And why that value?
@Audioguru again

 

Yes, it is. It comes from the phone system and it was standardized based on the lines that ran on the poles (I think it was 18 AWG @ 5" separation). It became the standard impedance for audio circuits and the reference for signal levels.

 

620Ω for designing it is then. Thanks.

 

If you are interacting with other professional / broadcasting equipment, spring for a 604 ohm, 1% part. Often, pro gear inputs have a 604 ohm input resistor to GND or a 604 ohm output resistor. If your input/output resistor is 620 ohms, that is enough of an error to be visible on a VU meter. Back in another life, I kept a TV station's routing switcher alive and well. Lotsa tweaks all over the building so that everything switched seamlessly. Went through a lot of brand new video and audio equipment replacing 5% resistors with 1%.

ak

 

The bible for all things audio related is "Audio Encyclopedia" by Howard M. Tremaine.



page 1573

24.27 Why has 600 ohms been standardized for input and output impedances?

In past years, several different values of impedance, such as 150, 250, and 500 ohms, have been more or less standard for input and output circuits for sound equipment. During the early days of fm and television sound transmission, because of equipment available at the time, it was easier to meet the frequency requirements of FCC relative to equalization, using a 150-ohm impedance. Also, this was the first attempt to standardize on a given impedance for both input and output circuits. However, as the VU meter was based on a 600-ohm circuit, special input transformers were required for these meters, as discussed in Question 10.34.

Equipment manufactured with the last few years (1970s) use 600 ohms for both input and output impedance, except in the instance of microphone preamplifiers, which use 50 ohms (in older equipment this was 30 ohms). Some European microphones specify a load impedance of 200 ohms, others 1000 ohms. Power amplifiers employ both bridging and 600-ohm input impedances, with output impedances ranging from 4 ohms to 600 ohms, depending on the service requirements on the amplifier.

 

I agree on them using 600 ohms for a phone line that can be many miles (kms) long with quite a lot of parallel capacitance that cuts high frequencies too much if the cable impedance is higher.

 

https://www.fmsystems-inc.com/the-600-ohm-audio-standard-where-did-it-go/

 

I also have The Audio Cyclopedia.
Although it is very dated, it nevertheless is a treasure trove of audio information.
Highly recommended.

It provides the historical details and reasons for many things that we have today.

 

But consumer and professional audio are two different worlds. Audio levels are higher in professional. Signals are balanced (XLR) not unbalanced (RCA). Consumer input Z is about 10K ohms, not 600. Professional equipment is 600 ohm in 600 ohm out.
The "tape loop" is close or near professional levels.

 

620Ω║18KΩ=599.4Ω
If you need to tweak it test select the 18KΩ.
Paralleling resistor can yield almost any value, IMO you don't need specialty resistors.

 

600Ω output is common (balanced line with 300Ω on each side). Matching between the two output of the balanced line is more important that the absolute value - and 300Ω is an E24 value.
600Ω input impedance isn't - generally it's 10k (or 20k for balanced line), for microphones, line inputs to mixing desks and inputs to power amplifiers. (see Self D., Small Signal Audio Design, Elsevir 2010).
We're not terminating transmission lines here, we're ensuring op-amp stability into a potentially capacitive load (the cable) and ensuring maximum CMRR.
600Ω is important for measuring signal levels, because the measurement is derived from the original telephone standard, 0dBm = 1mW into 600Ω = 0.775V, but 0.775V is taken as 0dBm nowadays, regardless of what the load impedance is.

 

>620Ω║18KΩ=599.4Ω
If you need to tweak it test select the 18KΩ.

Or 1k2||1k2 = 600R

 

>620Ω║18KΩ=599.4Ω
If you need to tweak it test select the 18KΩ.

Or 1k2||1k2 = 600R

But it's a balanced line, so it's driven by TWO 300Ω resistors.

 

>But it's a balanced line, so it's driven by TWO 300Ω resistors.

My comment was referring to the inaccuracy of the 620||18k, rather than the application of the thread

 

My comment was referring to the inaccuracy of the 620||18k, rather than the application of the thread

In that case, I'd agree with 2 x 1.2k in parallel, because that gains the greatest benefit of resistors in parallel decreasing error due to tolerance, and also good for stockholding!
This website
http://sim.okawa-denshi.jp/en/rercal.php
will calculate optimal resistors in parallel, potential dividers and loads of other stuff.

 

I've written a program in C that does that with the E96 series

 

620Ω║18KΩ=599.4Ω
If you need to tweak it test select the 18KΩ.
Paralleling resistor can yield almost any value, IMO you don't need specialty resistors.

No matter how many digits a calculator spits out, that still is a +/-5% tolerance value.

Better to use one 604 ohm resistor at an input, and two 301 ohm resistors at a balanced output.

ak

 

No matter how many digits a calculator spits out, that still is a +/-5% tolerance value.

Better to use one 604 ohm resistor at an input, and two 301 ohm resistors at a balanced output.

ak

Professional audio inputs are generally 10k unbalanced or 20k balanced.

 

Professional audio inputs are generally 10k unbalanced or 20k balanced.

Agreed. The vast majority of professional audio equipment is designed with "bridged" impedances, with the source Z significantly lower than the load Z. The old 600-ohm impedance matched connections hasn't been a thing for many decades.

 

Agreed. The vast majority of professional audio equipment is designed with "bridged" impedances, with the source Z significantly lower than the load Z. The old 600-ohm impedance matched connections hasn't been a thing for many decades.

Yes. Why introduce a 6dB loss when you don't have to? And that means that it is no longer essential for output impedances to be exactly 600Ω.
The attenuation with 600Ω output and 20k input is 0.257dB, and with 620Ω it increases to 0.265dB, a difference of only 0.008dB. Much more important for the output impedances to be equal, because that really affects the CMRR.