1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

LED...The Perfect Bypass Capacitor

Discussion in 'Shock Brother's DIY Amps' started by XTRXTR, Apr 8, 2021.

  1. XTRXTR

    XTRXTR TDPRI Member

    Posts:
    82
    Joined:
    Jul 13, 2020
    Location:
    Northern Colorado
    Per Valvewizard other stuff page. He has several uses for LED. #2 on the cathode he states it replaces the resistor and capacitor and it acts as a perfect bypass capacitor.

    I wonder if anyone can explain what he means? I have a Blue 470nm LED on V1b cathode of my 2204 without any resistor or capacitor and it does sound really good to me. Clean tones with plenty of bass, dirty (full preamp on) also sounds rather nice with good control from the guitar volume and tone pots. 2.74Vdc on the cathode is 500mVdc higher than the spec'd voltage from Mark Huss schematic

    Has anyone here tried this mod? What do you think of it? Why does it work so well? I would like to understand it from an electrical perspective.

    I have played with this in all three modes of the 3-way cold clipper switch and have not found a downside. Why isn't this a standard on some tube amps already?
     
  2. edvard

    edvard Tele-Afflicted

    Posts:
    1,985
    Joined:
    May 15, 2016
    Location:
    Bremerton, WA
    tradition.gif
     
    stepvan likes this.
  3. TeleTucson

    TeleTucson Tele-Afflicted

    Posts:
    1,835
    Joined:
    Aug 6, 2016
    Location:
    Tucson, AZ
    LED's look like diodes electrically, and the fact that they emit light under forward bias is irrelevant here.

    However, depending on the material/design emission wavelength, the forward bias turn-on will be different than a typical silicon diode, and the reverse breakdown is usually quite low because of the high doping levels.

    So, in a nutshell, for applications like your advocating, it's more about the materials and doping that result in a particular I-V characteristic that you may find sonically appealing, nothing to do with emitting light.
     
    SRHmusic and XTRXTR like this.
  4. edvard

    edvard Tele-Afflicted

    Posts:
    1,985
    Joined:
    May 15, 2016
    Location:
    Bremerton, WA
    A while ago, on the DIYStompboxes forum, someone came up with the bright idea to bias FETs with a diode on the Source pin instead of a resistor and cap. The idea was that the LED would maintain its forward voltage (0.06 - 0.09 depending on the type, IIRC), thus looking like an ideal resistor/bypass-cap combo whose voltage at the source/diode junction would never change when adjusting the Drain resistor for optimal bias (FETs are horrendously fiddly).

    I hope that makes sense...
     
    XTRXTR likes this.
  5. XTRXTR

    XTRXTR TDPRI Member

    Posts:
    82
    Joined:
    Jul 13, 2020
    Location:
    Northern Colorado
    I'm not really caring that it emits light either. I was trying to determine why it was referred to as a perfect bypass capacitor. I'm sure I could find a resistor value that gives me the same VDC on the cathode but I doubt it would sound the same. Perhaps you gave me the answer and I just don't understand the explanation.
     
  6. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    724
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    It's a constant voltage source that determines the bias.
    Electrolytics can dry out etc. The perfect electrolytic is no electrolytic :).
    I think it is mostly used for hifi (Merlin has a section on it in his hifi book). In a guitar amp we want to determine the bias point for the clipping we want and use the bypass cap as a means to limit bass.

    In some cases a diode is preferred, other hifi solution incorporate a constant current source, and sometimes a resistor (+cap) is preferred.
     
    XTRXTR and SRHmusic like this.
  7. SRHmusic

    SRHmusic Tele-Holic Silver Supporter

    Posts:
    876
    Joined:
    Oct 19, 2020
    Location:
    North Carolina, USA
    As others say above, the diode will have a typically 0.6 to 0.8V drop for a wide range of currents. This gives a nearly constant voltage drop compared to a resistor where the voltage varies linearly with current.

    Note: In this case the diode needs to be used where the tube quiescent current is set by a different method. A resistor in the cathode leg provides negative feedback to stabilize the bias point. If you just swap out the resistor and cap for a diode you should recheck the bias point.

    Also as @Tom Kamphuys notes, the bypass cap and resistor allow you to roll off the bass response of the stage.

    So it's not a 'perfect capacitor' but more of a more constant voltage drop type of bias. A Zener might be interesting here, too.

    Edit: @Tom Kamphuys is correct, below, that LED forward voltages are higher, around 2 to 3V. These are a GaAs type of diode, while the silicon diodes have around 0.6 to 0.8V forward voltage.
     
    Last edited: Apr 8, 2021
    XTRXTR likes this.
  8. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    724
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    To add to that:
    The 1N4148 and alike are around the 0.7V. LED's are around 1.8V, a more common bias point. Some LED's are 3.0-3.5V.
     
    SRHmusic, robrob and XTRXTR like this.
  9. scooteraz

    scooteraz Friend of Leo's

    Posts:
    3,550
    Joined:
    Apr 17, 2007
    Location:
    Peoria, AZ
    LEDs have an intrinsic capacitance of about 50 pF on average. That combined with the voltage drops noted above create the bypass cap/resistor combination. However, LEDs come with their own issues, so the well known properties of just having a cap and a resistor makes as much sense as using an LED.

    See this article for use of an LED in a LRC circuit.

    https://www.edn.com/an-leds-intrinsic-capacitance-works-in-a-650-mv-lrc-circuit/
     
    XTRXTR likes this.
  10. XTRXTR

    XTRXTR TDPRI Member

    Posts:
    82
    Joined:
    Jul 13, 2020
    Location:
    Northern Colorado
    Thank you everyone for chiming in, I'll read everything I can find. I have to digest this info so far. Stay tuned for further questions please.o_O

    I think the 470nm LED I'm using has 2.74Vf which I think is why I measure 2.74Vdc on the cathode pin. The specs say 2.9 to 3.5. I have another LED on V2b plate the cathode follower for the tone stack, Listed as #1 in the same article, I mention it because it has the same drop 2.74V.
    The tube limits the LED current in both cases...right, other wise it would runaway and burn out.

    I have not seen any clipping, the guitar input signal is about 500mV max during strumming, I'll have to look at it more closely to be sure but I haven't heard anything like a clipping, especially while in the 39k mode on the cold clipper 3-way switch. I'll put the probe on the V1b plate next time. Perhaps the 34 yr old .68uf lego white cap I removed was bad because the tone is so much better than it was. I'll test out a new cap and resistor on the same cathode and see.
     
  11. Digital Larry

    Digital Larry Friend of Leo's

    Posts:
    2,857
    Joined:
    May 30, 2017
    Location:
    Silicon Valley, CA
    I don't think that the capacitance of your typical LED figures very much into how it works in this particular application. My view of electronics is that, ok, everything has some effect, just like there is a gravitational attraction between you and the moon. Which is why I do my weigh ins when the moon is overhead!

    I don't mean to disparage the original user of the term but it seems pretty far off the mark.
     
    SRHmusic likes this.
  12. King Fan

    King Fan Poster Extraordinaire Ad Free Member

    Posts:
    6,345
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Sometimes the light-emitting part of LED is a plus. Merlin reminds us in his trem chapter that it can give us a visual frequency / rate indicator: “By replacing the cathode resistor with an LED we eliminate the need for the large bypass capacitor, and obtain maximum gain and minimum output impedance at all frequencies in one fell swoop. A red LED will be suitable in most cases, providing about 1.6V bias.”

    23246850-DA60-4AB3-ABE9-8220C0403C96.jpeg
     
    robrob, FenderLover and XTRXTR like this.
  13. XTRXTR

    XTRXTR TDPRI Member

    Posts:
    82
    Joined:
    Jul 13, 2020
    Location:
    Northern Colorado
    @King Fan I remember reading that section and I forgot about the LED on the cathode of the oscillator. Inferring from that then its a "perfect bypass capacitor" because it provides a great gain level even at 2.74, but also provides the full audio spectrum. So rather than an expensive and large cap like a Solen at 1uf or 3uf to get the bass I can use an LED I bought in a bundle for pennies per unit.

    I think I might have an old red LED around in my extras bin, if I can find it, it might be worth the try. I like the bass tone I'm getting and the bright cap is doing its thing. I want to verify using a switch from an RC to LED that its not just my brain playing tricks. Maybe do a switch with a red LED to blue and RC using a 3-way selector.
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.