Tube driven effects loop and other questions

Discussion in 'Shock Brother's DIY Amps' started by NSB_Chris, Feb 16, 2020.

  1. NSB_Chris

    NSB_Chris Tele-Meister

    Age:
    55
    Posts:
    188
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Hello All!
    I have been in the planning phase of my next build which will be a version of the RR800V. I liked the first one I built so much I am trying another one. This chassis configuration will be somewhat unorthodox, but that is something for another post. I am planning to put in a tube driven effects loop to be able to put time based effects after the preamp.

    First question is the circuit. My research found the circuit shown below as possibly the best.
    The easy part is I want to be able to completely bypass it if it is not used. I am thinking that gain adjustments before and after the loop are beneficial to be able to control the input level to the effects in the loop. Thoughts?

    Second question has to do with providing some indication of the drive level in the effects loop. Does anyone see the value of adding something like a VU meter either in the chassis or in a stomp box configuration in the effects loop to be able to confirm that the drive level is in a reasonable level for the pedals in the loop? I am guessing the answer is no and the advice is just do it by ear and make sure you don't hear unwanted distortion coming from the time based effects. However, with a heavily distorted input signal, sometimes it ends up being a subtle fissy lunatic fringe coming from the slightly over-driven pedal. I thought maybe it would be beneficial to have some sort of measurement of the input signal from the preamp. I see a lot of discussion of the limited value of VU meters.

    Thanks for any advice/opinions!
    Chris

    Reply #8 of: https://el34world.com/Forum/index.php?topic=10208.0
    [​IMG]
     
  2. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    I don't know what an RR800V is, but if it is anything like a 2204, here is a tube buffered loop.

    FX Loop AC CF #4 non MV.png
     
  3. NSB_Chris

    NSB_Chris Tele-Meister

    Age:
    55
    Posts:
    188
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Thanks! The RR800V is Rob Robinette's JCM800 micro that is essentially a 2204 preamp with a 6V6 power amp, so you are spot on. The real difference is that I have two 2204 preamp channels that mix together after the master volumes. This mixed signal would go to the effects loop bypass switch.

    The circuit you have there has a few fundamental differences to the one I posted:

    Yours has the send level adjustment pot at the input of the cathode follower and the one I posted has it after. Any advantage one way or the other?

    On your circuit, what is the purpose of the 100k resistor between the input side of the return level pot and the wiper? Results in the top side of the voltage divider going between 0 to 90k and the bottom side going between 1M and ground. Trying to wrap my head around what that does over the simple voltage divider.

    I noticed a large number of reference circuits for effects loops had a 22M resistor and a capacitor on the return stage running from before the grid stopper to the plate. What is the purpose of that?

    Sorry for all the questions! Hoping to learn something from this and avoid a lot of iteration getting this to work.
     
  4. Badside

    Badside Tele-Meister

    Posts:
    379
    Joined:
    Jul 5, 2011
    Location:
    Montreal, Qc, Canada
    I've never seen this before either, but thinking about it I can see one interesting use case: in effect, the Return control also changes the input impedance. So assuming you're driving the return with a line-level device, you'll probably leave the control lower and this will effectively reduce the input impedance to as low as 100k which is better for line level devices (lower input impedance = less treble loss due to cable capacitance), while going all the way up for a guitar level device will increase input impedance to 1M which is what these devices expect.
    Where it gets trickier is that this also changes the taper of the control a LOT because the 100k is always in parallel with the "wiper to top" resistance.

    Either way, it's... interesting. Personally I just omit send and return controls and tweak resistors values so that it doesn't overdrive pedals and that the volume is the same whether the loop is used or not. Just less stuff to worry about. I only use my loop with pedals anyway, so I don't really care. I still keep the return input impedance to 100k or less because pedals can handle it.

    Only purpose I would see here is to get rid super high frequencies (less chance of hearing radio in your amps), although a 22M resistor is so ridiculously high it might as well not be there. Maybe it's a typo and it should read 2.2M? It does say check for errors. If I understand correctly, this is based on a Dumble loop and... well there are lots of myths about those and people believing this kind of stuff is secret to the tone. All a 22M resistor is good at is adding noise (noise is proportional to the resistance value). Just my opinion of course, but there are simpler ways to low-pass your return signal.

    Personally... I've moved to MOSFET driven loops in recent builds. So simple, and in buffer mode they basically have no impact on tone (you're sending that signal through solid state effects anyway). I still recover with a tube stage (first rule of MOSFET: never for signal gain). This frees up a triode to do other things.
     
  5. Mexitele Blues

    Mexitele Blues Tele-Holic

    Age:
    42
    Posts:
    839
    Joined:
    Mar 7, 2018
    Location:
    Westminster, CO
    That appears to be a local negative feedback loop at the return stage. I'm not sure that a 22M here is going to allow a meaningful amount through to clean up the signal.
     
  6. NSB_Chris

    NSB_Chris Tele-Meister

    Age:
    55
    Posts:
    188
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    This is good stuff. Thanks for the input!

    That makes sense. I never would have thought of that. That may be why they chose 250k for the return pot on the schematic I referenced. It is just put somewhere in the middle between 1M and 100k. I would only be using pedals, so perhaps I should just use 1M pot on the return side of whatever I end up putting in.

    I originally thought the 22M resistor was a mistake, but I saw it several places. Could actually be an original typo that has been propagated. I can't say my search was exhaustive, but I didn't see anyone question it.
     
  7. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    How are you mixing them? Any chance you could post a schematic of some sort?
     
  8. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    The most fundamental difference is that I took a set of parameters and designed a loop that would meet or exceed those parameters. The loop that you posted either had no parameters to meet or the parameters were unacceptable.

    The loop you posted is taken from a Dumble amp and may have been designed to work in conjunction with the rest of the amp. It has some flaws, however, that exist no matter what the intent was.

    One of my parameters is that the the loop must perform adequately even when the guitar player has set the controls as stupidly as possible. One of the things he is going to do is to put the largest signal possible into the cathode follower. The loop must be designed so that the largest signal possible doesn't clip the cathode follower. Had the loop been fed from the plate of a 12AX7, a resistor would be necessary between the plate and the send pot. Since the 2204 feeds the loop from a tone stack, the tone stack acts as that resistor.

    In normal operation, the guitar player is going to have to turn the send level down in order to get the proper signal level for his device and the cathode follower will be operating well within its capabilities. The Dumble loop allows the user to overdrive the cathode follower and still get a useable signal level for his device.
     
    NSB_Chris likes this.
  9. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    My loop is an adaptation from one that had master volumes in it and I left the return circuit the same. You could take the 1M pot and the 100K resistor out and just put a 100K pot in instead.

    FX Loop AC CF #3 w MV.png

    The Miller effect is going to shear high frequencies off and how much it shears depends on what you have in series with the grid. The more resistance you have in series with the grid, the more the highs are attenuated. A 1M pot is a problem because it starts putting too much resistance in series with the grid part way through its rotation. A 100K resistor in parallel with the series portion of the pot keeps it from becoming large enough to cause problems. The only reason to use a 1M pot is that it is a dual ganged pot and the other pot must be 1M.

     
    NSB_Chris likes this.
  10. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    I'm having a lot of trouble with the font size changing at random today. A guy told me how to fix that one time, but I have forgotten.

    The output impedance on most effect devices is extremely low. This is a good thing because they can drive really long cables without high frequency attenuation. It also means that they can drive return circuits with low input impedances. Driving into a 100K input impedance is child's play for these devices.
     
    NSB_Chris likes this.
  11. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    It's local negative feedback. But 22M and 220nF? And a tone-shaping circuit within the NFB? It may have done something that Dumble wanted, but it doesn't do anything that you want.
     
  12. Badside

    Badside Tele-Meister

    Posts:
    379
    Joined:
    Jul 5, 2011
    Location:
    Montreal, Qc, Canada
    Well, I can't say what the original intent was, all I can say is that this happens as a byproduct of this approach.
    But like Ten Over said, better just always have a low input impedance for the FX loop return (I use 100k), this reduces "tone suck" when running long cables (due to the cable's capacitance forming a low-pass filter with the grid leak resistor). Pedals are active device and won't care. It only matters if for some reason you wanted to plug a guitar or bass straight into the FX loop return. I've tried it for fun, doesn't sound particularly good.
     
  13. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    The input impedance of the loop return circuit has very little to do with high frequency loss from the cable.

    Let's say you have a 50' cable coming out of your effects device and going to your loop return. The cable has a capacitance of 30pF per foot for a total capacitance of 1500pF. At 15KHz, a 1500pF capacitor has an impedance of 7.07K, so the cable is the equivalent of a 7.07K resistor to ground from the point of view of the effects device. The input impedance of the loop return circuit is in parallel with the 7.07K. A 100K loop input impedance makes the impedance seen by the device 7.06K. A 1M loop input impedance makes the impedance seen by the device 7.07K.
     
  14. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    The output impedance of the loop as well as the output impedance of the effects device has a huge effect on the high frequency loss from cables.

    The Dumble loop has a 250K send pot so that the lowest output impedance at the worst case is 62.5K. If you want a cut off frequency of 15KHz and your output impedance is 62.5K, then the maximum capacitance is 170pF. At 30pF per foot, the maximum cable length is 5.67 feet. That won't even reach the ground with a full stack.
     
  15. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    The output impedance of my loop is 6.5K. For a 15KHz cut off frequency, the maximum capacitance with 6.5K is 1630pF. That's 54.33 feet at 30pF per foot.

    The output impedance on most effect devices is even lower, sometimes as low as 100 Ohms.
     
  16. NSB_Chris

    NSB_Chris Tele-Meister

    Age:
    55
    Posts:
    188
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
  17. NSB_Chris

    NSB_Chris Tele-Meister

    Age:
    55
    Posts:
    188
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Good point! I am liking your circuit more and more. Since I would be adding this between the mixing resistors of the two channels (with master volumes) and the phase inverter, wouldn't it be best to just use your first circuit?
     
  18. Paul-T

    Paul-T Tele-Meister

    Posts:
    300
    Joined:
    Nov 23, 2018
    Location:
    London
    For reference, some of those reading this thread might be interested in Sluckey's active FX loop (with added reverb).
     
  19. Ten Over

    Ten Over Tele-Holic

    Posts:
    523
    Joined:
    May 13, 2015
    Location:
    Central California
    I did the math a little quickly and made some assumptions, but here is one to consider.

    FX Loop NSB Chris.png
     
  20. NSB_Chris

    NSB_Chris Tele-Meister

    Age:
    55
    Posts:
    188
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Edited:

    Do you think that it is a bad idea to have both the master volumes and the send level before the send 12AX7?

    Since I am not using a PPIMV, I am not in much danger of overloading the send stage with your revised schematic.

    Thanks for looking into this so quickly! I am going to lay out that configuration.
     
    Last edited: Feb 17, 2020
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.