I'm aiming for around 5k total resistance and around 78dB attenuation in a 47-step, 46-resistor attenuator.
The bottom values are challenging ones. 0.67, 0.12, 0.15, 0.17, 0.2, 0.25, 0.3, 0.32, 0.45, 0.47, 0.62, 0.91, 1.15 Ohms). I can find some, but for others the selection is mostly wirewound and MOX.
Are there any 1/10 to 1/4W metal film resistor kits that focus on small values like these?
The bottom values are challenging ones. 0.67, 0.12, 0.15, 0.17, 0.2, 0.25, 0.3, 0.32, 0.45, 0.47, 0.62, 0.91, 1.15 Ohms). I can find some, but for others the selection is mostly wirewound and MOX.
Are there any 1/10 to 1/4W metal film resistor kits that focus on small values like these?
Doubtful, have you considered an attenuator using separate coarse and fine adjustments, like this one?
https://glass-ware.stores.turbify.net/atstat2.html
https://glass-ware.stores.turbify.net/atstat2.html
0, 0.1, 0.12, 0.15, 0.18, 0.2, 0.22, 0.24, 0.27, 0.33, 0.36, 0.39, 0.43, 0.47, 0.5, 0.56, 0.62, 0.75, 0.68, 0.82, 0.91Ω ±1% 1/4W 21values X 20 pieces each.
https://www.amazon.com/Resistor-Assorted-21Values-consuming-Elements/dp/B0CH9NJYSN
If you want this to be "high end" go for SMD resistors. With these you find the closest tolerances and smallest tempco. And because of size it is easy to use two in series to get the values you need.
First- 78 dB atteunation on the switch (or Pot) is a bit unrealistic. The previous stage would need enormous headroom for that to be useful. Second the switch contact resistance will be significant and not that constant. Also the stray capacitance will be an issue. The usual practice for getting large frequency flat response like this is cascaded attenuators. Add a second wafer to the switch and switch in a 40 dB atteuation at the appropriate point. This is typical practice in audio analyzers dealing with large dynamic ranges.
Most commercial steeped attenuators for recording and broadcast applications had a taper to zero at around -40 dB.
Those low value resistances are used for current shunts for measuring current. https://www.mouser.com/c/passive-co...-sense-resistors/current-sense-resistors-smd/ You can build with series parallel what you want but it gets expensive fast with the precision shunts.
Most commercial steeped attenuators for recording and broadcast applications had a taper to zero at around -40 dB.
Those low value resistances are used for current shunts for measuring current. https://www.mouser.com/c/passive-co...-sense-resistors/current-sense-resistors-smd/ You can build with series parallel what you want but it gets expensive fast with the precision shunts.
Suppose you have a two times 100 W amplifier, two 90 dB at 1 W, 1 m loudspeakers and a room with a 1 m reverberation radius. The maximum sound pressure level is then 110 dB from only one loudspeaker, 113 dB with two as long as they can be assumed to add in power.
40 dB below 113 dB is 73 dB, way too loud to step to complete silence if this stepped attenuator is to be used for domestic volume control rather than for mixing console channel gain setting. It could be even more, as I assume now that the maximum gain is only just sufficient to fully drive the main amplifiers.
78 dB below 113 dB is 35 dB, which is a rather more reasonable number if this stepped attenuator is to be used for domestic volume control. It can again be higher if there is some gain reserve somewhere.
If the attenuator is just a big voltage divider with taps connected to a big rotary switch, the switch contact resistance is not part of the voltage divider and does not affect the attenuation much, as long as the input impedance of the next stage is much greater than the contact resistance.
40 dB below 113 dB is 73 dB, way too loud to step to complete silence if this stepped attenuator is to be used for domestic volume control rather than for mixing console channel gain setting. It could be even more, as I assume now that the maximum gain is only just sufficient to fully drive the main amplifiers.
78 dB below 113 dB is 35 dB, which is a rather more reasonable number if this stepped attenuator is to be used for domestic volume control. It can again be higher if there is some gain reserve somewhere.
If the attenuator is just a big voltage divider with taps connected to a big rotary switch, the switch contact resistance is not part of the voltage divider and does not affect the attenuation much, as long as the input impedance of the next stage is much greater than the contact resistance.
I love it, but my faceplate was machined for a single volume control. A relay attenuator isn’t totally out of the question either, but introduces some enclosure space problems.
Thank you for the suggestion!
It’s for a series type attenuator, Goldpoint V47C. The lowest resistor sets the min volume level, then the resistance of each step adds and increases logarithmically, in this situation to about 5k ohms.
Last edited:
If you use a ladder attenuator, ie two resistors per step, then you only need two resistor values. Ok one more for the last step terminator. 2dB steps is about 80% so you need 1K series resistors and 20K shunt resistors, with a 4K (3.9K) terminator. It's much easier to buy 47 1K and 47 20K resistors than all those odd values, and not a lot of money since they come 100 for a couple bucks. (note 1k= (100%-80%=20%) *5K, 1/4k-1/5k=1/20K)
https://en.wikipedia.org/wiki/Resistor_ladder
https://en.wikipedia.org/wiki/Resistor_ladder
Thank you for the suggestion. It wasn’t easy to find one with enough steps.
The Goldpoint V47C (what I’m planning to use the resistors in) acts in similar manner to a potentiometer by moving the center pin connection along the resistors with each click. The thought in mind here is 1.5dB steps up to about -40dB, then 2dB the rest of the way.
The Goldpoint V47C (what I’m planning to use the resistors in) acts in similar manner to a potentiometer by moving the center pin connection along the resistors with each click. The thought in mind here is 1.5dB steps up to about -40dB, then 2dB the rest of the way.