Stepped Attenuator

[EWBrown]

I've gotten a couple of the ALPS stepped attenuators from this seller in Hong Kong:

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... gory=39783

These are pretty decent. $14 each plus shipping. He has them in 50K, 100K and 250K, (stereo)

If you want to roll your own, check this page from Triode Electronics,
he has the "blank" switches and there is a calculator for figuring out the resistors needed.

http://store.yahoo.com/triodeel/controls.html

/ed brown in NH

[erichayes]

Hi All

Before spending a relatively large amount of money and/or time on a stepped attenuator, try lowering the value of the control to 50K and see if the sonic characteristics change. I suspect you might be a victim of the Miller effect which can cause mild to severe treble cut depending on where the control is set. A stepped attenuator won't help, but sometimes a lower value pot will.

Eric in McKinleyville

[Thermion]

I think that the stepped attenuator is the right solution. Get those Rat Shack pots out of your signal path. If you roll your own, you can select any value, 10K, 50K, 100K, etc., that you want. You can even pick the attenuation rates between steps to fit your listening style.

I purchased one of the 24 position dual switches from Triode Electronics to use in my linestage. This is a very good quality switch. I haven't ordered resistors yet, but I used an excel spreadsheet to taylor the attenuations and calculate resistor values.

I found lots of good data on stepped attenuators on Goldpoint's site at http://www.goldpt.com/index.html.

JT

[gogzhad]

Hi,
Hoping Ed will see this - I need to make a passive attenuator for my headphones - I play CDs on my laptop while studying for background sound to deaden the ambient noise - problem is the output is much too loud - need an attenuator.
Ed - do you think those alps attenuators from the ebay guy would be suitable for such an application?
tia,
dave

[EWBrown]

The E-bay attenuators are 50k to 250 K which would be too high for headphones, which are typically 8 to 32 ohms. These would require a much lower total resistance.

A simple cathode (or emitter) follower between the line output and phones would make up for the impedance imbalance, without providing any voltage gain.

Do you know the output impedance of your CD player?

I bought one of the "fancy" step attenuators from kyc111 a few months ago, built in a "radial" configuration, it is about 2 1/2 inches diameter, rather than the longer "slim" version shown above. Very nice, each step uses two resistors in a true divider there the $14 ALPS ones are 24 resistors in a series string, which behaves more like a potentiometer.



/ed brown

[gogzhad]

I should have mentioned I was thinking of using this in a circuit I saw on the headwize page (http://headwize2.powerpill.org/projects ... y7_prj.htm) wherein a shunt resistor is used from the pot wiper to ground - circuit calls for a 100k pot. I was mainly interested in whether the channel to channel tracking was reasonably good, and if the action was smooth and noise free - I went ahead and ordered one - at that price, pretty low risk.
Anyway, Ed, don't have a clue about the output impedance of the CD since it's the output of my laptop computer most of the time - but it drives the crap out of my headphones - I want some background sound and I get down to the very bottom of the volume control to get the output I want, but there's no fine tuning and it kind of drops off a cliff right there near the bottom. Anyway, probably already written way too much about such a minor deal - but I knew you'd have an answer.
thanks,
dave

[EWBrown]

The $2.99 radioshack pots are OK for the price, but they don't track very precisely, and I've seen one which was intermittent (it had a hairline crack in the phenolic). As the old adage says - you get what you pay for.

The step attenuators use 1% tolerance resistors, and track very closely, when I was checking them out with a Fluke 87 DVM, I found less than 0.2% difference (2 parts in a thousand). Typically, they change the audio level 2 to 3 dB per step, an increment which is about the minimum detectable by ear.

A simple way out is to use a 5K or 10K audio taper pot, either stereo, or two monos, and just use that to feed the phones from the line output, it may not be the most elegant, or best solution, but it will give you the means to fine tune the level from the computer audio output. My SWAG is that the line level maximum is around 0.7 to 1.4 V P-P, and more if it is intended to drive external speakers.
The pot could be wired as a standard voltage divider, or as a simple variable series resistance.

/ed brown in NH