• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 48 subscribers
  • Views 1541 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.

LMP7300 hysteresis problem

user4411225
Prodigy 90 points
Other Parts Discussed in Thread: LMP7300

Hi ,

    I encountered a problem when I used the LMP7300. As illustrated in the following Fig, LMP7300 monitors the bus voltage. I set the Positive Hysteresis is zero, while the Negative Hysteresis is Vref*R11/(R11+R17) = 122mV, which is less than max hysteresis 130mV. However, when I turn up the Vin makes the voltage of input Pin -IN is larger than Vref, the output is still in high voltage. It can not work. When I tied the Pin HYSTN to Vref, It can work well. So I'm doubtful that the key of problem is the hsysteresis setting.

    Pls help me find the problem. Thx!

    ps:sorry for my poor Engilsh.

 

  • 0
    TI__Mastermind 49966 points

    Hello User,

    Looks like you swapped the inputs compared to the example circuits (which normally have the reference voltage on -IN). This would change the definition of positive and negative hysteresis directions as the comparator sees it. Have you tried swapping HYSTP and HYSTN inputs?

    Can you provide a scope photo of the test points?

    BTW: Never apologize for learning a second language...your English is fine.

    Regards,

  • 0 in reply to Paul Grohe
    Prodigy 90 points

    Dear Paul,

            Thanks for your quickly reply. It's indeed that I try to swap the inputs for getting a inverting output.But I can show you how I caculated.

    In comparator side:

             ViL=Vref-Vref*(R11/(R11+R17))=1.926V

             ViH=Vref=2.048V

    In  bus side:

             ViL_bus=ViL*((25+1)k/1k) =26*ViL=50.073V

             ViH_bus=26*ViL=53.248V

           Look at my simulation. Red is Vbus, green is output.  What I want that when Vbus is higher than 53V, the comparator output switches from high to low, after that,  the Vbus drops from 53V to 50V, the comparator output switches from low to high.

             

            I have no idea about AC test lacking of programmable power supply. So it is meaningless to show you the test scope.  The test result is that when Vbus is larger than 53V even 55V, the comparator output keeps high. No matter what I adjust Vbus input, the output has no reponse.

            Thanks again.

    zhipan

  • 0 in reply to user4411225
    TI__Mastermind 49966 points
    Hello Zhipan,

    How high do you have to go with Vin until it switches? Does it ever switch?

    If you cannot generate >55V, then apply a 1.8- 2.2V DC voltage directly to the negative input (pin 2).

    Have you confirmed that the reference voltage is correct (2.048V)? Can you provide the voltages on pins (particularly 1, 2, 4 & 5) under both high and low Vin conditions?

    Regards,
  • 0 in reply to Paul Grohe
    Prodigy 90 points

    Hi Paul,

         When Vin is 53V, the output will switch. But it did not.

         According to your advice, I have a test with 0~2.2V DC voltage directly to the negative input. As illustrated in following figure, Ch1 is -IN(Pin 2),Ch2 is OUT(Pin 4), Ch3 is HYSTN(Pin5),Ch4 is REF and +IN(Pin 7 and 1). You can see negative input is a triangle wave whose high-level is 2.2V and low-level is 0V. Out is still 5V. HYSTN is 1.96V which is divided by the resistors. REF and +IN are 2.048V which is right.

        Is it right to swapped the inputs ? The internal block diagram of comparator is not shown in datasheet. I do not know wheather the feedback of amplifier is changed. 

          

         Thanks a lot. 

    Best Regard.

    Zhipan

  • 0 in reply to Paul Grohe
    Prodigy 90 points

    Hi Paul,

       I find when I switch +IN and -IN, the HYSTP and HYSTN are also switched. In my case, I want to set the negative hysteresis, but test result shows it is a positive hysteresis. I can recalculate the resistors to repair circuit. Because this conclusion is not in datasheet, can you help me check it ? Thanks.

      

  • 0 in reply to user4411225
    TI__Mastermind 49966 points

    Hello Zhipan,

    Sorry for the delay. I have been stuck in meetings.

    What is your new resistor values and/or schematic? Then I can verify it.

    Regards,

  • 0 in reply to Paul Grohe
    Prodigy 90 points

    Hi Paul,

        As illustrated in following figure, the bus sample resistors are replaced.

  • 0 in reply to user4411225
    TI__Mastermind 49966 points

    Hello Zhipan,

    Your original circuit should work.

    By inverting the inputs, the definition of "positive" and "negative" reversed as to the input signal. You are actually adding a negative offset to make the positive Vin threshold go higher (because the comparator inputs and logic see it as a negative slope).

    I was finally able to go out into the lab and try it out, using precision voltage sources and 6 digit DMMs (to eliminate resistor tolerance issues).

    I set up the circuit as follows:

    V+ = +5V

    +IN to VREF

    -IN to Voltage source with DMM

    HYSTP = VREF

    HYSTP = Voltage source with DMM, set to +1.92300V

    VREF measured 2.04823V

    The positive trip point (~53V) was correct at 2.048V.

    However, the negative trip point (50V) was at 1.946V, a +23mV difference. That is much higher than the spec'd 1mV offset. Even with that, that translates to a 50.6V trip point. I'll have to figure out what is going on...

    Can you manually sweep the input voltage and accurately measure (with a good DMM) the voltages on the pins?

    The DC accuracy and offset of Tek digital scopes can be off by many percent if you do not do the self-calibration routine every day, and gets worse using x10 probes. Usually the cal is found in the Utility menu called "Signal Path Calibration". It takes several minutes to run. Not to be confused with probe calibration!. The cal should be done before attempting any accurate DC measurements. I would highly recommend a good (>4.5 digit) DMM for static measurements.

    Regards,

  • 0 in reply to Paul Grohe
    Prodigy 90 points

    Hello Paul,

            You are so kindly help me verify the circuit. Thank you very much.

            I will verify the recaculated circuit with new resistor value next week.  I will use a good DMM an OSC to measure it.

            During in my design period, I simulated the inverting circuit. However the pspice model of LM7300 showed the same feature with noninverting circuit. So I suggest the model should be fixed and the inverting input should be added in your datasheet.

            Thanks again!

    Best Regards!

    Zhipan