Negative Feedback Resistor value and noise floor

Discussion in 'Amp Tech Center' started by markeyd123, Nov 29, 2019.

  1. markeyd123

    markeyd123 Tele-Meister

    Posts:
    231
    Joined:
    Jun 27, 2017
    Location:
    USA
    I am trying to understand the role of the negative feedback resistor in a Fender 68 Custom Princeton Reverb Reissue amp. It has a 5.6k resistor in R10 which is the negative feedback resistor.

    How would a lower value resistor (like the 2.7k I believe is in the original Princeton Reverb) impact the noise floor?

    I am under the impression that more negative feedback means (a bunch of things, but I am concerned with background noise, hiss, etc.) less noise. If more neg. feedback equals less noise, then a lower value resistor would cause there to be less noise, and a higher value resistor would mean more noise??? Or do I have this backwards?

    Thanks in advance for setting me straight.
     
  2. King Fan

    King Fan Friend of Leo's

    Posts:
    4,222
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    You’re right, less resistor, more *negative* FB, less noise. At least if the tail resistor is the same.

    Why, if I may, do you ask? Noisy amp?


    Sent from my iPad using Tapatalk
     
    markeyd123 likes this.
  3. markeyd123

    markeyd123 Tele-Meister

    Posts:
    231
    Joined:
    Jun 27, 2017
    Location:
    USA
    Well, this amp in particular has been known to have some excessive noise and in this model they doubled the value of that resistor as compared with the traditional Princeton Reverb. I understand that if there is an underlying issue causing excessive noise that should be addressed first and I did have a tech look at it. At that time there was also an issue with the input jack causing some noise which was fixed . . . but there still seems to be some excessive noise. This may be an issue again, but I don't think so. I've gone through the tubes a million times. I tend to think the input jack is just shoddy, maybe, or it's that resistor which causes less negative feedback . . . since it is double the value of the traditional PRs value. I'd like to replace it with a 10k pot.
     
  4. Uncle Daddy

    Uncle Daddy Tele-Holic

    Posts:
    988
    Joined:
    Sep 26, 2015
    Location:
    Maldon, England
    Uncle Doug has a good Youtube video explaining negative feedback.
     
    markeyd123 and muscmp like this.
  5. LightningPhil

    LightningPhil TDPRI Member

    Posts:
    76
    Joined:
    Dec 1, 2018
    Location:
    Nottingham, UK
    To fight noise, I usually start with the power supply. The HT for all the preamp valves hang off C29, which is 22uF. This sits behind a filter comprising C27 (also 22uF) and R57 and R59. Both of which are 18k ohms. Increasing both these caps would lower the roll off frequency of the filter and since they’re behind resistors, not overload the rectifier. It would have some effect on hiss level. How much, can’t say, but expect you’d notice and it wouldn’t change the character of the amp much at all.
     
    markeyd123 and aerhed like this.
  6. Wally

    Wally Telefied Ad Free Member

    Posts:
    33,386
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    Markeyd123, if you wanted to install an adjustment pot there, I would suggest replacing that 5.6k top resistor with the original value 2.7k and put a 5K pot in series with that resistor. That would allow you go from the stock vintage Prin Rev 2.7k resistance to 7.7k....even less NFB than you have now. Decreasing strength of the NFB signal always increases the noise. That noise is not noticeable while playing, ime. If one wants the sonic content that comes with decreased NFB, the one learns to live with the noise.
     
    markeyd123 and King Fan like this.
  7. King Fan

    King Fan Friend of Leo's

    Posts:
    4,222
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    ^^^ what Wally said. I hoped someone smart would suggest a base-value resistor for your pot, and, FWIW, Fender's marketing description says they reduced NFB in the CPR to increase touch sensitivity.


    Sent from my iPad using Tapatalk
     
    markeyd123 likes this.
  8. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,521
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    The NFB should only be magnifying a problem that shouldn't be there. I built a PR earlier this year with a NFB switch, and though it raises the noise a tiny bit, it wasn't anything that was noticeable at any playing volume.

    My guess would be the cheap, tiny resistors they use on the board,along with any other cost saving measures. I switched to 1 watt carbon film resistors a couple years ago and haven't worried about white noise in amps since.

    Sent from my moto g(6) using Tapatalk
     
    LightningPhil and markeyd123 like this.
  9. markeyd123

    markeyd123 Tele-Meister

    Posts:
    231
    Joined:
    Jun 27, 2017
    Location:
    USA
    It is true that the noise is not noticeable when playing. Just a pet peeve because I know the amp could be quieter, as it has been quieter, while idling. For the edification of anyone reading this with the same issue: I had an epiphany . . . I remember someone saying to make sure the nut around the input jack is tight because this ensures a ground connection of some sort. I'll try that too when I get home later as it's a super easy fix. Maybe it came loose over time again.

    King Fan - I'd feel better knowing that the noise (or at least some of it) is a trade off for tone, as opposed to a problem with the amp. I'm OK with that, so thanks.

    Thanks Wally, I like the idea of the resistor in series with the pot. I didn't realize this was the way to achieve that, but now I see what you are saying.

    Lightning Phil - What would you change the 22uF caps to? What value? Thanks
     
    LightningPhil likes this.
  10. LightningPhil

    LightningPhil TDPRI Member

    Posts:
    76
    Joined:
    Dec 1, 2018
    Location:
    Nottingham, UK
    47 uF caps would be a good start. If looking through specs before purchasing them, appropriately small (in size) caps could probably be found. Alternatively, and I’ve done this on amps with glued on caps so as to not have to break the glue and keep things reversible, stick another couple of 22 uF caps on the back of the board. Good for an experiment, but do use some heat shrink over any exposed leads and keep it neat.

    The filter to the pre-amp is 2nd order, which has a 12 dB per octave roll off. So, by doubling the capacitance and halving the roll off frequency, anything above the original roll off that’s being injected into it from the output section should be decreased by 12 dB. And since an amplifier with some hiss coming out of the speakers must have the hiss in the output section (otherwise how does it come out of the speakers, even if it’s generated elsewhere in the circuit), this may have some effect. The net result will likely not be the full 12 dB reduction as its treating a symptom and not a cure. But it’s very easy and reversible.

    For cathode resistors, jsnwhite619 is right, better resistors are less noisy. Though I prefer the wire wound type for low values and metal film for high. But here again, “doubling” the value of the bypass caps from 22 uF to 47 uF would shave off 12 dB of any noise being generated by the resistor. For the normal 1.5k ohm and 22 uF combination, the roll off starts at 5 Hz. Decreasing this to 2.5 Hz should have no musical ramifications. Lots of amps use 47 uF anyhow...

    And now for something on the topic of the original question:)

    The difference in negative feedback resistor values is 5.6k vs 2.7k. Which is a factor of 2. And since it’s simply a voltage divider with the 47 ohm resistor, that’s a 6 dB change. So swapping in the original 2.7k resistor would reduce gain by 6 dB, along with any background hiss and noise. Tack another 5.6 k resistor on the top of the original as an experiment and see if you like it. Then maybe add a switch rather than a pot if you like both the original and tinkered with solutions.
     
    markeyd123 and Fluddman like this.
  11. markeyd123

    markeyd123 Tele-Meister

    Posts:
    231
    Joined:
    Jun 27, 2017
    Location:
    USA
    Initial problem solved:

    Great info, and I will likely experiment with all the great info provided above just because I do like to tinker, . . . I still like the idea of adjustable negative feedback in particular . . . noise, or no noise.

    . . . but for the sake of anyone that found this thread because they have a problem with background noise/hiss from this amp and are looking for a solution (The '68 Custom Princeton Reverb Reissue) I did find a simple fix: The nut on the face of the amp securing the input jacks were tightened down seemingly well. Still, I used a wrench and really tightened them with some elbow grease. Not enough to break anything, but with more torque than I would generally think I would need. The amp quieted down seemingly as a result!

    On an unrelated note (while we're here) another problem people complain of with this model is that the amp (speaker) farts out with higher output pickups like humberbuckers especially. This is fixed but jumpering the R34 resistor as per this thread:

    https://www.tdpri.com/threads/fender-68-custom-princeton-reverb-r34-location.580559/

    This mod makes the Bass knob much more usable as you can really go from 0 to 10 as opposed to "9 to 20" where 0 seems more like 9 should be before jumpering the resistor. It doesn't hurt anything . . . everyone who has this amp should try it as it's very easily reversed. Just solder one end of a wire to one side of the resistor, and the other end of the wire to the other side of the resistor. Thanks again everyone.
     
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.