what is input impedance?

i have a preamp with a knob from 3k down to 150. this knob is on the latest model of the art preamp. what is its function and how would i use it properly? it seems to shift eq or phase...so whats default or off....
thank you

ill keep it on 3k then. thx

FWIW, It has been my personal experience to have learned that Mr Drury seems to know far more about the nuanced difference between canned beer brands than alternating current signals.


Here is an actual article written about input impedance on microphone preamps that was written by someone who knows what he is speaking about:

http://rupertneve.com/com...s/impedance-and-noise/

When I was in basic electronics training (in the Air Force) it was hammered home about matching "Z" for maximum power transfer (volume) as Herb notes.  My old ARRL (Ham Radio) handbook also goes into impedence matching.

Even for speakers, there is no impedance matching.
In the world of RF engineering, signals are extremely small and power is precious. You don't want to waste anything, not even 10%. Most circuits only have to handle a very limited bandwidth. Look at your FM radio. The range is about 90 - 110 MHz. The center is 100 MHz. So 20 MHz bandwidth over 100 MHz center frequency means 20% fractional bandwidth. Compare that to audio: 20 - 20,000 Hz (3 orders of magnitude), i.e. 19,980 Hz bandwidth over 10,010 Hz center frequency - almost 200% fractional bandwidth. So, the concept for audio is: make sure the frequency response if flat. Wasting energy is, thus, the accepted solution. The efficiency of loudspeakers, for example, is incredibly low.

In RF engineering the goal is to use the signal power most efficiently. Transformers and other lossless impedance converters (filters) are easily built since higher frequencies mean smaller inductors and transformers. Also, transmission lines can be used as narrowband transformers. This works because you can make a cable a quarter of a wavelength. (100 MHz has a wavelength of 3 m (10 feet) in vacuum. In a coaxial cable, wave propagate more slowly so that 1 wavelength is only 2 m (about 7 feet). For audio, you would be looking at miles or hundreds of miles. 

In the world of audio, the standard approach is to make the source impedance low and the load impedance high. This approach ensures that the source 'enforces' the frequency response. Low output impedances are easily achieved with transistors (<100 Ohms), and so are fairly high input impedances (> 2000 Ohms). This is not really efficient, but extremely predictable. 
There are (at least) three exceptions: 1. Ribbon mics. The output impedance is extremely low, and so is the signal level. Here, using a transformer is very beneficial. A 1:20 transformer will give you 20 times the output voltage and 20^2=400 times the output impedance. This way, you get into the right range. 2. Condenser mics. The output impedance is so high that it would never work without an active (FET) impedance converter. 4. Electric guitars. The output impedance in really high, and it is inductive. If the input of the amplifier had a low impedance, there would not be a lot of signal left. Another point is that the capacitance of the cable and the inductance of the pickup form a resonant circuit the gives you the desired frequency response that boosts the mids. This effect would go away with low input impedance.

Let's talk about power amplifiers and speakers: If you move the membrane of a speaker, you induce a voltage in the coil. If the terminals of the speaker are not connected to anything, you can easily move the membrane. If you short circuit the terminals, a large current will be driven through the coil. This current creates a forces that will make it harder to move the membrane. Thus, if you amplifier has a higher output impedance and if the cables a thin, speakers can reasonate according to what the mechanical design predicts. If the output impedance of the amplifier is low and the cables have a low resistance, the speakers have to follow the electric signal.

Back to the mic preamp: Choose the high impedance setting unless the low impedance setting gives you a more desirable sound.

Wilko

Great stuff, Jeff, Wilko & Mike.

This old dog is still learning new tricks.

I would suggest that if one does not have a monitor system and room that provides authority and accuracy down to 20Hz that one will never hear the phase smear and distortion occurring in the lows and mids as the mic is loaded down with too low and impedance bridge.

The higher impedance seems bright because the signal is not being swamped with freakish bass anomalies.

Frankly speaking the ISA series is a bright sounding preamp... that is it's nature. It's said to be loosely based on a Rupert Neve design... Mr Neve doesn't provide multi impedance choices on his new gear.

It's very easy to turn the hi frequencies down with EQ or mic placement but it is impossible to restore the quality of the bass response once it has been turned into mud.

Restoration tools such as the Transient Shaper 64 will not help... it will only make the mud different.

I think the general advice to experiment freely is great advice... but if one is going to purposefully handicap their microphones frequency response it might be best to only do so if one has the ability to hear what is really changing with some level of accuracy.

If you ignore this in a mix... then the result will be mystery meat down in the low end when you finally hear playback on a big bad system.

I take some small pride in the fact that my better mixes can played on a huge PA and sound tight and solid. I like it when the bass can be perceived as a dangerous percussive weapon.

A lot of people spend years wondering why their bass seems mushy...

all the very best,
mike

True, unless you change the nonlinear behavior. The load impedance will have an impact on how and how much the ribbon will move and deform. Thus, you may notice differences in the nonlinear response which you cannot model so easily.
But, I frankly don't know if any audible change in the nonlinear response is expected.
 
Wilko

Hi AT,
 Mr. Rupert Neve states that in his learned opinion all mics perform best when looking into a high impedance... he explains the reason in easy to understand detail in the two papers I linked to above.

 That is why he designs his current preamps with an extraordinary high 10kOhm input and then simply provides adequate gain on the preamp.

 There are 2 ways to look at it.

1) the mic designers are probably dong their best job at making the mic sound great when it looks into today's *convention* of 1.5kOhm.

2) Rupert Neve has discovered for himself that all of them sound even more accurate and dynamic when they see even more impedance than that.

 Mr Neve explains how we are on a journey or transition from the days of "matching" impedance to fully realizing the practice of "bridging" impedance.

 He explains that back when he started the practice of bridging in his designs that he used a 1:5 bridge.

 Now a days a 1:10 bridge is the common convention.

 Mr Neve demonstrates that he thinks a 1:1000 bridge works fantastic.

 Now, one must remember that this is not necessarily saying that the 10kOhm input gives you the exact sound you want... but it does give you the widest flattest frequency range and the most accurately captured transients and dynamics particularly in the (what I consider critical) bass and low mids. The idea is if you capture something that way that you may sculpt it easily in post and have fantastic transients preserved until you wish to do your thing with them.

 Now this is not to discourage someone from tailoring EQ with microphone choice and placement... and in fact many EQ response characteristics in a microphone design are the result of acoustical phenomena at the capsule and in the resonant chamber formed by the grill assembly... so that is one step removed from the electronics we are speaking of... but it is connected as well. It's all just math being turned into art.

 So again, I all am for free experimentation and anarchy in the art studio but I am warning people that in this instance the effect they think they are enjoying may come at the expense of some gross loss of detail and authority in the bass and lower mids and I am suggesting that once that character is lost and printed to the disk... you will never really get it back no matter how much software ones buys or time they spend efforting a repair.

 Most ribbon mics these days ARE made for 1500 ohm inputs... or higher.

 A great example of the shift in design philosophy is the AEA The Ribbon Pre offered by Wes Dooley the famous RCA ribbon mic expert who owns AEA. The input on his, made just for Ribbon Mics preamp is 18kOhm. Wes thinks you'll really like his 270ohm AEA R84 mic when it sees 18,000 ohms. Now that's a "bridge".

 So, the engineering need to provide a good match/bridge for an ancient ribbon mic is usually only necessary if indeed one has a ancient ribbon mic.

 The rest of the time it might be considered a good idea to get a solid signal and then destroy it in post at your leisure.

 But that's not to say that some artistic anarchy can't be both fun and useful!!!

 all the very best,
mike




editing spelling

Hey Mike, I read the Neve stuff.  Like most of what he sez, it makes sense (when I can follow it).  Good stuff and thanks.  I like the way the Portico preamps sound, so who's to argue w/ Mr. Neve?

And yea, Bill, sometimes doing things wrong sounds right (ain't that an old R&B song?).  But you gotta know the rules to break 'em.  I do know that I used the lower impedence settings at times wanting a less bright sound.  But I don't use the ISA as much since I got the Portico.
@

I agree with all of you.... it's all art... get crazy!!!

I just like listening to tight bass tones a whole bunch.

all the best,
mike