10 db or 20 db attenuator?

@almarg 
True, true...
B

I suggest this. https://www.schiit.com/products/sys
I’m using one between my Rogue RP-5 and Nuforce STA200 which has super high gain (My ZU Speakers are 101 db). The attenuator allows me to lower volume a bit so tube hiss is not so loud. The unit is dead quiet and passive. You can control how much volume you need instead of replying on fixed settings. I tried the Harrison Attenuators and noticed a very slight "smeary" sound that I cannot explain. The SYS does not add any audio color or anything from what I can hear. Just a pure signal transfer. For $49, you can’t go wrong. Just my 2 cents

Thanks both for taking the time to respond. 

I also tried the Harrison Attenuators recently.  Bought 2 sets of the 10 db versions so that I could try them in line and get 20 db attenuation, as well as 10 db.

The sound quality with 10 db was slightly smeared, but some might say bearable.  With the 20 db set up the sound quality was terribly muffled - like listening with a cushion in front of the speakers.

The little Schiit device sounds interesting

The Harrison Labs attenuators use resistor values that are far lower than the values used in the Rothwells, and consequently they will present too heavy a load to be optimal for many components which may be used to drive them. Especially many tube-based components.


For example, in this thread a member reported measuring the 12 db single-ended Harrison attenuator as consisting of a series resistor of about 6.75K between its input and output, and a resistor of about 2.2K in shunt between its output and ground. That would mean that the component providing the signal would see a load impedance that is between about 8K and 9K, the exact value depending on the input impedance of the component receiving the signal. Many components, especially many of those that are tube-based, will not do well driving such a low impedance.


In contrast, I’ve measured the single-ended version of the Rothwell 10 db attenuator as consisting of a 22K resistor in series between its input and output, and a 10K resistor in shunt between its output and ground. That would mean that the component providing the signal would see a load impedance in the area of approximately 30K, the exact value again depending on the input impedance of the component receiving the signal. Also, I’d suspect that the balanced version has a similar pair of resistors for each of the two signals in the balanced signal pair, resulting in a balanced input impedance of roughly 60K. And I suspect those figures would be even higher for the 20 db Rothwells.


Also, btw, the 10K specified input impedance of the Schiit Sys (which will actually be even lower when the input impedance of the component receiving the signal is taken into account) would be too low to be optimal with some signal sources, again especially many of those that are tube-based. And as a passive preamp having a maximum output impedance of 5K the cables on its output side should be kept short. (No cables would be used on the output side of the Rothwells, since they are designed to be plugged directly into the input jacks of the component receiving the signals).  Finally, of course, the Sys is single-ended and the OP is looking for something balanced.


Regards,
-- Al