Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[klaus]

From an inquiry by a poster whose name was redacted:

Originally Posted: Fri, 15 September 2006

Much attention has been given to finding the optimum impedance match between a microphone and preamp, especially with the rise of impedance-selectable pres.

Working with a wide range of both mics and pres, I am wondering how one determines - from a technical perspective - which pre would best suit a specific microphone?

The base research that I have done yields guidelines, such as: "the input impedance of the pre should be 10x the source impedance of the mic..."
 
Sonic factors notwithstanding, what are the "rules" for impedance matching?
 

[maarvold]

Although I'm not an expert, I will tell you one thing Steve Hogan told me.  Steve redesigned much of the current line of Jensen Transformers and works on all of my gear.  Klaus, if you feel it's necessary, I will see if I can get Steve to post this same information (although certainly in a more accurate and concise form).  Steve said:
Any mic preamp that uses a Jensen [input] Transformer, or a similarly-purposed microphone input transformer, will exhibit high frequency ringing when using any of the modern Neumann [transformerless] mics--the ones that have a 50 Ohm output impedance.  That would include TLM170, TLM193, KM140, KM184, M149, TLM103, TLM193, TLM49, TLM67 and TLM50.  These microphones must be loaded differently to avoid the ringing. 
I have had Steve add a Lo-Z switch to all of my preamps that fit this description (14 channels). 

[John Roberts {JR}]

The general rule of thumb for modern interchangeable, general purpose mics and preamps, is for an interface termination, 10x the mic's source impedance to deliver a "bridging" termination (for mostly voltage transfer). Mic source impedance is not a single specified number. I generally use 150-200 Ohms nominal for mic source impedance, and 1.5-2k for preamp input terminations.

Straying from a bridging termination can make differences in level, and perhaps subtle frequency response and/or reactive interactions (with dynamic mics). Perhaps of interest to esoteric preamp makers who want to sound "different". 

Mic designers are (or should be) generally aware of the nominal terminations and design to work properly in that context, or not. Terminating a mic differently than the mic designer's intended nominal is operating off the reservation, and your mileage may vary.

JR
 

[Jim Williams]

Mic's output impedance are all over the map, some use iron, some don't and those tend to be lower like 47 ohms for 414TL's and 22 ohms for Schoeps.

Input impedances are also all over the map, some with 1200 ohm iron, some with 2200 ohm loaded transistors, some at 4 k or above. Gordons are at what, 1 meg ohms? So far no one has complained about those, yet.

To optimize input impedance for a mic you will need to set up some test gear. Square waves are used to see tilts in response and ringing. Then you add RC elements to stop it, usually imperically. Then you will need to add switches for every mic preamp to add/create all the permutations of loading, it better be a rotary with lots of steps. Then create a chart to show ideal loading positions for every mic on every mic preamp.

Or, just load lightly, press record and call it a day...

[maarvold]

Actually, after speaking with Steve Hogan, I found out I should have said "high frequency overshoot" instead of "ringing"... but it still equates to 'tizzy', accentuated top end, even if only slightly so.  Re: Gordon pre's, their default input impedance position is 1 MegOhm, but they have a lower setting; Grant Carpenter's position is that the wants the mics to be 'unaware' that they are 'looking at' any load.  But it is also my understanding that, because they are not transformer coupled, the overshoot problem with the Neumann 50 Ohm output impedance mics does not occur.  For the record, I will say that the Gordons are very good.  But I will also say that I currently generally don't use my 6 channels of Gordon's any more; rather I use my [990-ish] highly modified Precision Analog (made of 'unobtanium'), Jensen input transformer-coupled (in Lo-Z mode) with my modified TLM170's and TLM193's and am supremely happy.  FWIW. 

[Steve Hogan]

Because this link was sent to me by an inquiring client, and it seems that I might be able to shed further light on the impedance issues regarding microphones, I will offer some basics.

First the theory (which varies in the real world, as I will explain later):

Microphones produce relatively low-level signals that must be amplified in order to be brought up to useable levels. The game plan is to increase the level of the desired signal and minimize any electrical noise coming from the mic.  We also want to interface our microphone to the preamplifier in a way that maintains the best fidelity.   

Exceptions to this generalization abound because some recordists intentionally modify the loading to certain microphones in order to change the way they sound.  That's totally OK in the name of artistic license.  The following discussion, however, relates to those who desire to capture the music with the most faithful fidelity possible without intentional coloration.

   First some definitions:

Load impedance is the complex magnitude and phase presented to the microphone by the preamp circuitry. It can be thought of as a complex impedance placed in parallel with a preamp input with theoretically infinitely high input impedance.  Infinitely high impedance when placed across a signal source will not cause any drop in signal level.

This impedance is not just a resistive load, but contains capacitive and inductive reactance which makes the impedance vary with frequency.  In general, for example, a transformer-based preamp input will have an impedance that is quite flat in the midrange but becomes lower at both low and high frequencies due to the transformer inductance at low frequencies and the input capacitance at high frequencies. Usually the midrange impedance is on the order of 1000 to 2000 Ohms. When you switch in the resistive pad to such a preamp, then the microphone will see a much more resistive load because the pad resistors interface between the microphone and the input transformer so the mic sees the resistors in the pad as the load.

Microphone output source impedance is a little bit more difficult concept to initially grasp. One may think of the microphone's output as a Zero Ohm signal generator in series with an impedance. This generator can be thought of as having infinitely low output (source) impedance -- one that can be loaded with 1 milliOhm and still not drop in voltage level. In series with that zero Ohm generator is a complex impedance (having magnitude and phase) which is the output source impedance of the mic.

 In some very low output impedance transformerless microphone designs, this equivalent series impedance magnitude may be as low as 12 Ohms.  Or it may be greater than 600 Ohms in some ribbon mics.  There are a lot of transformerless mics that have output impedances in the 30 to 50 Ohm range, and most microphones that incorporate an output transformer have output source impedances in the 150 to 250 Ohm range. Depending on the microphone, this output source impedance may be mostly resistive, somewhat capacitive and maybe inductive, depending on the microphone's design.

  Since the microphone's output source impedance lies between the theoretically perfect microphone internal signal generator and the load, there will be some loss across the source impedance caused by the input impedance of the preamp. The source impedance of the microphone and the load impedance of the mic preamp form a voltage divider between the theoretically zero Ohm source microphone and the theoretically infinitely high input impedance (Z) mic preamp.

Matched Impedances:
Although the term "impedance matching" is incorrectly and loosely used to describe any adjustment to the impedances of the microphone/preamp interface, the term "matched impedance" has a very specific meaning which applies to all electronics including mic to preamp interfacing.

The source is considered to be "matched" when loaded by an impedance exactly equal to its output source impedance.  It turns out  that when the load Z and the source Z are exactly the same magnitude, the maximum POWER is transferred from the source device to the load.   The voltage level between the zero Ohm generator and the infinitely high impedance load drops exactly 1/2 or -6dB when the source and load are equal.  Consider the case of a CB radio (Oh, am I old) with a 50 Ohm output Z feeding an antenna with a 50 Ohm input Z in order to transfer maximum power from the transmitter to the antenna.

Because we do not need to transfer power to the mic preamp, but instead we want to maximize the signal Voltage getting to the preamp from the microphone, we generally use the following method to interface mic to preamp:

Bridging Load:
 As a rule of thumb, an output source is considered to have a "bridging load" or to be "bridged" when the load impedance is greater than or equal to ten times the source impedance.  Thus a mic preamp with a 1500 Ohm input impedance will present a bridging load to a microphone with 150 Ohm output source Z.  Such a load will cause the signal voltage level to drop -0.83dB instead of the -6dB that occurs with a matching load.  A bridging load improves the signal level 5.17 dB over the matching configuration. A 5 dB improvement in signal-to-noise ratio is definitely worthwhile.


   If all this seems hopelessly complicated, be of good cheer! One does not have to understand all the engineering details to understand how it all interacts.  If you understand the principles of how the microphone and its output source impedance can interact with a preamp and its complex input impedance, you may be able to make decisions on interfacing which improve your recordings.

Since this post is already rediculously long, I will continue in my next post, unless you have had enough already.

[klaus]

Thanks for the thorough reply.

What needs further mentioning and explanation is the fact that mis-matching impedances between mic and pre increases distortion.

Another aspect: electronically balanced, rather than component or transformer balanced, mic output configurations, regardless of their theoretically superior linearity, are more susceptible to noise.

[Inlinesix]

First I am sorry for my bad English, i dont speak english.

I am designing some mic pre amps and still have many doubts about this.

The transformers that I have to use the microphone preamplifier input has the following characteristics:



- DC resistance in the primary = 200 Ohms

- DC resistance in the secondary = 950 Ohms

- Value = 1: 6

- 20dBm

- Electrostatic shielding


In some tests I did with an attenuator



This association of the resistance R3 with the DC resistance of the transformer primary, thus giving a total resistance of 85.7 ohms

Analyzing a Jensen scheme, I saw that the PAD resistor is 169 Ohms and the DC resistance of the primary trnasformador JT-115K-E is 19.7 ohms, thus having an overall resistance of 17.64 Ohms.

I was wondering if I ignore this DC resistance of the transformer primaries or should take this into account when calculating my circuit.


In my project I have two gain stages and control the gain level with a potentiometer between these two stages, so I do not need PAD, but the tests I did, the PAD is necessary to leave the more linear frequency curve.

These preamps are for SM57 microphones, and I need you guys help me choose the correct values of these resistors to the PAD and that the attenuation is the lowest possible.

tanks !!

[polypals]

Hello George,

My sincere compliments to jump into the deep.

May I suggest for you to get acquainted with electronics before you go on designing circuitry.
This is not meant to discourage you but to open a whole new world of possibilities to you.

The dc resistance of a transformer is not of major importance for the application in audio.

[polypals]

Thanks for the thorough reply.

What needs further mentioning and explanation is the fact that mis-matching impedances between mic and pre increases distortion.

Another aspect: electronically balanced, rather than component or transformer balanced, mic output configurations, regardless of their theoretically superior linearity, are more susceptible to noise.

First of all there is no matching of impedances, there is simply the matter of optimum voltage transfer to the input of the preamp.
This is achieved by keeping the source impedance at least ten times lower than the load from the pre amp.
Please explain how distortion can be influencend as long as the rule for optimum output transfer is followed.

The beauty of the European system lies in the universal application.
Regardless of the source as long as its impedance is considerably lower than the load of the pre amp the match is good.

Designs with an input transformer should be suitably damped to avoid overshoot or ringing as a result of ultra low impedance sources.
I am stating European system because it was a deviation from the US maximum power transfer system coming from the telephone industry.
Older American mikes have a 600 Ohm source impedance to give maximum power to the mixer or network.

[Inlinesix]

Hello George,

My sincere compliments to jump into the deep.

May I suggest for you to get acquainted with electronics before you go on designing circuitry.
This is not meant to discourage you but to open a whole new world of possibilities to you.

The dc resistance of a transformer is not of major importance for the application in audio.

I know I have to study more electronic circuits. Whenever I learn something new I think I'm just an ignorant.

I always learned everything myself, I never had a teacher.

You said dc resistance is not so influences the audio, but it is not what it seems when it comes to OT for guitar amplifiers.

In such cases I noticed much difference. So I asked the question.

I understand that in low-level signals that does not make a significant difference.


tanks

[klaus]

David Satz posted on the Neumann forum a really clear and informative summary of microphone impedance strapping found through the 1980s on U87, KM84 etc.

There are actually three possible settings for the output impedance of the KM 84 and most other Neumann "fet 80" microphones: 50, 150 or 200 Ohms. In Europe, 200 Ohms was the de facto norm, with 50 Ohms as an alternative if you needed to drive long microphone cables; the lower source impedance helps avoid high-frequency losses and other problems that can occur due to the distributed capacitance of the cable. The 50 Ohm setting also has 6 dB lower output levels for any given SPL, which can help prevent overload in the input stage of a marginal preamp, mixer or recorder.

Most fet 80 microphones sold via Gotham Audio Corporation (the long-time U.S. distributor) were set to the third option, 150 Ohms, although many have been rewired in the years since then. The two secondary windings of the output transformer were paralleled as in the 50 Ohm setting, and a pair of ca. 50-Ohm series resistors, normally bypassed with solder bridges, were "un-bypassed" by opening (unsoldering) the bridges.

That setting has the lowest output levels of the three--more like those of a dynamic microphone. When the KM 84 was introduced in the 1960s, some U.S. broadcast stations still had old mixing desks with microphone inputs designed for dynamic microphones. Mr. Temmer, the head of Gotham Audio, was concerned about overloading those inputs, and also about some high-frequency response variation that occurs in certain preamps when the output impedance of the microphone isn't what the preamp's input circuit was designed for. He didn't want Neumann microphones to be blamed for any possible adverse interactions with other equipment. As a result, if you were a U.S. customer, you had to special order the microphones if you wanted them set for 50 or 200 Ohms--and then (as I experienced myself) someone from Gotham would call to quiz you about what kind of equipment you intended to use the microphones with.

But nowadays, if you're using a good modern preamp or recorder, it should be perfectly safe to use the 50 or 200 Ohm settings. Either one is preferable to 150 Ohms for avoiding interference in the microphone cable; if resistive pads are needed to prevent overload, they should always be placed immediately before the part of the circuit that is threatened with overload, rather than at the output of the microphone. 

[Hermetech Mastering]

Just got a European/Tuchel pair of KM84s from 1977, and am presuming they have the 200 ohm outputs. Question is, regarding the mic pre, (a Chandler TG2 which has switchable input impedance), which setting will give me the flattest/best sound, 1200 ohm (standard) or 300 ohm? Have just been using 1200 so far and liking the results, but am about to ship the mics off to Andreas Grosser in Germany for a service. The 300 ohm setting is about 5dB louder on the same source, but am worried it won't sound as good/have higher noise/be less flat etc.

Theoretically, which setting would be best for these mics?

Thanks in advance,

Gregg

[klaus]

The old rule of 1:5 still applies in most if not all cases: with  200Ω output strapping, the mic pre's input should be at least 1000 ohms.

As to output strapping on the KM84: I still like the sound of secondaries of the output transformers best in series (200Ω) rather than parallel (50Ω), with in-line pads removed.

[Hermetech Mastering]

Thanks Klaus!

How easy it to remove the inline pads on a KM84?