should be available here: www.ti.com/.../lm25085-5085quick-calc However, this site is almost empty for me - nothing to download.

    Regards
    Tadeusz
  • 0 in reply to Tadeusz Tatery
    TI__Guru* 76852 points

    8204.LM(2)5085_Quick_Start.xls

    Hi Tadeusz,

    That link works fine on my browser. I attached the quickstart calculator file.

    I suspect the inductor is the root cause of the issue - maybe you can test with an inductor that is adeqautely rated.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Intellectual 310 points

    Hi Tim,

    Today link works fine for me as well.

    Regarding "quickstart file", there is probably mistake in L1 calculation. This file use Ton_atVinmax for Lmin and Lnom calculations however equation 18 from LM5085 datasheet use Ton_atVinmin. This is probably reason why minimum recommended value for L1 is 15,6uH according to  "quickstart file" while calculation according to equation 18 from datasheet says 14uH (I used 15uH in my hardware).

    Next discrepancy is in R3 calculations. According to "quickstart file" recommended value for R3 is 93,1k but as "User selected value for R3" there is 19,1k. Why? My calculations according to equations from datasheet (26, 27, 28) gives me 119k.

    To be honest now I'm totally confused which calculations are correct. I attached excel file with my calculations so you can verify if they are correct.

    By the way, do you have any suggestion about proper value of inductor saturation current?

    Regards,

    Tadeusz

    LM5085_calc.zip

  • 0 in reply to Timothy Hegarty
    Intellectual 310 points

    Hi Tim,

    I've changed L1 to HM78D-128150MLFTR which has: inductance (parallel): 15uH, RMS current (parallel): 5.8A, saturation current (parallel) 9.1A, resistance (parallel): 0.036R and unfortunately this hasn't helped at all. With this new inductor almost all parts on the PCB are getting hot (inductor and Q1 are really hot). Previously only inductor was hot (but not so much hot as now).

    After power up output voltage and current drop two times (I've tried few times and only time after which voltage drops is different):

    Zoom to first drop (voltage dropped to about 4V8, current drops to 4,3A):

    Zoom to second drop (voltage dropped to about 4V6, current drops to 4,1A):

    DC signal is yellow, ripple output voltage is blue.

  • 0 in reply to Tadeusz Tatery
    TI__Guru* 76852 points

    Hi Tadeusz,

    I noticed that channel 2 with the AC coupled output voltage (showing the ripple) is very similar to a load transient response.

    Does the FB voltage (pin 3) remain at its regulation threshold of 1.25V when the output decreases? Please send the inductor current waveform too.

    Regards,
    Tim

  • 0 in reply to Timothy Hegarty
    Intellectual 310 points
    Hi Tim,

    It is strange but circuit has started working (L1 = HM78D-128150MLFTR). I have stable 5V with 4,4A on the output (parts are hot but it is possible to touch them for short time). However, R1 has to be 6k2 instead of calculated 1k91. Did you had a look at the attached calculations? Any conclusion why "quickstart file" is using different values for some parameters calculations than datasheet? And, which calculations are correct?

    Regards,
    Tadeusz
  • 0 in reply to Tadeusz Tatery
    TI__Guru* 76852 points
    Hi Tadeusz,

    I will review the quickstart file data. Please check the inductor current. This may indicate why R1 has to be so large.

    Regards,
    Tim