Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

Preamp hiss

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Preamp hiss

    Any suggestions how to reduce the hiss of this circuit?
    (The input of a diy bass amp)

    Click image for larger version  Name:	20250117_232349.jpg Views:	0 Size:	97.7 KB ID:	1009668I thought it was below the hum noise, testing at home with a small FR speaker and headphones.
    Tuning the gain down and the master volume up - some residual hum noise from psu and the circuitry shows up.
    ​​​​
    Then I took the amp at the rehearsal room to the cabinet.
    15" mid bass + a horn compression driver.
    It started to pour out...
    Plugging mate's headphones, plugging the line out into mixing console and through stereo speakers - the hiss is there but not that annoying.
    Turning up the compression driver (bi-amping the bass cabinet) and it comes bold and loud

    I just got an idea to lower the gain pot value and the preceding attenuator resistor, so eventually the hiss is lower...
    There is a voltage divider cause the gain is way too much for a bass amp (6n1p) tube.

    Otherwise I like the double triode topology, cause it adds grit when distortion just opens up slightly.
    The first triode seems to need double less gain to do that without the attenuator resistor.
    How to lower the gain ?
    There is not much difference lowering the plate resistor on simulator.

  • #2
    Why are you using Q1 ? The impedance at R13 is low enough to drive R18/RV1. The IRF830 is too big for the job. Find a better MOSFET.
    WARNING! Musical Instrument amplifiers contain lethal voltages and can retain them even when unplugged. Refer service to qualified personnel.
    REMEMBER: Everybody knows that smokin' ain't allowed in school !

    Comment


    • #3
      The MOSFET does nothing to the signal, except it may allow me to lower R18/ RV1. F ex. 200k/ 100k or 100k/ 50k.
      Would lower resistance result in less hiss?

      Could I also get less gain so I don't need R18 at all?
      What about R21 - the second grid resistor?

      I will try another thing as well - a less sensitive midrange driver that plays up to 8-10k instead of the compression driver.
      Given the hiss is not that severe in headphones/ amp line-out - mixing console, it might 'mask' the hiss.
      If that works + reducing the preamp hiss would result in much lower hiss.

      Comment


      • #4
        Before changing anything it's best to identify where the hiss is coming from. What happens if you remove the tube - is it still noisy? I sometimes clip lead a small capacitor to ground (maybe 10nf) at different points to determine where hiss is originating.



        ​​​​​​
        Last edited by Mick Bailey; 01-20-2025, 01:26 PM.

        Comment


        • #5
          Do you really need that huge treble boost in the preamp?
          Speaking of C1 which mostly shorts R18 and so counteracts the attenuation at high mid and treble frequencies.
          This emphasizes the hiss.
          How is the hiss with C1 connected after R18?
          How is the hiss with the input shorted to ground?

          To lower gain without changing the sound you could split R23 for voltage dividing.

          Can we see the rest of the circuit?
          Last edited by Helmholtz; 01-20-2025, 03:38 PM.
          - Own Opinions Only -

          Comment


          • #6
            The treble boost is so that the tone does not sound dull when gain is turned down.
            SInce the gain is too much, it is allready turned down by R18 at max setting...

            In general I am not satisfied by the cabinet signature in terms of top end and mids. 15" + tweeter. The choice was pragmatic.. 1x15" speaker is cheaper than 2x10" and definatelly cheaper than 4x10". The 10"s have the mids and punch I prefer and much better coherency with the tweeter.

            I am bi-amping the cabinet. This allows me to turn up the tweeter (the hiss also coming up) and I am tempted to do so. But still it sounds thin on top and can't really fill the gap between the 15" and the compression driver

            Don't you mean to split divide R13 and how it is compared noisewise with the R18 attenuator ?

            The rest of the circuit is op-amps. 10 band EQ - send return loop - crossover at 2k5 - two channel limiter (FR/ tweeter).
            I don't even have the whole schematic in one drawing cause I put it up and build it module by module..
            But the hiss apparently increases when I turn up the gain.
            At gain to zero the hiss is gone and the hum noise floor takes over. I thought this was the noise floor when testing at home. But the compression driver revealed the hiss as much more apparent..

            The bass guitar onboard electronics makes it worse. The hiss goes up when I turn the treble up. I can test with a passive bass guitar but that's another issue and optimizing the amp preamp hiss worths the effort. The hiss would be much better with a better onboard bass electronics.

            Unfortunately I can't do anything unless I take the amp back home.. I might try swapping tube, shorting the signal (the 10nf cap test) and some simple stuff at the rehearsal room in the weekend...
            Last edited by emosms; 01-21-2025, 08:27 AM.

            Comment


            • #7
              I read this: https://music-electronics-forum.com/...it-plate-loads

              Originally posted by Chuck H View Post
              Thank you gents.

              One thing I would mention is that I've used split loads A LOT. One of my designs uses a 5X attenuation on one stage and a 3X on another. I don't notice any hum. What I do notice is very slightly less hiss because I eliminated two voltage dividers and their higher series resistance.
              Not clear how much 'less hiss' and eventually more psu hum would be if I move the attenuation to a plate split load and remove the interstage attenuator (R18)
              The attenuation now seems about 1/3.
              The reduced output impedance of v1 probably won't have any effect, because the mossfet buffers the output.

              Comment


              • #8
                But it could also be that the MOSFET low z output skews the snr, preserving hiss that otherwise would naturally roll of if the valve is loaded....
                I kept the first mossfet to preserve the most top end possible, and the second as a safe and solid way to pass the signal to the op amps.

                Comment


                • #9
                  How is the hiss with C1 connected after R18?
                  How is the hiss with the input shorted to ground?​
                  Please answer these questions.
                  Please also check the hiss with the input grid directly shorted to ground.
                  Last edited by Helmholtz; 01-21-2025, 04:43 PM.
                  - Own Opinions Only -

                  Comment


                  • #10
                    Originally posted by Helmholtz View Post
                    Can we see the rest of the circuit?
                    It would be very helpful to see the whole circuit as well as gutshots of the amp. Otherwise everything we see is filtered through a tiny lens.

                    Comment


                    • #11
                      Here are the 'guts':

                      Click image for larger version  Name:	1000003302.jpg Views:	0 Size:	1.50 MB ID:	1009783
                      The next circuits are on different drawings but it doesn't matter. When I turn the preamp gain down the hiss is gone.

                      Gain all way down, master all the way up - there is hum which I thought is 'the noise floor'. But the hiss comes much more annoying and louder through the compression driver. (Ok on headphones and line-out - mixing console, ok through the 15" as well).

                      Unfortunately I can't do any tests until the weekend
                      Attached Files

                      Comment


                      • #12

                        Where is the output from C24 connected? Also, having zener diodes in the signal channel is not the best way to limit. These diodes are a source of noise. No wonder noise generator circuits are built on them.
                        Click image for larger version

Name:	dataurl251836.png
Views:	89
Size:	365 Bytes
ID:	1009814

                        Comment


                        • #13
                          C24 is connected to a 10 band op-amp EQ.
                          Shorting input to ground does not reduce hiss.
                          Most of the hiss comes up when I turn up the gain pot (before the second triode).
                          It goes worse when I turn the bass guitar onboard EQ to the max (the treble pot).

                          The problem is nicely 'masked' by replacing the compression horn driver with a 5" midrange driver (on the picture).
                          The midrange should play up to 10kHz according to specs.
                          Moreover it sounds MUCH much better tonally, unlike the compression driver (thin and harsher sounding).

                          Click image for larger version  Name:	20250202_224215.jpg Views:	0 Size:	131.6 KB ID:	1010205

                          Much of the hiss is caused by the bass guitar onboard EQ.
                          I tried another bass, treble EQ all the way up - the hiss level is audibly lower.
                          The other bass could be switched to passive mode - the hiss is considerably lower and almost gone.

                          I am still open to preamp improvements, but since the amp is not at my place and we go to the rehearsal room in the weekends (sometimes skipping) the process of testing/ making adjustments will take longer ...
                          Last edited by emosms; 02-02-2025, 09:08 PM.

                          Comment


                          • #14
                            Originally posted by x-pro View Post
                            Where is the output from C24 connected? Also, having zener diodes in the signal channel is not the best way to limit. These diodes are a source of noise. No wonder noise generator circuits are built on them.
                            Click image for larger version

Name:	dataurl251836.png
Views:	89
Size:	365 Bytes
ID:	1009814
                            The zenners are not DC biased (C24 decoupling).
                            How they could possibly go into 'breakdown mode' and exhibit noise?
                            The noise is constant and amplified by the gain pot. Not amplitude dependant (if they go close to a breakdown at large AC signal as they are supposed to and eventually limit it)

                            Comment

                            Working...
                            X