filtering?

Discussion in 'Wireless & RF Design' started by mik3ca, Jan 23, 2008.

  1. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    I have made a radio that can pick up remote stations when I use a battery (that has 6V left in it). If I change the battery to a DC adapter rated at 6V, 300mA, then all I get is a 60Hz hum mixed in with the radio. How do I remove it?

    My design is similar to: http://www.tricountyi.net/~randerse/superrgn.htm but I swapped the emitter and collector of the transistor, and connected the inductor to ground, and connected the 0.01uF and 10K in parallel to +ve instead of ground, and omitted the .01uF capacitor between the second NPN and ground. I also converted this stage into a push-pull amplifier stage to gain more volume.

    I think that I need some sort of filter so that I can convert adapter DC voltage to battery DC voltage.

    Can someone help me?

    thanks.
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Sounds like you're using a "wall wart" 6v supply. Some of these have built-in regulation, some don't.

    You can probably get rid of most of the hum by using a relatively large electrolytic capacitor (say, 100uF or larger) connected from your Vcc to ground. Make sure you observe correct polarity on the cap. The cap should have a rating of 12v or better (rule of thumb for electrolytic cap voltage rating is 2x the voltage to be applied across it.)

    If the hum doesn't go away, your "wall wart" may have a bad rectifier in it, or your circuit may be drawing near the maximum current that it can supply.
     
  3. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    I already have a 1000uF, 16V rated capacitor connected between +ve and ground.

    my power amplifier is similar to this circuit:

    http://web.telia.com/~u85920178/begin/bias-10.gif

    the exceptions are:

    I used a short instead of a second diode.
    C1 is 47uF
    R1 is 1.2M ohm
    C3 is 100uF
    R2 is 8K
    TR1 and TR2 are 2N2222's
    TR3 is a 2N4403

    this whole amplifier is connected to the last circuit I showed you at the point where the coupling capacitor comes in.
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You should not have replaced the 2nd diode with a short; the D1/D2 pair adjust the biasing across the transistors, keeping a constant voltage drop of around 1.2-1.3v. With just one diode, the voltage between the two bases is between 0.6-0.7V. Your amp is likely drawing far more power than you intended, and certainly not functioning as designed.

    Look at the circuit again. Notice that TR2 and TR3 amount to two PN junctions from base to base? That's why there are two diodes; the same voltage drop as the PN junctions in the transistors.
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Humor me. ;)

    Try putting a diode between the + lead of the wall wart and your filter capacitor.

    If the noise goes away, you have a shorted rectifier in the wall wart.
     
  6. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Hi MStechCA,
    I know it is you. I don't know anybody else who is playing with a lousy super-regen "radio" and this "amplifier" that is full of distortion.

    I simulated your amplifier and it was horrible. The one diode causes crossover distortion (not high current like SgtWookie says).

    I added a 2nd diode to reduce the crossover distortion, added a resistor and capacitor to "boostrap" the driver transistor, added an input resistor, re-biased the input voltage and used a 2N4401 to match the 2N4403 output transistor. It is much better.
     
  7. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
  8. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    lousy? LOL

    even if it is 1N4007 diode?

    Must the capacitance between the anode of the first diode and the cathode of the last diode (from the diodes in series) be as small as possible? is that why they use 2 diodes?

    I know this rounds the top of the waveform, but at this point, I don't care, because the only thing I am interested in now is picking up a remote station. The quality of the sound doesn't concern me yet.

    My input resistor in my design (which I didnt mention before) is a 0.15uH inductor because I want to give out maximum output while cutting off RHF. (RHF = ridiculously high frequencies).

    Why do you use only 330K? I had to use higher resistor values in order to obtain more gain.

    Why does a 2N4401 make it better?

    also, why are your two resistors coming from VCC 820 and 920 ohms? aren't those too low of values?
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    A 1N4007 has roughly the same forward voltage drop (0.6V - 0.7V) as any other PN junction silicon diode, such as an 1N914, 1N4148, etc. It's mostly their current rating, PIV, and physical size that make a difference when choosing one for an application.

    When you're choosing transistors to use in a push-pull amplifier like this, you should try to select a pair that are well matched for gain (hFE) as mismatches will lead to considerable distortion. 2N2222s have an hFE of around 200. A 2N3905 is just slightly higher, but would be a reasonable match. The 2N4403 you originally selected has an hFE of around 335, which is the big reason why the waveform was so distorted. The 2N4401 at 300 is a much better match.

    Transistor hFE's are specified as a minimum, and typical value. In reality, the transistors themselves can vary fairly widely. You can use a multimeter that has an hFE function to manually match your transistors for the best circuit performance. Harbor Freight has a couple of models of DMMs with hFE that are occasionally on sale for as low as $2.99

    But let's get back to your 60 cycle noise problem.

    Did you try using a diode between your positive wall wart lead and your 1000 uF cap, cathode to the cap?
     
  10. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    I actually solved the problem. I cut the size of the antenna and disconnected it from my furnace vent (equivalent to a cold-water pipe). It is the furnace vent that is causing much of the noise.

    I actually didn't try using the diode, but I will.
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Oh! I had no clue you'd created a ground loop.

    When you were using the battery, your circuit was "grounded" via the antenna.

    When you started using the wall wart, you probably created a ground loop.

    If the noise is gone, the diode won't improve things, but it will reduce the available supply voltage from 6 to around 5.3 to 5.4v. At that point, the only benefit would be to prevent damage in case the supply was connected with the leads reversed.

    Your antenna will be most effective if it is 1/4 wavelength.
    For the following frequencies:
    118 MHz, or the middle of the band, that's about 28.4".
    108MHz is 31",
    128MHz is 26.16".
    If you're listening to one particular frequency most often, you should make an antenna tuned for that frequency. The "general chat" pilot frequency is 123.45 MHz, optimal 1/4 wavelength for that is 27.12".
     
  12. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Hi again mik3ca,
    I am sorry if I offended you. We have talked about your super-regen radio's reception problems and audio amplifier's distortion enough on the other forums. I was hoping that you would understand my simulation of your amplifier and my fixes to it here. I won't talk about them any more.
    I am glad you fixed your hum problem.
     
  13. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    You might try using a grounding rod, if you have access to such. Cold water pipes are no longer approved as the sole grounding means for residences because they don't always provide a decent ground path. The same might be happening with your furnace vent.
     
  14. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    I just noticed that when I replace the two pull-up resistors in my schematic with a pull-up resistor and a 0.15uH inductor connected in series, it seems that I can pull in weaker stations better.

    I'm wondering if the wall adapter emits substantially high frequencies that prevents the receiver from working.
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    In HF circuits, very small changes can result in either improved performance, or greatly degraded performance. It's up to the person with the circuit in hand to figure out what works best.

    ALL circuits are compromises of sorts, trading current for speed, voltage for amplification, etc. But when beginning to delve into higher frequencies, un-planned for things can happen. Circuit layouts must be more and more precise, types and composition of things like capacitors must be considered. A multilayered ceramic capacitor does not respond at 500 MHz like it does at 50 MHz; one marked as 47pF will likely measure close to 147pF at higher frequencies, and just go up from there.
     
  16. mik3ca

    Thread Starter Active Member

    Feb 11, 2007
    189
    0
    I have a miller effect circuit in my amplifier. the emitter of a transistor is connected to ground through a capacitor and resistor in parallel.

    I tried 8K resistor in parallel with a 1000uF cap and it seemed to cause the hum. I reduced the 8K to 1K, and the hum was reduced, but I think that my ability to tune into weak stations have been reduced.

    So it seems that I need to find the best resistor (the highest possible that won't cause the hum) that also allows me to tune into weak stations.
     
Loading...