input impedance question

Started by snk, September 08, 2020, 01:29:17 PM

Previous topic - Next topic

snk

Hello,

I am building Sabrotone Kompressor and parametric eq.

I have found a thread here, saying that the input impedance of the eq circuit was was too low, which led me to ask myself a question : I am not a guitar player, and I plan to use the compressor and the eq with studio gear (the signal chain would probably be : drum machine > eq/compressor > mixer, or analog synth > fx pedal  > eq/compressor > rack fx > mixer), so which would be the best input impedance for these circuits ?
- Should I go for a 1M input impedance (and use a 1M resistor at the input of both circuit) ?
- Should I aim for 600 ohms (somewhat studio gear input impedance standard)?
- Should I keep the circuits as they were designed and see what happens?

Thank you!

antonis

>- Should I keep the circuits as they were designed and see what happens?<

That .. :icon_wink:
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

snk

Thank you, Antonis.

It is the discussion about the eq's very low input impedance which got me hesitating.


antonis

As you might know, resistance & noise aren't bossom buddies.. :icon_wink:

IMHO, you may raise R10 value up to 470k - 510k without any noticeable (by ear) noise increase..
(100k bias resistors are used for bipolar input op-amps both for low noise and input current offset..)
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

snk

Interesting, thank you : This is something I overlook a bit too often :)

So, in this circuit, picking a low value (100k-ish) for R10 would keep it low noise, but chosing a 1M-ish value would increase the impedance and make it more suitable in a guitar effect chain, right?

antonis

"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

snk

Thank you Adonis, that's very helpful.
I might use a socket and try with 100-330K, and then with 1M and see how it sounds.
(I may even use a switch if both values find their use in my studio)

snk

Funnily, I just found this post from you :)
Quote from: antonis on November 29, 2019, 07:43:25 AM
I've built many of single supply RG's parametric EQs variations with absolutely on issue..!!!
(just use the reasonably lowest resistor values for bias & gain and good op-amp use practice..)

antonis

#8
 :icon_biggrin: :icon_lol: :icon_biggrin:

When preceding circuit/effect output impedance is in the range of few tens of Ohms, 100k or 1M input impedance make no practical difference..


P.S.
>I might use a socket and try with 100-330K, and then with 1M and see how it sounds.
(I may even use a switch if both values find their use in my studio)
<

As long as you use an FET input op-amp, there shoudn't be any bias issue..
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

MaxPower

Adonis? Nudge, nudge say no more!

As long as no one calls me Maxipad...
What lies behind us and what lies before us are tiny matters, compared to what lies within us - Emerson

PRR

> - Should I go for a 1M input impedance
> - Should I aim for 600 ohms

600 ohms is 1938.

Modern studio interface is <50r source and >10k load. So just leave it at 100k.
  • SUPPORTER


duck_arse

I feel sick.

bartimaeus



i was looking over the schematics for the pedals you're making, and i'm curious why the EQ has both a 1M resistor and a 100k resistor on the input? so isn't the input impedance actually ~90k since they're in parallel?

my guess is that the 1M is there to prevent any switch pop, and the designer didn't care that it lowered the impedance slightly compared to the original 100k. but maybe i'm missing something, since the 100k is also part of an active highpass filter?

antonis

You don't miss anything..  :icon_wink:
(except for the main purpose of 100k resistor which is for IC2 bias..)

IMHO, C8 value could be lowered down to 100nF, for a cut-off frequency of 16Hz..
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

PRR

> 100k is also part of an active highpass filter?

Is it?? Compute the cut-off.

On my thumbs, looks like 0.16Hz. It isn't blocking anything except stray DC.

Why so low? Unlike instrument amps with few stages (and interstage networks), this looks like Studio Gear. How many stages from microphone through effects, mixer, compressor, recorder, playback, mastering room, disk lathe, to your Grammy Award? Say a hundred. If each stage is 0.1dB down at 20Hz, a hundred stages is 10dB down at 20Hz, likely 5dB down at 45Hz, a significant shave of the musical bass.

There ARE other ways to do this. Sometimes it comes down to what we did last time, what your boss wants to see, what parts are in over-stock, or just random chance.
  • SUPPORTER