My radio is ALMOST complete

Discussion in 'Wireless & RF Design' started by mik3ca, Feb 2, 2008.

  1. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    My circuit is shown as a GIF image (an attachment).

    My problem lies mainly in the components circled in red and purple.

    It seems that if I use capacitors for the red components and resistors for the purple components and if the capacitors are of the correct values and the resistors are the values that you see, then I will be able to pick up Z103.5 and 103.9 (the Hawk in WoodStock).

    the problem is that I don't know what values I should choose.

    I think that one resistor should be lower than the other resistor, but I am lost on the values.

    I understand that large capacitor values pass more base, and raise the volume, but I can't pick up remote stations. (but I can pick up 104.5 Chum FM)

    Can someone suggest values for the circled components and explain why those values should be used?

    I will think about adding a second diode in series with the one there when I see myself being able to pick up Z103.5.

    Thanks
     
  2. darenw5

    Active Member

    Feb 2, 2008
    45
    0
    I wouldn't add a 2nd diode there. A silicon transistor conducting current has a 0.7V voltage drop between its emitter and base. Much less, and the transistor is "off". The diode D1 already there, in the current flow from R5 through Q2 and R4, maintains a 0.7V drop between the bases of Q3 and Q4, not enough for both to be "on" at the same time, just one. Two diodes would let Q3 and Q4 both be on, and you may note that Q3 and Q4 are connected straight to the battery (or maybe through an on/off switch, which i'd recommend) - so i think you'd have a very short battery life.
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Where's your antenna? :confused:

    Good scoop on the diode - you would not want to put two in series.

    In the "good old days", we used Germanium diodes. 1N34, if I remember correctly (oops, wait - let me put my teeth back in....)
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Let me correct myself. :rolleyes:

    Take a look at D1 again. It's across the base of Q3 and Q4.

    There should be another diode in series with D1. Otherwise, the transistor's transitions will be very poor.

    Also, I see you trying to match a 2n2222 with a 2n2907. You should've used 2n3904/2n3906 instead; more amplification and a much better match on hFE.

    Working backwards, the rest of the circuit doesn't seem to make a lot of sense. ???

    There's no detector, no IF section... I don't get why you have C1 being 1000uF, since it's parallel to the battery - unless you're actually using a "wall wart" supply that's 100 feet away from the circuit. I'm not sure why your C5 is so large either. There is no load resistor in series with Q3/Q4, which means they could draw a LOT of current.

    Did you start out with an existing design and start making changes to it, or did you start from scratch?
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Hi Sgtwookie,
    It is a single transistor and single tuned circuit super-regen radio that has had many of its parts randomly changed around from the original version here:
    http://www.tricountyi.net/~randerse/superrgn.htm

    Its original amp has two diodes to reduce crossover distortion and has a lot of negative feedback to reduce overall distortion. Its output transistors operate with nearly 10 times the base current it has here.

    I simulated the original audio amplifier and this modified one and they are both biased incorrectly.
    Here is the original audio amp:
     
    • amp.gif
      amp.gif
      File size:
      10.1 KB
      Views:
      53
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Thanks, Audioguru - and yes, I agree that the changes appear to be quite random.

    mic3ka, I'm afraid that your circuit in it's current state will simply reduce fully-charged batteries to fully discharged batteries, albeit at a very slow rate.

    It appears that you used Cadsoft's Eagle Layout Editor to produce your schematic. While that software is good (yet quirky) for schematic capture and laying out PWB's, there is no way of simulating the circuit's performance without purchasing expensive add-ons.

    There are several decent SPICE programs available for free on the Web.
    One of them is LTSpice with SwitcherCad, by Linear Technology.
    The download page is here:
    http://www.linear.com/designtools/software/switchercad.jsp

    Analog Devices has a version of Multisim available for download here:
    http://www.analog.com/en/DCDesignToolsDisplay/0,3091,%5F1206,00.html
    It's quite a large program.

    Texas Instruments has Tina-TI, another schematic capture and SPICE program.
    http://focus.ti.com/docs/toolsw/folders/print/tina-ti.html

    There's also Circuitmaker Student - it's no longer supported, you can't update the library, and are limited to 50 components. It's handy for "quick and dirty" though.
    http://my.ece.ucsb.edu/bobsclass/2C/Simulation/circuit_maker.htm

    You really need to get some tools like these.
     
  7. darenw5

    Active Member

    Feb 2, 2008
    45
    0
    Yes, two diodes will more or less eliminate crossover distortion. But the biasing and battery drain, hmm... I too have simulated this circuit, using QUCS (http://qucs.sourceforge.net/) with the radio part and the audio part in separate projects. This is one of my favorite toys. (haven't touched spice in a long time) It's great to try good and bad circuit modification, and not a thing burns to a crisp. Don't know how well QUCS would hold up in professional work; spice certainly has the mindshare and model libraries and everything.
     
  8. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The original audio amplifier in the article produces a whopping 50mW into an 8 ohm speaker at clipping with a 9V supply. When it is biased correctly for symmetrical clipping its power output is 100mW.
    An LM386 little audio amp IC produces 450mW with much lower distortion, has automatic bias adjustment and costs less.
     
  9. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    sorry, I forgot to indicate it.

    I have it connected to the base of the RF transistor.

    I find that connecting it to the emitter (which I did before) will overpower the circuit (at best, changing the frequency).
     
  10. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    everything up to the first coupling capacitor (and inductor in series) is my detector.

    My receiver is a modified super-regenerative receiver.

    that's C4. and yes, I'm using a wall wart supply thats about 1 foot away from the circuit. I'm tempted to up that value to 4700uF. Something tells me that C4 serves as a filter, and I think that the higher I go, the better, since low frequencies coming from the wart will be removed. (let's see if I can get down to 1Hz. LOL)

    C5 is large because I wanted to create a short circuit for AC.
    Basically, I'm aiming for 0 ohms for AC at audio frequencies.

    I started out with an aircraft receiver design which can be found at:

    http://www.tricountyi.net/~randerse/SUPERRGN.GIF

    then I modified it.
     
Loading...