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.

When configured as an E-Fuse LM5069 fails after first trip.

Other Parts Discussed in Thread: LM5069, CSD18540Q5B, CSD17559Q5, LM5066

We are using two LM5069's on a power supply that we are developing; One for each rail of the supply (24VDC and 12VDC). When testing the trip setpoints of 6A and 7.5A, respectively, the circuit will trip once but fails to reset. After power cycling the supply power, the fuse conducts again, but as you increase the loading the circuit acts almost as if the MOSFET is operating in the linear region, and has a proportional voltage drop across the FET. The circuit never seems to recover from this state, so I'm assuming that it has failed.

Our test load is an adjustable DC load with very little capacitance and inductance. 

Thank you in advance for your help!

  • Paul,

    There are several problems with the circuit.  

    1) BSC052N03LS is a terrible FET to use for hot swap apps.  Its SOA plot is extremely poor and not suited to your application.  Guaranteed to fail with a power into load or power into short scenario.  TI has some excellent selections.   CSD17559Q5 will be a good choice for the +12v application.  For the +24v application, you need to use a higher Vds rated FET than 30v.  50v or higher is recommended.   CSD18540Q5B,  60v rated.  

    2) Pwr limit is set too high on both designs.  Reduce Rpwer (R125 and R230) to 61.9k and make the FET changes recommended.

    3) Can you provide Cout that this design feeds?  Assumed 220uF in attached.

    4) You will need a TVS on the input and a power Schottky on the output, appropriate for the voltage ranges you work to - if you intend to survive hot output short circuit.  During short circuit breaker action, di/dt will drive Vin very high and Vout below gnd, potenially damaging the LM5066 or FET.  The TVS protects the input.  5.0SMDJ15A is  sufficient  and an SMC size may also work but needs to be tested in your circuit.  A Diodes Inc B540 super barrier output diode will protect agains -Vout from exceeding -1v and damaging the out pin of the LM5066.

    5)  a .01uF or 0.1uF input capacitor, placed right at R201/R219 + side will filter against Vin noise.  Keep current sense routes short and keep the Vin route wider as it also provides bias current to the LM5069 and will improve current sense accuracy.

    Try these.  Play with the spreadsheet.  It is very useful.  Good luck.

    Brian

    LM5069_Des_Calcr E2E 5_27_16 12v_adjusted.xlsx5857.LM5069_Des_Calcr E2E 5_27_16 12v_adjusted.xlsxspreadsheets and this can be adjusted as needed.

  • Paul,

    Looks like I posted the same spreadsheet twice.  Here is the 24v sheet.

    BrianLM5069_Design_Calculator_REV_B E2E 5_27_16 24v.xlsx

  • LM5069_DesCalc E2E 5_27_16 24v_Adjusted.xlsxAnd the adjusted 24v sheet with new Rpwr and FET.

    Brian

  • Brian,

    Thank you so much for your exceptionally helpful response! We'll get to re-rolling the design and doing some more testing on very short order.

    -Paul.

  • Hi Paul,

    By the way, WELCOME to the E2E forums! Hope you have a great experience. I believe Brian covered everything quite well but if you're looking for additional resources to learn more about robust hot swap design, we have a lot of great material at "www.ti.com/hotswap".

    Under the Technical Documents tab, we have some recommended application notes. You can take a look at the recommend "Robust Hot Swap Design' article which is our "master" document for hot swap design.

    We also have these design calculator tools which Brian had attached. The latest versions for all of our devices can be found under the "Tools & Software" tab.

    Lastly we have a series of video tutorials to explain each step of the design calculator tools. They are broken down into 5-10min segments and can be found under the "Support & Training" tab.


    Overall, if you have any feedback regarding our collateral (what worked well or what we can do to make it easier for you), then be sure to let us know here on E2E!

    In case you were unaware, E2E also has blogs for discussions about power devices. For example some blogs I had written for hot swap design:
    e2e.ti.com/.../simplify-robust-hot-swap-design-using-design-calculator-tools
    e2e.ti.com/.../choosing-the-right-sense-resistor-layout


    Thanks for considering our products and we hope to hear from you again on E2E!
    Alex