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.

LM25088: Hardware-Fault and/or Design Fault?

Part Number: LM25088
Other Parts Discussed in Thread: LM5088

I created two versions of a large PCB, both with a[26V down to 15V]-12V DC-DC up to >5A output current based on the same LM25088 schematic but different layouts.
Both times I had issues with reliability and getting the DC-DC to work at all.
Both times I had issues with getting a reliable output voltage:

  • between ~7V to ~20V with 100Hz frequency with 13.4V RMS on V1 (Version 1) but >5A maximum output (not surprising that I am not able to filter that out with even a large C and L...)

  • constant 9V output only up to a <1.5A with intended osciallting output with 400kHz on V2, but of course wrong voltage (9V instead of 12V)
    or
  • also ~9V to ~17V with 100Hz frequency with 13.4V RMS on V2 similar to V1 behaviour but >5A maximum output


V2 of the PCB was made with a professionell Pick&Place-Machine, proper Solder Paste and industrial solder oven. Therefor I am assuming
there has to be some error in Schematic and/or Layout for both V1/V2, but I can't figure out what/were.
The design was taken out of the TI Webench Designer and used on another PCB, where it worked without any issues (a few years ago, constant exact 12V for up to 8A).
I hope to have only copied the schematic and made it a little more readable, but of course I have to have went in the wrong direction at some point.
On the V2 layout instead of the 10mOhm resistor on the bottom as in the layout picture,
the part was placed between the backside of the LM25088 and the Diode Kathode to make sure the trace resistances are low enough.
Pin 10 was pulled to ground with a 0 Ohm resistor to make sure there is no difference between different LM25088 versions.

The parts used:

  • All capacitors should be rated for over the maximum voltage (35V to be sure)
  • Diode: MBRD20100CT-13 (DPAK package)
  • Transistor: BUK7S1R0-40H (PLFPAK88 Package)
  • Inductor + Capacitor Output Filter: SRP1270-6R8M + EEEFT1V0101AP (100uF 35V)

If any oscilloscope measurements might be helpfull, I'll add them as requested.

I would appreciate any idea/any tipps where I can start "debugging" the hardware to test some part of the functionality and get towards solving the issue.
I would be glad if someone sees a Schematic/layout fault/issue, because then I will be able to fix it fast.
I have spent quite some time comparing working and non-working versions of the DC-DC on different PCBs, but could not draw any logic conclusions. Therefor starting over with fault-finding might be the best idea.

A little more info to:- constant 9V output only up to a <1.5A with intended osciallting output with 400kHz on V2, but of course wrong voltage (9V instead of 12V)
With a load that wants to draw up to 5A of current the start-up seems to work up to 1.5A, but then oscialltes between higher and lower voltage with a frequency of a few Hz.
Without any load, the voltage drops to oscillating 0V to 4V with a few Hz frequency. If anyone has had similiar issues, I would appreciate a short info about what/why.

Top View of Layout (All "Power" parts, also added 10mOhm on all gnd vias to the Diode Input):

Bottom View of Layout:

Schematic:

  • Hello David 

    I need more clear descriptions about the behavior to investigate the issue more. 

    "between ~7V to ~20V with 100Hz frequency with 13.4V RMS on V1 (Version 1) but >5A maximum output (not surprising that I am not able to filter that out with even a large C and L...)"

    1) Would you please describe this behavior and the test conditions more clearly. ? 


    "constant 9V output only up to a <1.5A with intended osciallting output with 400kHz on V2, but of course wrong voltage (9V instead of 12V)"

    2)   What's the FB pin voltage when VOUT is constant 9V ?

    3) What's your target switching frequency ? RT=33kOhm now. 

    4)Please double check your component selections using LM5088 quick start calculator https://www.ti.com/lit/zip/SNVU062 , and send me the excel file. 

    - Eric Lee

  • To maybe created a clearer picture I will only post both behaviours with the same test conditions of V2:

    Both times an inductive load 12V with continous current of 4.3A was used. I am noticing an additional frequency (to the switching frequency of ~400kHz?) of 1.3kHz.
    For the whole overview i measured each single Pin of the IC. The folder in each Zip contains oszi-pictures of all pins, because it is not clear to me which might be relevant. Id rather provide more data than necessarry than providing to little.


    V2, able to deliver full current of 5A of continous current but at oscillating voltage


    12V 4,266A Dauerlast.zip


    V2, not able to deliver 4.3A, also switching on and off with frequency of ~1.5Hz


    12V andere Platine Spannungszusammenbruch.zip


    About switching frequency and the LM5088 calculator:

    I dont really mind a different switching frequency/different characteristics because most of my supplied devices do not care about voltage deviations or a little bit of noise. But a voltage jumping from 9 to 13V might be a little bit suboptimal.

    I really only care about ~24V Volts in (26V max,, ~15V min.) to 12V with a close to constant voltage of ~12V. I would like that from 0A up to 8A, effective Current might be around 3A.

    I noticed in both cases the GND-voltage not being optimal, (maybe around +-200mV) which might be because i connected the cooling Ground-vias directly to the GND-pins via traces. I could remove these with a scalpel, but am not sure if this will help.

    I hope you can get more information out of all the Oszi-pictures than I can, I cannot find the cause for the unexpected behaviour.

    What parts (Capacities and Resistors) were actually used on both boards is hard for me to say, I could create another Board with the exactly rated parts, but that will take some more weeks, so I hope to find the problem before then.

  • Hello David 

    'Pin 10 RES.JPG' clearly shows that the buck converter is restarting at every about 600ms due to an over current limit.- Please refer Figure 7-8. Current Limit Restart Timing Diagram. 

    Please send me the completed Quick Start calculator with your electrical specifications. I'd like to check if compensation components are selected properly.

    Also, my recommendation is to use a better oscilloscope when you debug the board. The waveforms are not clear now.

    - Eric Lee

  • Hello
    I am closing this thread since I have not heard from you for sometime.
    If you are still trying to resolve the issue, please feel free to open a new thread or reopen this thread. Thanks
    - Eric Lee (Applications Engineering)