• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 2 answers
  • Subscribers 51 subscribers
  • Views 495 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.

THS4521: Low cost conversion of +/-1.25 bipolar signal to 0 to 2.5 V unipolar signal to pass through unipolar mux TMUX1308-Q1

Adam Torma
Mastermind 45340 points
Part Number: THS4521
Other Parts Discussed in Thread: TMUX1308-Q1, ADS131M04

E2E,

I want to design a circuit for my BMS system. There is a TMUX1308-Q1 MUX on the front end for a bipolar +/-1.25V signal.  However, the mux can't pass negative signals through the signal path. Therefore I need to level-shift my signal to address the issue with the mux. Did you know a cheap solution to do this? Or did you know somebody who can help me?

 

Thank you,

Adam

  • 0
    Guru 47515 points

    well certainly an FDA will level shift to an output common mode you set, but it will also be generating a differential signal, if that is not what you want, maybe just an op amp running inverting gain of 1 with your desired output common mode on the V+ input, but then we need the source impedance, and speed required. 

  • 0
    TI__Guru*** 140126 points

    Adam,

    What is the desired output range? Is inversion OK? If anything works, then this is lowest cost. Level shift with attenuation. 

    If you need gain of 1 or more , then an op amp circuit is needed. What is the desired signal bandwidth?

  • 0 in reply to Ron Michallick
    TI__Expert 3230 points

    Hello all,

    special thanks for your feedback. An operational Amp even for 5 Cent is much to expensive. All what would be possible is a transistor + some resistors.

    The best solution I've found is our ISO77xx series. On the input we use 3.3V power supply and on the output +/-2.5V. These solution works.

    But our ISO is in the range of 70 cent means 12 cent per channel this is far to expensive.

     Best regards

    Olaf

  • 0 in reply to Egon Hirsch
    TI__Guru*** 140126 points

    Olaf,

    Could you restate the input signal and the desired output signal again?

  • 0 in reply to Ron Michallick
    TI__Expert 3230 points

    Hello Ron,

    the power supply of the MUX is +/-2.5V and the power supply of the MCU is 3.3V. So the cheapest solution I found was our

    ISO77xx series. The input of our ISO is supplied with 3.3V and the output with +/-2.5V. This works fine, but the price per channel

    is in the range of 12 to 15 cents this is far too expensive.

    Best regards

    Olaf

  • 0 in reply to Egon Hirsch
    TI__Guru*** 140126 points

    Olaf,

    The power supply description is very helpful, but I still do not know data direction. My best guess it that data (analog voltage) flows into the MCU.

    With the MUX powered by +/-2.5V and MCU powered by 0V and 3.3V, I see that the control lines for the MUX will need to be level shifted for the MCU to control the MUX’s EN, A0, A1, A2 control lines. Is why the digital ISO77xx devices can do the job?

    However, there is also a level shift need for the analog channel MUX “D” pin the MCU analog pin. The ISOxx can’t do that.

    Is my understanding correct?

     

  • 0 in reply to Ron Michallick
    TI__Expert 3230 points

    Hi Ronald,

    the D pin from the MUX goes to the ADS131M04. This ADC has an internal -2.5V power supply so no level shift is necessary.

    As you wrote the MCU has to controll EN, A0,A1 and A2 this works perfect with our ISO only the part is too expensive.

    Best regards

    Olaf

  • 0 in reply to Egon Hirsch
    TI__Guru*** 140126 points

    Olaf,

    I’m sorry this took so long to resolve. It is much easier now that I know all the details of the problem.

    The MUX data sheet says for a supply of 5V (aka +/-2.5V) the VIL must be lower than 0.95V and the VIH must be greater than 1.25V. That is just a 0.3V dead zone to avoid. From the MCU ground point of view the VIL is less than -1.55V and VIH greater -1.25V.

    I have two solutions. One is passive (green box) and scales the MCU pin to have VIL and VIH near -1.82V (270mV margin) and -0.92V (330mV margin). This works if the voltage supplies are fairly accurate. I added two optional capacitors to speed up edges under parasitic capacitance load. The divider is fast, but I can’t say for sure if the speed of TMUX switch speed will be affected. You could open a new post for that so TMUX device support can answer.

    The other solution is active (PNP). This solution (blue box) inverts the signal. The VIH is a full +2.5V and will be fast. The VIL is a full -2.5V and the speed is based on R4 value and parasitic capacitance.

    easy2.TSC

  • 0 in reply to Ron Michallick
    TI__Expert 3230 points

    Hi Ron,

    this is what I looking for, excellent job. Special thanks for your efforts.

    Best regards

    Olaf