Connect with us

RF audio receiver diversity circuit

Discussion in 'Radio and Wireless' started by benbiles, Jan 26, 2014.

  1. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    Hi, I am trying to design a 'true diversity' RF receiver audio link.

    I have 2 RF modules that output line level audio.

    The modules have a Vdc RSSI indicator output. 0 - 2v

    I would like to Pan the audio from one module to the other by using a comparator to decide the higher RSSI signal ( with some hysteresis (( adjustable 5ms - 500ms? )
    and have the audio output panned rather than switched from the audio source A&B in approx 5ms - 500ms adjustable by variable resistor(s).

    Here's where i am at so far !!

    1, I have a rough idea that the +-1V audio outputs have to be converted to 0v -1.5v swing with OpAmps so it can pass through the pan circuit?

    2, I should use CMOS pair (nfet and pfet) with some RC to control speed of the pan from one audio source to the other and back ?

    here is a circuit diagram I have slashed together, but I think its still way off from working !!!

    any help / ideas would be really welcome !!!

    thanks,

    Ben
     

    Attached Files:

    Last edited: Jan 26, 2014
  2. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    Hi Ben
    Can you explain in detail what the circuit is suppose to do. It's just I am not sure of the purpose of the caps on the input to the FETs it appears you are feeding them with a low or a high DC level. What's D1 doing across the outputs?

    Your circuit says +/- 1V, you won't get this from a single supply unless you go through a capacitor from the output of the op-amps. And you will probably want to set the output at half supply so you will need supply the non inverting terminal with 1/2 supply voltage from say a buffer. Then you would normally couple this in through a capacitor to the input to block any DC. Put a series resistor after the 1/2 supply feed and the then input goes after this.This helps with PSR. I personally find it much easier to use a split supply, which is where most op-amps seem to work the best.
    Look Interesting though.
    Adam
     
  3. davenn

    davenn Moderator

    13,345
    1,774
    Sep 5, 2009
    agreed, the RSSI is a DC voltage and it isn't going to be able to get to the FET's to control them via caps


    It appears as tho he was using it to stop one output from feeding back into the other output
    problem is that a diode in an audio line like that is likely to cause distortion of the signal

    Dave
     
  4. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    caps and diode removed!!

    Hi Arouse1973!!! and Davenn !! Thanks so much for looking at this for me !!

    woops yes the diode should not have been there, It was an earlier mistake I made.

    The caps where there simply because I nicked the soft switch idea from a musicians Boss anti click pedal !!! so yes, there gone too now..

    I'll try and explain what this circuit is supposed to do....

    The two audio RF receiver modules ( radio mic receiver ) receive/demodulate the same radio signal through there individual antennas. In this case it could be WBFM or 2.4 ghz packet radio depending on the modules.

    The audio output of the different modules is always standard (ish) line level audio -10bd ( +-1v ) approx!!

    The idea is to decrease the chance of audio dropout by using 2 receivers & 2 antennas as redundancy. The audio output of my design should always come from the receiver with the highest signal strength (RSSI )

    So the comparator output should go high if audio module A's RSSI Vdc is higher than audio module B by a percentage. comparator output should go low if module B's RSSI Vdc is higher by a percentage.

    There should be some hysteresis added so that the inputs don't switch erratically when both modules receive strength ( RSSI vdc ) is too close

    The problem with switching the from one audio source to another with say the an ADG884 IC would be a clicking popping sound as the audio switches over.

    So I'm trying to switch the audio source by quickly panning over with the NFET PFET's and some RC controlled delay..

    Arouse1973, I tried to add the half power buffer to the op amps in this drawing.. not sure
    what you mean by buffer? should i use a transistor so half the input voltage? or am i close with the resistors?

    What kind of capacitor value should I add on the outputs of the opamps?

    Would this kind of audio (AC?) get through the cmos NFET PFET with no added DC ?

    Do I have to remove the DV voltage after the audio is passed though?

    would C5 & R7 + C6 & R8 control the switch speed of NFET & PFET ? or is that totally wrong?

    once again thanks so much for your help on this !! I would be great to get to build this and show you guys it working :) :eek:

    Thanks BEN
     

    Attached Files:

  5. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    When ever you express in terms of decibels it is useful to give the unit the ratio refers to. In this case it could be dBmV or dBV. I assume dBV because you give +/- 1V so this means the output level will be 10dBV smaller than 1V. That works out to be 0.31V, so also wondering how your going to be -0dBV (1V) lower than this. Or am I not understanding what you mean?


    Resistors are commonly used but they do reduce the PSR. It's a compromise for a less complicated circuit. Using split supply is much better as any small variation in the supply voltage does not effect the output. Google op-amp buffer and you will see what I mean.

    Try 10uF-100uF depending on the lower frequency response you want.

    Give me a sometime and I'll try and work something out for you. Can you give me the part number of the RF modules you are using.
     
  6. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    audio output power

    ahh, maybe i need to re-think this! Maybe i should normalize the audio inputs somehow since different boards will output different audio levels...?

    do these numbers make more sense?
    30mW RL = 32Ω or 50mW RL = 16Ω
    1k Ohm Output impedance

    anyway, I'll Google op-amp buffer and try add that to the drawing tomorrow if I can understand how it works.

    maybe I can PM the part numbers to you later? there making up some adapter boards so there not for sale as such yet i don't think..

    basically some i will test first draw up to 35mA at 3v each so hopefully I can run them on 2xAA batteries. I don't know yet what power regulator the modules use..:confused:
     
  7. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    Hi Ben
    This is something like I was on about. You could try just joining the outputs together as shown with out the switching circuit. As long as their is no more than say 5 degrees phase shift between the two outputs it should work. You would also benefit from an AGC circuit. But I won't be able to work on that until a bit later. That needs a bit more thought.
    Adam
     

    Attached Files:

  8. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    VCA's instead?

    I'm starting to think this might be a better off using VCA's so I can PAN the output from source A - B - A and so on

    If the comparator switches the transistors that increase voltage to the VCA's with RC time delay then maybe I would get a PAN effect as the VCA's decrease or increase Vcc input.

    Maybe this would work? that's if the VCA's let no audio though when Vcc = 0v ?

    I'm not really sure how to set the RC time delays so that the volume increase and decrease over time properly.. but that would prevent any audio clicks if it worked..

    There's also a very low power / low noise dual channel VCA IC if found quite cheap.. so I could get away with using only 2 VCA's if it works!

    Can anyone see any reason why VCA's are the wrong way to go about this ?

    I don't really care much about the output level of the audio as long as there's no distortion!
    :rolleyes::confused::p
     

    Attached Files:

  9. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    Ben

    Put this together for you, it should work ok if you keep the input between 200-400mV. Set the input for 300mV and the output should be 4V pk pk. So if your input goes up and down a bit it should be fairly constant. You should only see a variation of about 20mV on the output which has been divided down from the 4V pk pk to 800mV pk pk by the resistor dividers.

    You may have to fiddle with the values a bit to get it to work for your design.

    Thanks
    Adam
     

    Attached Files:

    • AGC.PNG
      AGC.PNG
      File size:
      10.9 KB
      Views:
      299
  10. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    buffer ?

    Hi Adam, so is this circuit a buffer? copy input audio and output at 800mv swing to be hooked up to switch or pan section afterwards?

    Am I correct in thinking this takes the load of the modules audio output?
    just trying to work out why I need this buffer.. will it normalize the audio input from v2 between between 200 - 400mw to an output of 800mv ( within 20mv ) so its an AGC buffer ?

    I'm trying to understand how J1 is working.. effecting the hysteresis of the opamp? is this controlling the gain amount? a kind of feedback?

    could I scale all the values down to make it all run on 3volts ? or does the input voltage need to be that high for gain control?

    sorry, I am a real beginner as you can see!! just trying to learn how this is working before i move onto the panning stuff..

    I got the idea on another forum I should switch audio from A-B or B-A in circumstances where RSSI vdc changes quickly and pan otherwise to avoid sudden dropouts. I was thinking I could control the rate of pan somehow with a 555 timer and the VCA's but no idea how to detect sudden drop in RSSI !! anyway still thinking about that one!!

    thanks again for your help, I'll keep looking at your buffer and try figure out how its working !!
     
  11. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    buffer ?

    Hi Adam, so is this circuit a buffer? copy input audio and output at 800mv swing to be hooked up to switch or pan section afterwards?

    Am I correct in thinking this takes the load of the modules audio output?
    just trying to work out why I need this buffer.. will it normalize the audio input from v2 between between 200 - 400mw to an output of 800mv ( within 20mv ) so its an AGC buffer ?

    I'm trying to understand how J1 is working.. effecting the hysteresis of the opamp? is this controlling the gain amount? a kind of feedback?

    could I scale all the values down to make it all run on 3volts ? or does the input voltage need to be that high for gain control?

    sorry, I am a real beginner as you can see!! just trying to learn how this is working before i move onto the panning stuff..

    I got the idea on another forum I should switch audio from A-B or B-A in circumstances where RSSI vdc changes quickly and pan otherwise to avoid sudden dropouts. I was thinking I could control the rate of pan somehow with a 555 timer and the VCA's but no idea how to detect sudden drop in RSSI !! anyway still thinking about that one!!

    thanks again for your help, I'll keep looking at your buffer and try figure out how its working !!

    I found out that the TI VCA810 VCA's won't cut it for Audio !! I think too much THD at the top end or something.

    Anyway, THATS 2162 VCA's run at minimum +-2.5v ( 5v ) non operational at 4.5v
    so that's the 3v idea out the window ! Looks like 9v is way forwards and down-convert for the 3v module.

    So I need a Vdc ramp up and 'hold' when the comparator goes from low to high and ramp down and hold when comparator goes high to low.. to be sent to VCA2 gain control.

    and then I would need to reverse that Vdc ramp effect to control VCA2 gain with a inverting transistor?

    anyone know a good place to start ! I'm sure this ones been done many times before..

    I guess a 555 timer and RC but I'm not sure how to hold the Vdc output indefinitely until next input change!
     
    Last edited: Jan 31, 2014
  12. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    heres a drawing with AGC's and dual chanel VCA's

    here's a drawing with AGC's and dual chanel VCA's

    Thanks Adam for the AGC !! I couldn't fit in your schematic on one page x 4 so I just draw in blocks for now.

    The LT1007 looks like an amazing high quality opamp for audio re THD !!!

    so I need to sort out ramping up and down the VCA's gain next for panning the audio.

    The 2 x THAT2162 VCA's will hopefully do the job with just +9v without the negative voltage.
     

    Attached Files:

  13. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    Hi Ben

    You won't necessarily need the buffer, if you do then you remove the potential divider to the input of the buffer and the output of the buffer goes where the divider was. Use one buffer for both channels.

    The AGC will be your output and the two input channels will go into the first two op-amp circuits I sent you. If you join the two outputs with resistors they will mix together so you shouldn't need any switches. If one of them is slightly lower that the other the higher one will dominate and the output into the AGC should stay within limits. You need to adjust the gain of the first two op-amps so you have 300mV in to the AGC.

    J1 is used as a variable resistor and is operated in it's ohmic region. As the input changes the voltage on the rectifier output varies, this in turn varies the gain.

    As I said you might need to tweak it a bit. It might run on 3V but this might be a bit low.

    Adam
     
  14. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    audio cannot be mixed

    Hi Adam,

    Ok, great, I see where your coming from re the buffer / AGC

    I cant mix the audio channels together though since the whole point if the circuit is to switch or pan between RF receivers audio outputs to avoid RF dropout or RF fade / nasty noises ! Its a common trick on radio mic receivers to decrease the chance of the audio taking a hit.

    Unless you mean that the AGC of each channel can be controlled by the RSSI input and can actually shut off audio coming from one source completely letting only the stronger one through?

    If you look at my last drawing I think you will see what I mean. The reason you cant switch the audio is because of spikes in audio. that's why I'm going for a fast audio pan.

    some receiver systems switch incoming RF signals from two RF receivers before its demodulated & expanded into audio but I'll be using different modules so I cant really hack into them that easily..

    I'm thinking about using the IC for the pan circuits, do you think I would be better of using the opamps you mention in AGC and making the RSSI Voltage controlled gain that way ?

    http://www.thatcorp.com/2162_Dual_Pre-Trimmed_Blackmer_IC_Voltage-Controlled_Amplifier.shtml
     
  15. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    Ok connect the RSSI output to the front of the op-amp you want to turn off. Is this an option?
     
  16. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    T2 is not quite right. You need to use a PNP. Connect it like T1 but with the emitter to the power and use a base resistor.
     
  17. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    is that correct as in drawing ? I connected to base above resistor now its PNP

    anyway need outputs from tr1 and tr2 to mirror

    ramp up and hold output at +9V in 50ms
    ramp down and hold output at 0V in 50ms

    that way two VCA's gains go up while two fall creating the stereo pan

    unless there's a simpler way to do it ?
     

    Attached Files:

  18. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    Nope flip it over. For someone that just slapped the first design together, you are making a few fubars. Lol
    Adam
     
  19. benbiles

    benbiles

    12
    0
    Jan 26, 2014
    pnp npn

    LOL!!! I got worse at this than when I was a kid in the 80's !! At least then I new the difference between my pnp's and npn's !! although I did blow a few up :)

    anyway npn or pnp inverter err same thing?

    Yer, so about ramping these voltages.. I could use just simple RC and charge them up with transistor output?

    or should i use a 555 timer?

    I don't really need to ramp up/down the current in a perfect line but would a cap+resistor in parallel discharge current in the same way they charge?

    I'm using TINA- TI free version and these odd dots are appearing ( in yellow ) !! guess you get what you pay for !!
     

    Attached Files:

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Similar Threads
There are no similar threads yet.
Loading...
Electronics Point Logo
Continue to site
Quote of the day

-