You are here

Passive Monitor Pre-Attenuator

Customising, building or repairing your own gear? Need help with acoustic treatment or soundproofing? Ask away…

Passive Monitor Pre-Attenuator

Postby Pringe » Thu Apr 02, 2020 11:32 pm

I'm currently planning to make a passive controller - I've been doing a fair bit of reading up on the topic and I know it's been discussed a lot with the merits of passive/active etc. so I'll try to keep my question succinct but would be great to have some wisdom with some theory I'm trying to figure out!

I am using a 10k log 4-gang 24-stepped potentiometer and have checked the value of all resistors for good matching. My source impedance is 100 ohms and my load impedance is 30k ohms. I am planning to configure the attenuator as a simple voltage divider (as opposed to another common method where the pot is wired a variable resistor between +ve and -ve signals.). However if I wanted to insert a pre-attenuator voltage divider to give me a -20dB pad I have seen recommended values of 68.1k for R1 and 30.1k for R2 (followed by the 10K pot)....my question is regarding how this affects the source impedance 'seen' by the load. Am I right in thinking that this would become 100R + 68.1K + (0 --> 10k) or am I misunderstanding? I'm querying it because it seems very steep for a source impedance (following the 1:10 minimum rule) and obviously puts it higher than my 30k load impedance but am wondering if I've missed something somewhere.

Thanks in advance for any assistance!
Pringe
Regular
Posts: 61
Joined: Wed Sep 01, 2004 12:00 am

Re: Passive Monitor Pre-Attenuator

Postby Hugh Robjohns » Fri Apr 03, 2020 11:25 am

I think you've just discovered why passive attenuators are difficult to work with in a versatile monitoring controller ... :-)

Obviously you can choose different values for the pad resistors, but there will always be a compromises involved.

Of course, you don't need a completely separate R1/R2 voltage divider for this... you only need to switch in an extra series input resistor at the front of the attenuator itself to provide an additional 20dB attenuation...
User avatar
Hugh Robjohns
Moderator
Posts: 27016
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Passive Monitor Pre-Attenuator

Postby Pringe » Fri Apr 03, 2020 4:39 pm

Ha yes it's a good exercise in getting to grips with some of the maths involved! Definitely seeing the compromises too :D

Just been working through it - with the 68.1k and 30.1k I mentioned I noticed that R2 sits in parallel with the attenuator, giving a divider of R1 68.1k and R2 7.5k for the -20dB pad. Without R2 I calculated that I'd need a series resistor of 90k before the attenuator to give the same -20dB pad ( 20log (10k / R + 10k) = 20 ), does this sound correct?

I can see how having an effective source impedance too high will result in a loss of voltage at the amp input (although I'm struggling with the maths to calculate exactly how much!) and in effect raise the noise floor, are there any other disadvantages to this?

One other question: when input/output impedances are given, is it effectively a series resistance, or a resistance between the input and ground?
Pringe
Regular
Posts: 61
Joined: Wed Sep 01, 2004 12:00 am

Re: Passive Monitor Pre-Attenuator

Postby Hugh Robjohns » Fri Apr 03, 2020 4:46 pm

Pringe wrote:I noticed that R2 sits in parallel with the attenuator...

Yep... that's why I suggested you don't need it. The attenuator itself becomes R2, and you just need to add a switchable series input resistor to increase the attenuator's R1 value.

...does this sound correct?

Sounds plausible... try it and see.

I can see how having an effective source impedance too high will result in a loss of voltage at the amp input

It's not the loss of voltage you need to worry about, it's the effect of the low-pass RC filter it forms with the connecting cable's capacitance...

One other question: when input/output impedances are given, is it effectively a series resistance, or a resistance between the input and ground?

In an unbalanced system, the output impedance is the effective resistance in series with the hot side of the source's voltage generator (the cold side being grounded). The input impedance is a parallel resistance between the input signal and ground, across which the wanted signal voltage is developed.

In a balanced system you effectively have two unbalanced system that share a 'virtual' common reference, so the impedances are shared across both legs.
User avatar
Hugh Robjohns
Moderator
Posts: 27016
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Passive Monitor Pre-Attenuator

Postby Pringe » Fri Apr 03, 2020 5:10 pm

Thanks so much Hugh! I think I need to go and read up on calculating the RC cutoff point because I was visualising the circuit in terms of:

1: an RC low-pass filter formed with the source impedance and the input cable core-screen capacitance.
2: with a series resistor included before the attenuator, a R1/R2 voltage divider (with differing values depending on the position of the attenuator wiper)
3: an RC low-pass filter formed with whatever R at wiper is and the output cable core-screen capacitance.

From what you are saying though I think I can't visualise the circuit in blocks/nodes like this and need to take into account the whole combined resistance...do I just look at series resistance or would R2 come into play? Thanks again.
Pringe
Regular
Posts: 61
Joined: Wed Sep 01, 2004 12:00 am

Re: Passive Monitor Pre-Attenuator

Postby James Perrett » Fri Apr 03, 2020 9:16 pm

Might be worth having a play with some simulation software. LTSpice is one that is free (although I've not used it).
User avatar
James Perrett
Moderator
Posts: 9201
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page