| |
G8MNY > TECH 20.12.24 11:03z 95 Lines 4142 Bytes #179 (0) @ WW
BID : 22260_GB7CIP
Read: GUEST
Subj: Ultra Low Noise Mic Pre-amp
Path: SR1BSZ<IW0QNL<IZ3LSV<DB0ERF<DK0WUE<DK0WUE<K7EK<KA1VSC<GB7CIP
Sent: 241220/1056Z @:GB7CIP.#32.GBR.EURO #:22260 [Caterham Surrey GBR]
From: G8MNY@GB7CIP.#32.GBR.EURO
To : TECH@WW
By G8MNY (Updated Jun 21)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)
Years ago I made a stereo one of these, for recording the morning chorus, with
low Z moving coil mics (no good for amplified Electrets!), after reading about
the then "new" Ferranti ultra low noise wide band pre-amp IC the ZN459, A Mic
preamp, just one of many specialist ultra low noise applications for this IC.
De-coupling 6
ù
ÚÄ180KÄÄÁÄ180KÄÄÄ¿ The IC amp has 2 stages x100 & x10,
³ ³ these voltage gains are fixed,
³ Supply 3 ³ but the current gains & in & out Zs
³ ³\ ù ³ can be modified.
³ ³ \ ³\³ ³
³ -³X \ +³ \ ³ .----. 1= Dil pin 1,6,8
Input 4ùÄÁÄÄ´100>ÄÄÄ´X \ ³ / 3 4 \ 2= Dil pin 7
³ / ³10 >ÄÁÂÄù2 Output ( 2 5 ) 3= Dil pin 2
³ / ÚÄ´ / ³ \ 1 6 / 4= Dil pin 3
³/³ ³-³ /³ 10R '¿ Ú' 5= Dil pin 4
³ ³ ³/ ³ ³ ÀÄÄÙ 6= Dil pin 5
Input ³ ÃÄÄÄÄ)ÄÄÄÙ Output TO71 8 pin Dil conversion
Earth 5ùÄÄÄÄÄÄÙ 1R ÀÄÄÄÄÄù1 Earth ZN459CT ZN459CP
_³_
///
My one was the tiny TO71 metal can ZN459CT version. so 6 wires only onto the a
PCB. The plastic 8 pin dil version has different pin outs see above!
IC SPECIFICATION
Freq Response 0Hz-15MHz @-3dB
Gain 1000x 60dB +/-1dB
Noise Equal to a 40R in series (800pV/ûHz)
Current 3mA @ 5V (+9V via a 1K2), 6V Max
Supply Rej 42.5dB
THD 0.15% @ 1V RMS 10kHz
Output Z 75 Ohm
Input Z 7k with a de-coupling C to ground, 170R with a 180k bypassed!
MIC PREAMP CIRCUIT
ÚÄÄÄÂÄÄÂÄÄ1k2ÄÄÄ<+9V
3 ³ ³ +2u _
MicÄ¿ _ 6T 20u+ +³ ³ 2ÄÄÄÄÄÄÂÄ220RÄ´ÃÄ(_()ÄÄ¿
³ 200R<Ä(_()Ä((()ÄÄÂÄ´ÃÄÄ4 === ³ u22 ³ ³ ³ 10K
³WW POT ³ ³ FBead ³ 100u³ ³ 6ÄÂÄ´ÃÄÙ ³ ³ Log<ÄÄ Pre-amp
ÀÄÄÄÁÄÄÄÙ ³ ³ ³u1³ ³ ³ ³ GAIN output
³ ===33n ³ === 100K ³ ÀÄÄÄÁÄÄÄÄ
³ ³ 5 ³ ³ 1 ³ ³
ÀÄÄÄÄÄÄÄÄÁÄÄÄÄÄÄÄÄÄÁÄÄÄÁÄÄÁÄÄÄÄÁÄÁÄÄÄÄÄÄÄÄÄÄÄÄÄÄÁÄ<0V
Because there is very high gain, >1mV input is all that is needed to cause
onset of clipping @ 1V O/P. So I used a 200R WW pot on the front panel, in
front of each pre-amp, to preset the max input level as required.
A high degree of RF input filtering is provided with a series 6 turns on a
ferrite bead & a 33nF to ground, but not enough to degrade the AF on a low Z
source.
The input 20uF must be very low noise, & might need selecting on test (200R Pot
at min).
Much lower input impedance is achieved by AC bypassing the internal 180k output
to de-couple pin with a u22. So the normal 7k input is now in // with 180k/1000
due to increased current NFB.
Also on my ICs a 100k from de-coupling pin to ground, helped improve the bias
to give a better symmetrical output clipping.
The high gain broadband device also needs good power de-coupling of a u1 & a
100uF across the power rails right under the IC for stability. I used 9V
battery power via the 1k2 gives about 5V on the IC.
To stop any output nasties the amp output sees a minimum load of 220R on the
PCB in series with the front panel 10k Log Gain pot. This can then be
resistively mixed with other channels in the normal way.
FURTHER
I have not experimented with a series FET attenuator on the input fed
controlled with an AGC line derrived from the output, but I can see the
possible advantage, provided if it adds no noise.
Also see tech buls "Mic DJ Preamp", "Phantom Balanced Mic Preamp", "Phantom Bal
Mic Preamp+Gun Mic" & "DC sources for coax/AF".
Why Don't U send an interesting Bul?
73 De John, G8MNY @ GB7CIP
Read previous mail | Read next mail
| |