Mic Pre input impedence?

[Svart]

over the course of modifying my console I came upon two resistors, each connected from the base of each of the input transistor (2n4403 here) to ground. I assume they serve no other purpose than setting the input impedence? they are 3.3k.

I recall reading somewhere on this forum about the affects of input impedence on microphone responce and I also recall from memory that the value is lower than 3.3k?

well, just for shits and grins I removed these resistors on a few channels and replaced them with 600R. I placed a few different mics on these channels as i normally would and tracked various drums and sounds.

A very unscientific approach it is, but one to satisfy my curiousity just the same...

well I have to say that this IMPROVED what was coming out of my mics quite a bit. even the unusable sm57 can be used now. it took my other mics and made the audio more even sounding, allowing me to use LESS EQ (as some have read my other posts, you know i was having to EQ the hell out of some things)

anyone else have experience with this? :thumb:

 

[NewYorkDave]

Without seeing the circuit of your mic preamp I can't say for sure, but those resistors probably set the base-emitter voltage as well as the input impedance. Also, since the base draws current, the actual input impedance will be lower than the values of the resistors. If you replaced them with 600 ohm resistors, your differential input impedance is somewhat lower than 1200 ohms, which is too low for some mics. 1200 to 2000 ohms, and sometimes a bit higher than that, is the "general purpose" input impedance range for most modern mic preamps.

 

[NewYorkDave]

You're making a big assumption. Changing those resistors MAY have little or no effect on the collector current at all, depending on the circuit.

 

[dale116dot7]

In most cases, the base current will be very, very small. If it's a 'normal' transformerless mic pre, the collector current will be somewhere between 250uA and 2mA, and the base current will be somewhere between 1/300 and 1/600 of that, so very little shift in DC bias. Try changing the circuit around - put the 3.3k's back and go across them with about 1k.

 

[BYacey]

Try going higher. The lower the resistors on the bases the more self noise created versus signal. You want the transducer to be the predominant noise source. I would be inclined to try 10K to 100K, although you will probably need to juggle the collector or emitter resistors to bias the transistor in it's linear range. 1 to 2 ma is a good bet to keep noise down. Low value base resistors also tend to load the output on dynamic and ribbon mics.

 

[Svart]

actually, the mics that improved in sound quality were the dynamic ones. this resistor i speak of went from base to ground on both the input signals. there are no other resistors from the base of either BJT to anywhere, just the low pass cap (RF filter) to ground also.

I understand what you are saying however my reasoning for doing this is very unscientific, rather almost an educated guess due to my lack of time and resources and irresistable urge to give it a try :green: . Going UP un numberical value is lessening the load that the audio signal sees, it is also allowing the base to drift upwards in bias. going down numberically is increasing the load and bringing the base to a lower bias point, however the advantage to my mind is that small signals too weak to sustain amplitude are also pulled down more readily. the effect i saw was a little less input from the mic but more linearity was realized. I don't normally work with BJTs so I really don't have a place to speak here, I'd pick a FET or IGBT anyday over a BJT but again this was just a "stick it in and see what happens" trial which actually worked in my favor. it may not work for others.

I noticed no real change in some mics, while others improved in sound. I didn't find that the audio became worse in any of my testing.

Noise doesn't seem to be an issue, nor do i notice it. I'm not a purist so a small amount of noise increase is negligable for increased fidelity.

I'm just reporting opinions for the common cause, but I welcome all of your opinions and comments, so thank you and keep them coming.

:thumb:

 

[Svart]

any more opinions? I'm rather interested in learning more about this.

:guinness:

 

[rafafredd]

There are some vintage units, like the old neve stuff that has 1k2 input impedance.

 

[babyhead]

1200 ohms, which is too low for some mics. 1200 to 2000 ohms, and sometimes a bit higher than that, is the "general purpose" input impedance range for most modern mic preamps.

If this is the case, why are all of the Jensen mic input transformers 150:x ?

What am I missing?

:guinness:

 

[NewYorkDave]

The "150" refers to the source impedance of the mic. The mic should be working into a load impedance ten times (or greater) its source impedance; that's why 1200-2000 ohms is a fairly typical input impedance for a mic preamp.

A 150:15K mic input transformer will reflect an impedance of 15K to its secondary when its primary is terminated by the 150-ohm source impedance of the mic. The secondary should be bridged by an impedance of 150K or greater to avoid loading the mic.

 

[PRR]

> any more opinions? I'm rather interested in learning more about this

Put the preamp back the way it was. The design load for mikes IS about 2K (1K, 3K, whatever). And it is a lot easier to reduce the load, when desired, than it is to increase it if needed.

Put a 5K audio-taper pot and 47Ω resistor in a box with XLR ends. Shunt this across your 3K input, and you can try loadings from 2K to 47Ω.

But all you are really doing is EQ-ing the mike, without any real control of what you get. There is slight synergy because mike response bumps may also be mike impedance bumps, so heavy loading may reduce ringing more than the main sound. But a proper EQ downstream should do the same.

 

[dale116dot7]

And the DC shift in a standard transformerless mic amp is almost insignificant - and the the op-amp stage would null out most of the DC shift anyways. Just keep them symmetrical.

 

[dale116dot7]

Oh yes, Summit had a preamp module with a pot for mic loading. I found interesting things with dynamics as well - condensors were basically not affected, especially the transformerless ones.

 

[Svart]

found interesting things with dynamics as well - condensors were basically not affected, especially the transformerless ones.

Yes, I find this is the case entirely. however I am also using an oktava 219(has transformer) for snare that sounds MUCH better with 600R than with the 3.3k load. the overheads, oktava mc012s, are the same, an SM57(has transformer as you know) and m201(don't know about transformer..) i use also sound much better now.

I like what you suggested PRR, I'd rather make it adjustable too.

one more observation.. it doesn't *sound* like EQ change, it sounds more even and natural than EQ changes. it sounds more even, more like frequency based compression(i know it's not, just tying to convey the feeling).

:guinness:

 

[babyhead]

Thank you NYDave for the clarification. This whole electricity thing can be boggling. :?