• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 45 subscribers
  • Views 268 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADC interfacing

Deepak%20murali
Expert 1095 points
Other Parts Discussed in Thread: MSP430G2553

6087.MSP_problem.TSCHi,

  I have designed a low pass filter(of cut off frequency 20 Hz). I gave a sinusoidal signal of frequency 12.5 Hz as input to the filter.  I connected the pin 6 of the filtering OPAMP to the CRO and I was satisfied with the output that I got on the CRO screen.

But when I connect the pin 6 of the OPAMP to the input of inbuilt ADC in MSP430G2553(to pin-P1.1), the output on the CRO gets disappeared.When I disconnect the pin 6 of the OPAMP from ADC input, the output waveform re-appears on the CRO screen. Any impedance mismatch problem?

From the datasheet of MSP430 ,the input capacitance is 27pF and a series resistance of 1000KOhm is present. I think I am correct about this information.

I have attached the TINA file of my filter along with the input impedance model of the ADC(using a 27pF capacitor and 1KOhm in series). I didn't find anything wrong with TINA simulation, but then Why am I not getting the output?

Thank you in advance.

  • 0
    TI__Guru 61456 points

    Hi Murali,

    Most likely it is a high frequency oscillation due to the 27pF (or more) load. Op-Amps do not like capacitive loads directly on their output. Your circuit has a gain of less than one, so it is more susceptible to capacitive loading. There needs to be a small value resistor in series between the op-amp output and ADC input.

    Your model of the input is a little incorrect. There will be 27pF to ground on the input with a high input impedance. This may be enough to cause the output to oscillate, and this oscillation can be in the 100's of KHz to MHz - depending on the op-amp bandwidth. Depending on the capacitive load and output drive capabilities, this oscillation can be hundreds of mVpp to full-rail swings.

    You are probably using a DSO instead of a CRO...and DSO's can lie. Digital scopes alias on high frequencies and/or average it out. CRO's will show "fuzz".

    DSO scope default settings are intended to make the trace look pretty and clean - which is the opposite of what is needed to look for oscillations and noise. Set thevertical scale to 500mV/Div, Turn off averaging, turn on "real-time" mode, set the channel BW to the highest, set triggering to AC coupled and 0mV level, and set samples to peak detect. Start with the highest timebase (100ns) and go lower to find the oscillation.

    See me...you are a few cubes away...show me your setup...

    Regards,

  • 0 in reply to Paul Grohe
    Expert 1095 points

    Hi Paul Grohe,

       Thank you so much for your reply. But I am confused with your advice. In this statement "Your model of the input is a little incorrect. There will be 27pF to ground on the input with a high input impedance.", you are telling me that the 27pF capacitor is in series with a high impedance resistor(as for voltage input, input impedance must be very high). But by this statement "There needs to be a small value resistor in series between the op-amp output and ADC input.", you are asking me to put a small resistor between the OPAMP and the ADC deliberately. Will it be of any use in placing a small resistor in series with high impedance of ADC?

    And you are 100% correcting(amazing, were you present when I was using DSO?) about my DSO. I used DSO(Falcon) only. I didn't know that there will be difference. I just use to think that both CRO and DSO are for checking our output waveform and nothing more.

    From where can I get more info on what exactly I will be doing in these steps and why I have to do these things."Turn off averaging, turn on "real-time" mode" & " set samples to peak detect". I mean I wanted to know more about how to use DSO in a better way instead of just operating time scale and volt/div.