Stepping out

January 15, 2013
 by Paul McGowan

Perhaps the most expensive option a designer has to control the gain is called a stepped attenuator. These devices can range from the reasonably simple to the ultra exotic and expensive.

Cello’s 59 step platinum attenuator as well as Ayre’s 66 stepped device are exotic, expensive hand built beauties that exemplify the state of the art in volume control through a switch, rather than a pot. If you’ll recall in our post on pots, we discussed how the potentiometer is the most widely used volume control yet because of the way it’s made, is also the worst sounding of the available options. If the designer is going to use a resistive divider to control the volume, then a stepped attenuator can be the best option available.

A stepped attenuator is a simple multi-position switch that allows the user to change the volume by switching in (typically) a different combination of two resistors for each volume step. The more steps available, the more exotic the switch itself. Here the designer must pay close attention to the contact materials of the switch choosing between simple nickel, to silver, gold, rhodium or platinum depending on the budget and the sonic preferences of the designer.

A stepped attenuator can also be built with relays (which are just magnetically operated switches) and many high-end preamplifiers that include remote control use relays instead of a mechanical rotary switch.

The downside of a switched attenuator is the same as that of a pot: it adds a sonic signature. In fact, anything you place in the signal path is going to have a sonic signature that changes the music from what it was to something else. A pot or stepped attenuator can only degrade the signal it cannot make it better. The stepped attenuator has a very small sonic signature while the pot a larger one depending on its construction.

Now that we’ve covered the gain choices tomorrow we’ll move on in our list of 9 items.

Subscribe to Paul's Posts

5 comments on “Stepping out”

  1. Can you comment on a variable attenuator using a single hard wired variable resistance device like the one used in the old Melos preamps that used varying light shining on a device to vary the resistance?

    1. Using a photo variable resistor is a good way to go sonically – but they are quite difficult to synchronize between channels. You need a separate LED for each channel and the identical light output of one doesn’t give the same volume results for the other – requiring hand trimming and some loss of tracking.

  2. I’ve got two questions for your Paul. The first is the one I ask about bipolar transistors. In the process of producing a commercial recording, that is from the initial live take to the actual manufacture of the physical product whether it’s a vinyl disc, CD, DVD, or tape, hasn’t the signal gone through many potentiometers already? How many? At the very least in the master tape recorder recording amplifier and playback preamplifier, same for the mixdown recorder, mixing boards, any outboard or other processors such as Dolby noise recduction, and in the manufacturing equipment itself? Why does one more make all the difference?

    In looking at the circuit topology of all equipment showing the schematic diagrams of all of the transistors, tubes or whatever in a circuit, it’s easy to forget that for the output of every one of them, the current that supplies the signal for whatever comes next has come from the power supply through the bias resistors. Don’t they count? Don’t they have an effect? Same as those in a stepped volume control or potentiomenter?

    1. This is a great and often asked question Mark and I’ll do my best to answer but I am not sure it will satisfy you – and I haven’t all the answers anyway. Just a few educated guesses.

      In the recording chain the same effects we hear concerning pots, transistors, tubes, wires, and all things in the signal path have just as much an effect on the sound as the playback equipment does. Nothing different. So, if the recording process has 10 pots it goes through and two EQ banks, each successively adds its sonic signature to the sound. When you play it back you add yet another link in the sonic signature chain.

      What I think is missing here is a good A/B – because you’re comparing in a vacuum. Imagine if we had one group playing in a studio and two recording chains. One chain was the standard kitchen sink approach and the other was an extreme minimalist chain. Yo make a recording of each and then publish two CD’s. There’s no question in my mind that the minimalist CD would sound significantly better than the kitchen sink one.

      I would also remind everyone that this is exactly what folks like Blue Coast and Mobile Fidelity etc. do – they have super tweaked out minimalist recording chains and this is one of the reasons their releases sound so much better than a commercial recording.

  3. I understand what you are saying Paul but Blue Coast, Mobile Fidelity, and their like represent only a tiny fraction of the recordings on the market and the ones most people have bought and listen to most. Those vast majority of other recordings were produced in the conventional way using all of the devices that “high enders” (did I just coin a new phase?) disdain most. Would any of the transistors, volume controls, capacitors, and other circuit components used in the equipment that produced most of those recordings find their way today into the most expensive and presumably best performing playback equipment? I doubt it. So are you operating on “the straw that broke the camel’s back” theory that has it that just one or a few more such devices in the entire recording/playback chain is the vinegar that ruins the entire honeypot? If you want to talk about what for audiophiles would likely be their worst signal processing nightmare if they knew it (as they see it), think of what Dolby A professional does to an audio signal. How many times was it used for recording inherently dynamic range limited music like pop music unnecessarily because it was just part of the usual process?

    Vacuum tube signal processing some audiophiles insist is the only thing they’ll listen to produced some of the technically worst recordings made in the 1950s. You can hear the difference for example that RCA’s technology underwent from say about 1955 to 1960 when it appears that the bandwidth increased to the point where the top octave was finally on the recordings. I’d gues by the mid 60s most tube equipment had been replaced with bipolar transistor recording studio technology, and lots and lots of potentiometers.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram