Isolate ac input signal

Discussion in 'The Projects Forum' started by radhey, Feb 1, 2011.

  1. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    Hello guys.. I have to isolate ac signal(sinusoidal) within a range of 0-5 Volts before feeding the signal to a ADC, just for safety purposes, have to use optocoupler but i m not getting desired output. Please can somebody help me on this, i need the output to be equal or near to input signal. Can someone help me with the circuit please..
     
  2. blueroomelectronics

    AAC Fanatic!

    Jul 22, 2007
    1,758
    98
    Not with a typical optocoupler, the LED will need at least 1.2V to turn on and the devices aren't typically linear. They're designed to isolate digital signals.
     
  3. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    Ok thankss.. is there any other way of doing isolation, as in isolation of 0-5 volt ac signal..??
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    An audio transformer might do it.
     
  5. blueroomelectronics

    AAC Fanatic!

    Jul 22, 2007
    1,758
    98
    What frequency? What's it for?
     
  6. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    it is for low frequency...range of 100-200 Hz....
     
  7. blueroomelectronics

    AAC Fanatic!

    Jul 22, 2007
    1,758
    98
    Ok what's it for?
     
  8. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    actually i have to design hardware circuit for PC based oscilloscope...that oscilloscope will be used for automotive purposes so there generally we have low voltage signals ranging from +12 to _12 volts...i downscaled that signal to a value between 0-5 volt by using simple resistors.. but i have been asked to use isolation circuit for safety purposes. So before feeding it to the ADC i have to deploy isolation circuit...
     
  9. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    can you help me on it as in how should i proceed with this project...
     
  10. wayneh

    Expert

    Sep 9, 2010
    12,145
    3,055
    This is an ancient problem that applies to all oscilloscopes. "Active" probes are used to isolate the signal from the scope, but that doesn't solve the ground loop problem. You may want to read some background on oscilloscope methodology, and how probes are constructed.
     
  11. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    That seems like a remarkably small bandwidth, even for an automotive application. Are you sure that it will be sufficient?
     
  12. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    @wayneh is it required to go in for actual methodology followed by active probes of oscilloscopes..??..as in will it not be sufficient to use audio transformer for isolation as mentioned above...

    @Adjuster actually bandwidth is not specified xactly but yeah i have been asked to deign this oscilloscope for low freq signal, those in order of hundred hertz....
     
  13. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Assuming that's -12V, how did you do that with only a resistive divider? Also, the divider itself is going to dictate the impedance of the load at the 5V node. For the signal to remain exactly 0 to +5V the load would have to be infinite, but realistically will be determined by the values of the divider. The larger they are the more influence the load will have on that node.
     
    Last edited: Feb 3, 2011
  14. wayneh

    Expert

    Sep 9, 2010
    12,145
    3,055
    Yes, a transformer will solve some problems and may be a good choice. But it introduces other problems because you are no longer measuring the "real" parameter, but something related to it through the transformer. You won't be able to measure any DC, only AC. Maybe that's fine. I think results may vary with frequency, but perhaps that's no problem either with your narrow and low frequency. FWIW, I don't think the O'scope crowd uses transformers but of course they're always talking GHz.
     
  15. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    radhey, have you left this thread?
     
  16. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    thanx for suggestions guys...in my projct i used simple resistors and dc voltage to clamp and downscale the signal from -10 to +10 volt and then used opto isolator to isolate the signal...and the final voltage is comin between 0-5 volt ac and current level is in nano amperes...

    can somebody tell me how to give this signal to input of PC mic port...and can the mic port sustain this current and voltage level..??
     
  17. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    Like someone said before, opto isolator is very non-linear, unless you use some auto-linearizing scheme with feedback. They are made for digital signal.
    Could you post your actual schematic?
     
  18. radhey

    Thread Starter New Member

    Feb 1, 2011
    11
    0
    here is my circuit...
     
  19. wayneh

    Expert

    Sep 9, 2010
    12,145
    3,055
    First, is it really a MIC input, or is it a line-level input? A powered mic can be used with a line-level input, and true MIC inputs on computers are a bit rare these days.

    Inputs to the line level jack should be divided down to ~1V p-p. The input will have an impedance of at least 50KΩ, maybe much higher, so use a resistor in series with the signal to minimize current. Start with 1M, or even 10M, and work down.

    BTW, you should use good shielded cables for this, and put the dividing and impedance-matching resistors inside a shielded box. I use an Altoid tin with two RCA connectors (for two channels), and then standard A/V cables to the computer. The tin is handy because I can easily change the resistors around without changing the mechanical aspects of the setup.
     
  20. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Didn't you say that this signal was going to feed the input of an ADC and that's why you needed 0 to +5V? Did I miss when it was changed to a Mic input on a PC?
     
Loading...