Will An Attenuator Help Reduce This Hiss?


i've got a little bit of tweeter 'HASSSHHHH' that i'm looking to reduce. as soon as i turn on the amp and preamp it becomes audible. placing the preamp into standby, mute, or on an unused input does nothing to change the level. changing the volume of the preamp does nothing to change the level. the only thing i've found that changes it is fully powering down the preamp - this eliminates it entirely.

i've been advised to insert a line level attenuator (endler, goldenjack, etc) between the amp and preamp at the amp inputs to bleed off some gain and reduce this noise.

will an attenuator reduce the 'HASSSHHHH' sound when reducing the level, or even muting the preamp, does not change things at all?

thanks for any inputs,
Scott
128x128srosenberg
Hi Scott,

It might help, but I'm not certain.

And as Ralph points out some people have reported adverse sonic effects from in-line attenuators, although others have used them successfully. FWIW a while back I had used a pair of 10 db Rothwells at the inputs of the amp I used at the time, with no adverse effects that were apparent to me.

It seems clear that the noise is being introduced at a circuit point in the preamp that is "after" the volume control. So adding a 10 db attenuator, for example, would presumably reduce the level of both signal and noise by 10 db. You would then turn the volume control up by a corresponding amount, which would result in the original signal level being restored but would leave the noise at the 10 db lower level.

As a rough rule of thumb, btw, 10 db corresponds to what would be subjectively perceived as half the volume. 20 db corresponds to what would be subjectively perceived as 1/4 of the volume.

If you presently turn your volume control more than 1/2 of the way up, or perhaps even somewhat less, I would be hesitant to use the 20 db value, because you might find yourself running out of range at the upper end of the volume control's range of settings.

The main reason for the uncertainty I indicated is that the mute function does not reduce the noise. I would expect that in most cases the location of the mute circuitry in the signal path would be very close to the output of the preamp, and that it would kill any noise originating upstream of that point. Which suggests the possibility that noise generated by the preamp is either radiating through the air to the power amp, or is coupling to the power amp through the ground connections between the two components, or possibly through power wiring. An attenuator would probably not help in those cases.

IIRC you are using a W4S DAC2 as your preamp, with balanced connections to the power amp. The balanced connections presumably reduce the likelihood of coupling via the grounds, but the noise produced by the digital circuitry in the DAC2 could conceivably be coupling through the air or through power wiring.

Best regards,
-- Al
thanks for all the great insights, all. to clarify a few questions...

the speaker i am hearing this trough is a Wilson Cub 2 used as a center channel. as ralph surmised, it is relatively efficient at 95 db.

Al, great memory! yes, at one point i was uing the W4S DAC2 as a preamp (can't say enough good things about that piece) however, i have since reconfigured my system enabling both 2-channel and multi-channel operation. the preamp driving the center channel is a Meridian 568.2mm and the power amp is a McCormack DNA-HT1. the interconnects between the amp and preamp are single ended, as the HT1 only offers RCA inputs.
Seems to me that the McCormick has about 30db of gain. This is going to be a bit of a trick.

Do you know how much output your source has? Is is enough to drive the amp directly?
Just to be different, Prometheus TVC's offer up to 54dB attenuation and isolation as well for under $500. Haven't read of any shipping backlogs for years. You can take a look at the Bent Audio stuff as well. If you're up for some DIY, I believe both Goldpoint and Dact have "fine tuning" naked attenuators. The problem with the inline attenuators is that they're only 10K Zin, which is not a problem in itself except in more complex parallel arrangements or with some tube and passive preamps.
NGJockey raises a good point about impedance compatibility. According to my not particularly accurate analog multimeter, though, my 10 db Rothwells consist of a resistor of about 22K in series between input and output, and a resistor of about 10K or 11K shunted from the output to ground. That will result in an overall input impedance, as seen by the preamp, in the vicinity of 30K, the exact value depending on the input impedance of the amp. As NG indicated, that shouldn't be any problem for a solid state component such as the Meridian.

The output impedance of the attenuator, seen "looking back" from the amp, will be in the vicinity of about 7K. Given that the attenuator plugs directly into the input jacks of the amp, with no intervening cable, that would only be a problem if the input impedance of the amp is low AND varies significantly as a function of frequency. Perhaps that is why some people report adverse effects from these things, and others don't?

Best regards,
-- Al