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.

TMCS1100: TMCS1100 gets faulty

Part Number: TMCS1100

Dear Ti team

kindly help me to resolve this issue 

we used tmcs1100a2 at smps output , it feedback is that it gets faulty afetr some time ,seen in many of the cards,

it starts to give fixed output voltage which corresponds to sometime 2A,to 16A i,e. 200mv to 1600mv, and why this happens is not known,

there is 10nf capacitor at output of tmcs1100 with 10K resistor in series and sometimes it behaves like it starts to give 1.6V fixed output while in there is no current flowing through it die in+ and in- pins,

and upon power reset, it starts to work normal, and some ic's output gets fixed to 1.6v or somewhere between 0 and 3v,

if you can , kindly help to resolve this issue 

  • lakh,

    Would it be possible to see a schematic of your setup?

  • Hello Carolus,

    below is schematic 

    Regards

  • lakh,

    A few things here. You mention in your first post that the output is connected in series with a 10k resistor. Is this 10k resistor swapped in for R23? R23 in the schematic provided is 0Ω. Also you mention 10nF of capacitance but the schematic is showing 100pF. O'm just trying to understand the signal chain differences between this schematic and whats been reported so far.

    Second, can you share the purpose of the ferrite bead in line with the ADC input here? I have not seen this done before, but just intuition tells me this may be an issue. The ferrite bead is going to filter high frequency content from the signal, but for an ADC input we need that high frequency content to properly settle to the correct value within the acquisition time of the ADC. In the course of your testing, have you examined the output of the TMCS1100 at the actual pins of the device vs. what is being delivered to the digital registers? It may be that the signal is being distorted by this addition. Could we get scope shots of the TMCS output at pin 7 of the device as well as the output on the "to ADC input" side of the bead?

  • Hello Carolous,

    In actual R23 = 0E resistance

    R31 = 10K resistance

     Also you mention 10nF of capacitance but the schematic is showing 100pF.

    C17 = 100pf and C64 is reduced to 10nf on board, because of the thought that hall may get loaded with this much capacitacne, but the problem still occured 

    yes FB9 and C64 to filter out high frequency noise, the traces from hall ic to microcontroller adc pins are more than 100mm, and i see some high frquency noise on adc pins, so put L and C to filter it out ,

    . In the course of your testing, have you examined the output of the TMCS1100 at the actual pins of the device vs. what is being delivered to the digital registers?

    - yes but with a multimeter i see 1.6V at hall ic output pin and adc shows counts accordingly as per 1.6V

     Could we get scope shots of the TMCS output at pin 7 of the device as well as the output on the "to ADC input" side of the bead?

    ok, i will put it on testin and wait for when problem occurs and capture these scope shots

     

  • Carolous, 

    what could be causing hall ic to damage,attaced image TMCS1100A2

  • lakh,

    This damage is most likely being caused by violations on the following curves:

    The ambient temperature must be kept to a certain level dependent on the level of current being passed through the device. To run the the device at 16A, specific attention must be given to the board layout as well as thermal calculations to ensure the ambient temperature does not exceed those listed in the above curves. We discuss thermal optimization in this app note

    This could provide a theory to potentially explain the behavior you are seeing. Upon startup, the device is in a working state, but beyond the defined ambient range of 125C, the part is not guaranteed to work in a linear fashion. That is obviously occurring here with the package being damaged to this effect. Then, it may just be luck that the part, after being allowed to cooled down, reenters a working state when the part is power cycled? 

    If you would like, you can PM me off the forum and I could have a look at your layout for you and recommend options for thermal management. 

  • can you recommend alternative part