Discuss B2B zeners X B2B Caps

Discussion in 'The Projects Forum' started by Macki, Jun 25, 2010.

  1. Macki

    Thread Starter New Member

    Jun 25, 2010
    Hello everyone,

    This is my first query/post on this forum. I have for quite some time consulted this site for various information and found the site to be a rich source for students or users with various interests, ambitions and/or background. My interest in electronics which I regret not pursuing in college, has mainly to do with audio amplification although I catch myself marveling over any circuit and not just its function but component layout or the overall design. It is all a work of art. Don't you think?

    What I have attached is probably a simple little DC blocking filter to many but there is some genius behind it. The network also couples the audio signal to output.

    At the input of the caps -otherwise the amplified audio signal- there is between 5-8 volt dc which is actually set by fixing the bias to the output devices configured in a quasi complementary manner. This 5-8 volt is actually output dc offset which is used to bias the driver stage upstream. Clever isn't it? Otherwise, this dc offset as you know should be zero or very close to zero.

    I would appreciate some discussion on this network, back to back zeners and back to back eletrolytics.

    With thanks and regards
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    The caps are in series with opposing polarities so that they can handle an AC signal to the speaker. The arrangement is pretty common. It lets you use inexpensive electrolytics in place of non polar capacitor types. In effect, you have an 11,00 uF cap in series with the speaker. Kinda hard to comment on any DC component with no idea of the amplifier circuitry.

    The zeners are not going to be useful. Take D7 for instance. Let the voltage on the cathode rise above 24 volts. What happens when the zener conducts? The speaker is, in essence, a 4 (or 8 or 16) ohm resistance to ground. With no limiting beyond the voice coil, the current passes through the zeners and the speaker.

    With a 4 ohm speaker, that might mean a path for 6 amps suddenly established. The result might be a fried output transistor and/or a blown voice coil, and perhaps both zeners smoked as well.

    If you are going to do signal limiting, it's a good idea to do so before the amplifier output.
  3. retched

    AAC Fanatic!

    Dec 5, 2009
    The zeners are signal clippers.


    Neat stuff. The zener clipper info is at the bottom of the linked page.

    What I dont understand is how, in parallel, they are doing much. I suppose they are providing feedback to the caps during breakdown sized signals, I guess that would cause audio distortion, but help keep the speaker from too high a voltage.

  4. Macki

    Thread Starter New Member

    Jun 25, 2010
    Thanks for the replies and link. Glad you found the circuit interesting.

    There is a 4-Amp fast-blow fuse in line with the input to the network which I cropped out of the attached image. The fuse should protect the speakers.

    As you have indicated, the network limits the signal to I believe 48V P2P. As you have also mentioned my take is that if this threshold is exceeded distortion occurs and output devices may be stressed beyond their intended dissipation to runaway.

    Interestingly, I have had to replace D8 a couple of times but not D7.

    Best regards
  5. John P

    AAC Fanatic!

    Oct 14, 2008
    I don't see how that can be true, since the zeners are in series with the speaker, not in parallel with a connection to ground. What I see happening is that until the voltage across the diode/capacitor combination reaches about 24V, the assembly is a capacitor, and would attenuate the low frequncies. But at a high drive level, low frequencies can drive the speaker via the zeners. So I wonder if this thing is designed to allow extra output amplitude for the bass, which might be needed if the speaker isn't very big, or if the desire was for a really thumping high-level bass response.
  6. Potato Pudding

    Well-Known Member

    Jun 11, 2010
    The Zeners are there to protect the Capacitors.

    They do provide some protection to the Speakers and Amp at the same time.

    Imagine the following scenario without those Zeners.

    Start with a Full Power lock beween one Rail and the speaker output.
    This could be caused by a DC input signal in an amp without input isolation.
    It might be a signal latching in the drivers ahead of the final output amp.
    It could be caused by a overheated power transistor that is in thermal conduction.
    It might be as simple as an overheated and dried out Filter capacitor that lowers one rail somewhere in the amp and causes an amplified offset error.
    Phase mismatch with the speaker might even cause an inductive bounce that actually sends the voltage higher than the power rails.

    Reasons not even required to be known yet - we have too much voltage trying to get out through the Capacitors into the speaker.

    The Capacitors are going to fail, and if they fail short before they fail open then you are going to lose the speakers too -voice coils burnt to charcoal -unless you pop a fuse instead.

    Now lets put the Zeners back in.

    They limit the voltage across the Capacitors so they will hopefully keep working. Whatever damage is done to the Speakers will be from 24 fewer volts.
    The problems of signal or final stage transistors that caused the full rail voltage are reduced because of the Zeners.

    The Zeners do not shunt output as would be expected with a clipper because the Capacitors are passing almost all of the AC signal. At least with a speaker connected.

    Lets try another situation where there is a problem. Lets short circuit the speaker connection to ground or just try to parallel 20 speaker cones of 4 Ohm each. (Not a good idea for anyone wanting to build their own speakers.) Now all the output voltage is across the capacitors. In fact the capacitors might try and do some voltage doubling at some frequencies. The Zeners might save those big Capacitors. It is a likely to be a sacrifice of the zeners and hopefully they provide a smoke warning that is noticed and the amp is shut down. The zeners should be cheaper than the capacitors. A Fuse should be even cheaper (if it hasn't been slugged) and the zeners will ALLOW the fuse to fail before the Capacitors (or final power transistors) do in a short circuit.

    Never slug a fuse. When a fuse opens figure out why it is blown. There is something wrong.
  7. Macki

    Thread Starter New Member

    Jun 25, 2010

    Thanks for the discussions.

    One thing I noted from your discussion was that the voltage across the two capacitors is 24v not 48v. My thinking was that since the caps are in series, however with their positives tied together, voltage across the two caps should be 48v while capacitance halves. The other thing I noticed was that the zeners are not clippers in this configuration.

    While searching the web on back to back capacitors, I came across an application guide for aluminum capacitors with the following paragraph on back-to-back caps which I thought would be of interest;


    (First page lower right)

    "If two, same-value, aluminum electrolytic capacitors
    are connected in series, back-to-back with the positive
    terminals or the negative terminals connected, the
    resulting single capacitor is a non-polar capacitor with
    half the capacitance. The two capacitors rectify the
    applied voltage and act as if they had been bypassed
    by diodes. When voltage is applied, the correct-polarity
    capacitor gets the full voltage. In non-polar aluminum
    electrolytic capacitors and motor-start aluminum
    electrolytic capacitors a second anode foil substitutes
    for the cathode foil to achieve a non-polar
    capacitor in a single case."

    The rectifying behavior of caps acting as if they are bypassed by diodes caught my attention...

    The analysis provided on this query in the previous replies were indeed both thought provoking and thorough. I was a bit taken a back at the extent of the workings of an assembly of caps and zeners I thought simple.

    Thanks again
  8. retched

    AAC Fanatic!

    Dec 5, 2009
    That may be worth looking into.

    I wonder if you can rectify AC without a diode drop using this cap method.

    I need to build another signal generator..SgtWookie broke mine.

    Well he helped me break mine. ;)

    back on topic, I have not seen this done, but has there been any experimentation or application of using this polar, back to back method as a bridge rectifier?

    I read the very informative PDF in the previous post, but it only mentioned that phenomena briefly.
  9. beenthere

    Retired Moderator

    Apr 20, 2004
    I looked at the article. As much as I respect CDE for knowing about capacitors, I have never heard of this capacitor rectification taking place. Since I have used back-to-back electrolytics for AC signal coupling, I have never observed the effect either.