• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Low distortion attenuator for building a passive pre-amp?

LaLaLard

Member
Joined
Jul 14, 2019
Messages
75
Likes
41
Location
Perth, Western Australia
Hi all,

Just wondering if you amazing folks could recommend some good attenuator for a passive pre-amp project. Using Hypex module for power amp part thus don't really want the attenuator to ruin the SINAD.

Ideally < $50, thanks.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Including a box, XLR connectors and 4 gang log type potmeter with excellent tracking ?
 
OP
LaLaLard

LaLaLard

Member
Joined
Jul 14, 2019
Messages
75
Likes
41
Location
Perth, Western Australia

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
stereo ? balanced (4-gang) ?, log (volume control) or linear (gain control)
Hypex themselves recommend to drive it from max 50 Ohm source R.
This rules out passive attenuators.
 
Last edited:

egellings

Major Contributor
Joined
Feb 6, 2020
Messages
4,005
Likes
3,245
I'm not sure just what kind of distortion that resistors can cause with small signals imposed upon them, outside of a tiny pinch of noise, and maybe contact problems for attenuators.
 
OP
LaLaLard

LaLaLard

Member
Joined
Jul 14, 2019
Messages
75
Likes
41
Location
Perth, Western Australia
stereo ? balanced (4-gang) ?, log (volume control) or linear (gain control)
Hypex themselves recommend to drive it from max 50 Ohm source R.
This rules out passive attenuators.
I'm currently using 2 custom built monoblocks built from NC500 module with a passive preamp (basically just an attenuator with RCA in & out) and the outcome is great. Basically I'm just trying to build another one with attenuator built-in.
 
OP
LaLaLard

LaLaLard

Member
Joined
Jul 14, 2019
Messages
75
Likes
41
Location
Perth, Western Australia
I'm not sure just what kind of distortion that resistors can cause with small signals imposed upon them, outside of a tiny pinch of noise, and maybe contact problems for attenuators.
true, but contact issue seems relatively common among attenuators and Ideally I want the noise from the resistor smaller than the Hypex modules (I recall the measured distortion of the NC400 module poweramp on this forum was 0.001%) which is quite low. No point in building with Hypex modules if noise from other part of the circuit is greater than that right?
 

egellings

Major Contributor
Joined
Feb 6, 2020
Messages
4,005
Likes
3,245
Erratic contact (switch or wiper) resistance will likely be the largest contributor to noise in an attenuator of reasonably low total resistance. It's good to get high quality there. Distortion, caused by nonlinearities in the resistor behavior, will likely not be a problem though, (it's way too tiny) unless you are using carbon composition resistors in a switched setup. Metal film resistors are cheap and work very well. So for me, a high quality switch and metal film is the way to go for optimum performance. Another possibility would be a chip made for the purpose that utilizes a thin film laser trimmed resistor network controlled with a digital signal. These use JFET switches, and the resistor and JFET together are trimmed so that the total resistance hits the target value. That might be the optimal choice--no contacts at all!
 
OP
LaLaLard

LaLaLard

Member
Joined
Jul 14, 2019
Messages
75
Likes
41
Location
Perth, Western Australia
Erratic contact (switch or wiper) resistance will likely be the largest contributor to noise in an attenuator of reasonably low total resistance. It's good to get high quality there. Distortion, caused by nonlinearities in the resistor behavior, will likely not be a problem though, (it's way too tiny) unless you are using carbon composition resistors in a switched setup. Metal film resistors are cheap and work very well. So for me, a high quality switch and metal film is the way to go for optimum performance. Another possibility would be a chip made for the purpose that utilizes a thin film laser trimmed resistor network controlled with a digital signal. These use JFET switches, and the resistor and JFET together are trimmed so that the total resistance hits the target value. That might be the optimal choice--no contacts at all!
This sound like a solution. I'll look into that! Thanks
 

raindance

Major Contributor
Joined
Sep 25, 2019
Messages
1,037
Likes
968
Alps blue log pot, 10K or so, from Parts Express or similar to avoid fakes.
 

JonP

Member
Joined
Jul 7, 2020
Messages
55
Likes
24
+1 on the Alps, for a easy solution. A stepped attenuator would probably be better on channel balance, and other above reasons.

Point to note, staying with lower resistance values (like 10K) is a good idea. A little story here... I got one to use in front of a chipamp board, just happened to be handy, it was a 100K. Didn't worry much about the high R value. Much time goes by, and I do some measurements on the amp with the volume control at lower setting... and I find a roll off above 7khz! Where'd that come from?? Usually, I'd have the pot either all the way up, or all the way down when plugging/unplugging cables, driving it with some source that I'd control for the volume level.

So, I hadn't noticed a lack of highs, till I was using it at a DIY speaker event, with long cables. We were puzzled by dull highs on a speaker, eventually went to the host's AV amp and it sounded much better. I went home to check out my amp, (LM3886, should be a good performer) and found the volume control related problem.

What was going on, was the high resistance of the pot, combined with the longer RCA cables (higher capacitance), was creating a RC filter. When you were at partial volume, it was such that it was down in the audio range, and impacted the frequency response of the amp! I wasn't expecting that at all, started to measure various cables with a good LCR meter, and found that they can be very different, and a 10-12' RCA typically has more C than you might expect. So, keeping the cables short helps (less C) but having a lower input impedance (lower R) also makes whatever RC filter you make, above the audio range.

I wonder if this effect might have something to do with creating some cable mythology... Some amps have very high input impedance, and some fancy audiophile cables might happen to have a lot less C per foot than others? "Ahh, now this cable removes another veil between you and the music...."

Final thought... see if whatever chip/JFET solution you look at has distortion specs, or is measured by someone. I had seen chip "programmable potentiometer" parts years ago, think I remember they were noisier, or had more distortion than most average audio amps back then. I dont' know if there's more "audio high end" versions available now that wouldn't be a step down for your Hypex.
 
Last edited:

EchoChamber

Addicted to Fun and Learning
Forum Donor
Joined
May 22, 2020
Messages
673
Likes
925
Hi all,

Just wondering if you amazing folks could recommend some good attenuator for a passive pre-amp project. Using Hypex module for power amp part thus don't really want the attenuator to ruin the SINAD.

Ideally < $50, thanks.
Assuming your sources are all digital, perhaps bypass an analog atenuator all together and use a DAC’s digital volume instead.
 

raindance

Major Contributor
Joined
Sep 25, 2019
Messages
1,037
Likes
968
+1 on the Alps, for a easy solution. A stepped attenuator would probably be better on channel balance, and other above reasons.

Point to note, staying with lower resistance values (like 10K) is a good idea. A little story here... I got one to use in front of a chipamp board, just happened to be handy, it was a 100K. Didn't worry much about the high R value. Much time goes by, and I do some measurements on the amp with the volume control at lower setting... and I find a roll off above 7khz! Where'd that come from?? Usually, I'd have the pot either all the way up, or all the way down when plugging/unplugging cables, driving it with some source that I'd control for the volume level.

So, I hadn't noticed a lack of highs, till I was using it at a DIY speaker event, with long cables. We were puzzled by dull highs on a speaker, eventually went to the host's AV amp and it sounded much better. I went home to check out my amp, (LM3886, should be a good performer) and found the volume control related problem.

What was going on, was the high resistance of the pot, combined with the longer RCA cables (higher capacitance), was creating a RC filter. When you were at partial volume, it was such that it was down in the audio range, and impacted the frequency response of the amp! I wasn't expecting that at all, started to measure various cables with a good LCR meter, and found that they can be very different, and a 10-12' RCA typically has more C than you might expect. So, keeping the cables short helps (less C) but having a lower input impedance (lower R) also makes whatever RC filter you make, above the audio range.

I wonder if this effect might have something to do with creating some cable mythology... Some amps have very high input impedance, and some fancy audiophile cables might happen to have a lot less C per foot than others? "Ahh, now this cable removes another veil between you and the music...."

Final thought... see if whatever chip/JFET solution you look at has distortion specs, or is measured by someone. I had seen chip "programmable potentiometer" parts years ago, think I remember they were noisier, or had more distortion than most average audio amps back then. I dont' know if there's more "audio high end" versions available now that wouldn't be a step down for your Hypex.
Yes, most tube amps and preamps have volume controls, which often determine input impedance, between 100K and 1M (1M being quite common in older gear). Then there are certain idiotic manufacturers who make passive preamps using 100K pots - which is typically twice or more the input impedance of a lot of consumer gear... this when the preferred ratio is that it is 10x LOWER :D

This is an interesting theory of the cable acting in conjunction with the pot to create a filter.
 

JonP

Member
Joined
Jul 7, 2020
Messages
55
Likes
24
Heh, Theory? It will do that. It was interesting to do the math... I did up the values in LTSpice and drew some graphs, and also there's a nice calculator for stepped attenuators/passive preamps, that DACT has: http://www.dact.com/html/ac_calculator.html Try it out. Actually, if the OP wants to build or buy a stepped attn, or passive pre, the site would be a good resource. They don't make cheap stuff, but others are more affordable.

Edit: You will see the most effect at around half volume, tapering slightly away to no effect at 0% and 100% volume. So, as I usually used it (100%), I didn't see it, but at the event, using it as a volume control, it showed up. Varying, with the different volume settings, with different speakers having different sensitivites.... A real lesson in knowing your setup, if you want to depend on it being predictable.

Got my own LCR meter a while ago, and just dug up some random RCA cables for some values, since I've forgotten the years ago measurements:

Cable A: Cheap 3' stereo RCA 270pF 90pf/foot
Cable B: Even Cheaper 5' stereo RCA 173pF 35pf/foot
Cable C: Digital 8' RCA (coax) 136pF 17pf/foot

So, you can see that the capacitance per foot can vary quite a bit, depending on the cable. The Digital cable is probably a true 75 ohm Coax, while the other audio stuff varies more.
I have used up to 12' of the Cable A stuff, which would be 1080pF, or 1.08nF, try that out with a 100K pot! I was using one of those for the DIY Speaker event. My apologies for unintentionally rolling off everybody!!
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The input resistance of the modules is 1.8k Ohm and is balanced.
In this case, when you really want to get the most out of the NC500, I recommend to use a 4-deck stepped attenuator with 2k total resistance.
They do recommend a 50 Ohm source.
I don't know the exact reason for this, could be to ensure specs are met. It is quite possible that it may work well with a higher value source (potmeter) though.
 

raindance

Major Contributor
Joined
Sep 25, 2019
Messages
1,037
Likes
968
Aha. Yeah that's because it needs a buffer. 1.8K is very, very low.
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,823
Likes
2,951
Location
Ichihara City, Chiba Prefecture, Japan
Last edited:

Kal Rubinson

Master Contributor
Industry Insider
Forum Donor
Joined
Mar 23, 2016
Messages
5,273
Likes
9,794
Location
NYC
Assuming your sources are all digital, perhaps bypass an analog atenuator all together and use a DAC’s digital volume instead.
Yes, especially appealing if you are using multiple channels. Those high quality 8/16 channel attenuators/pots get expensive.
 

JeffS7444

Major Contributor
Forum Donor
Joined
Jul 21, 2019
Messages
2,347
Likes
3,509
By the way, it's a mistake to think that passive "preamps" are necessarily higher-fidelity than active devices: I've experienced 100% distortion when using passives: No sound at all because I had an source component with high output impedance, and was trying to drive a longish cable - a real no-no. If I knew then what I know now, I'd have fully buffered that volume control.
 

mhardy6647

Grand Contributor
Joined
Dec 12, 2019
Messages
11,217
Likes
24,183
The comment above about source impedance* is, of course, really important.
Sometimes folks need gain, drive, and/or impedance matching -- and in those cases, a passive just ain't a gonna cut it.

For passive attenuation, from my perspective, it is hard if not impossible to beat a tapped transformer or autoformer, since it allows attenuation at constant impedance. It does require wide bandwidth, so they ain't cheap, but any constant-impedance stepped attenuator ain't a-gonna be cheap, even if it just uses resistors.

http://www.intactaudio.com/atten.html

010 by Mark Hardy, on Flickr

FWIW: My (stereo) "passive" is in the silver box on the left. That's an "Azden" branded (passive) source/dubbing switch on the right.
EDIT: Just to explain the silver box. The knob on the left is a ganged stereo TVC, the right knob is a three position source selector. One of the inputs to the box comes from the Azden (which can switch four inputs itself, one of which can be used in a dubbing loop).

_______________
* I.e., and more generally, the importance of complementary source output and load imput impedances. :)
 
Last edited:
Top Bottom