TTL voltage dropping from 5V to 0.8V when connected to circuit.

Discussion in 'The Projects Forum' started by Aditya Kumar Kasinadhuni, May 9, 2015.

  1. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
    Hello,

    I think this is a very simple question but I am not an electronic/electrical major so was wondering if someone could help me out.

    I am using a TTL (5V signal through a coaxial cable) from a function generator as a trigger to my microcontroller circuit (MSP430 Launchpad). In order to monitor the timing of the sequence, the TTL coax is split (using a coax T-piece) and with one connection going to the microcontroller while the other to an oscilloscope channel set to operate at 1 Mohm.

    In order to prevent the pins from being damaged, I am passing the 5V TTL signal going to the microcontroller through a voltage divider to ensure that its pins are seeing only 3V (their operating voltage). As long as the voltage is above 3V, the microcontroller detects the pin state to be HIGH and is triggered to perform other functions.

    This functioned well for a while, but suddenly from a couple of days, the TTL voltage that I am recording at the oscilloscope though starts with 5V, drops to about 0.8V after a couple of hours. Independently, the TTL when connected to the oscilloscope, is showing me a 5V signal. However, when I connect the remaining circuit at the T-piece, the voltage is immediately dropping to 0.8V on the scope and this is also not triggering the microcontroller (since it needs 3V). It makes me think that the circuit is drawing current but the microcontroller pins are configured to be INPUT which makes their input impedance extremely large (on the order of megaohms). So I dont quite understand the reason why the TTL voltage is dropping to 0.8V.

    I was wondering if someone could kindly throw some light on why the TTL voltage is decreasing? Your advise will be truly appreciated.

    Thank you.

    P.S: Please see the circuit diagram attached. allaboutcircuits.jpg
     
    Last edited: May 9, 2015
  2. crutschow

    Expert

    Mar 14, 2008
    12,993
    3,229
    I see no apparent reason for the voltage drop with time.
    Are you sure the micro input is in the high impedance state? What voltage do you measure directly at the micro input?
    You might try increasing the value of the micro input resistors to 20k and 30k.

    (Note that to show a shield cable connection you can draw a small circuit around the wire to indicate the shield and connect the ground to that circle. Typically you do that at both ends of the cable to show the shield connections. As you drew it, the ground is connected to the signal.)
     
  3. Brownout

    Well-Known Member

    Jan 10, 2012
    2,375
    998
    Don't use your scope to terminate the output signal (via the R-Test resistor)
     
  4. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
    Hi Brownout, I am sorry that part of the image was from an earlier version. Please see the new one.
     
  5. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
    Thank you. I have updated the attachment as you have suggested.
     
  6. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
     
  7. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    please check the polarity, ttl circuits are isolated internally by a reversed biased diode as the bottom layer. normaly it is ont conducting, but if connected backward will show around .8 volts.
     
  8. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,520
    1,247
    The first drawing shows the low side of the load going through a switch and 150 ohms to a signal called Cathode. The second drawing does not.

    ak
     
  9. Brownout

    Well-Known Member

    Jan 10, 2012
    2,375
    998
    Also, by connecting the scope's shield to the signal "cathode", you are connecting that signal to circuit ground, which probably isn't what you want to do. Use a multi-meter to measure the signal where is scope is presently connected.
     
  10. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
    Hi Brownout,

    The other alternative for me to measure the voltage across the resistor is to use an A-B approach. I created a math channel on the scope and connected channels A and B of the scope across the resistor. The grounds for these channels are then merged and connected to the microcontroller ground. Will this be okay? Or should the grounds be isolated from eachother?
     
  11. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
    It turns out the coaxial cable that connects from my function generator producing the TTL signal to the T-piece was bad. I checked the continuity of the center conductor of the coax with a multimeter and measured 1Mohm. Though that still doesnt not explain why I was seeing 5V or 4.3V when I connected that cable directly to the scope (without going through the T-piece). Can anyone please throw some light on this?
     
  12. Brownout

    Well-Known Member

    Jan 10, 2012
    2,375
    998
    It will probably be OK. I didn't suggest it because 1) I didn't know if you had another channel available and 2) I wans't sure if you wanted to keep the grounds isolated from each other.

    PS: This post references post #10.
     
  13. Brownout

    Well-Known Member

    Jan 10, 2012
    2,375
    998
    What do you expect to see when you connect the good coax directly to the scope? What are you seeing that you didn't expect?
     
  14. Aditya Kumar Kasinadhuni

    Thread Starter New Member

    May 9, 2015
    7
    0
    With the good cable, I am not seeing any voltage drop when I connect TTL from the function generator to the oscilloscope. I am recording the full TTL signal (4.85V in this case). Earlier with the bad cable, when I connected the TTL directly to the scope I was recording only 4.3 V. I wasnt sure that if the cable was bad, why I was actually recording only a 5V to 4.3V drop. I just wanted to be sure that there was nothing else that is happening here in the background.

    With the new working cable, connecting the circuit as shown in the figure, I am seeing 5V on the channel that is connected to the scope from the T-piece and my micro-controller is also being triggered correctly (meaning the input voltage at its pins is about 3V). Earlier with the bad cable I was only recording 0.8 V on the scope channel that connected to the T-piece.
     
  15. Brownout

    Well-Known Member

    Jan 10, 2012
    2,375
    998
    So, you're good, right?
     
Loading...