Old: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Some Measurements of the Input-Capacitorless Preamp

Post by mediatechnology »

I've been tweaking the THAT1570-based fully-differential input-capacitorless preamp and thought I would post some measurements.

These FFTs were done at 48 kHz, 16 bit, 65,536 points, 20 sample average.
For the noise measurements the input was terminated in 150R and phantom power was applied.

Image
DC-Coupled Preamp THAT1570 Noise at 9dB Gain 150R Source Phantom On

Image
DC-Coupled Preamp THAT1570 Noise at 60dB Gain 150R Source Phantom On

Image
DC-Coupled Preamp THAT1570 THD at 60dB Gain 150R Source Red is Generator Monitor Green is Output

Note the rising slope of 1/f noise present in the Soundcard itself at the generator monitor input.

The circuit is still on an open protoboard.
I used JR's trick of driving the gain resistor switch shield connection in common mode to greatly improve low-level 60Hz (and harmonics) performance.

I've found an interesting trait exhibited by most likely all CFB preamps when inductive tip/ring capacitors are connected directly to the preamp inputs.
I had some 1nF Xicon films that were particularly inductive that were causing an internal Colpitts oscillation of the 1570.
I've seen the same effect in the 1510/INA217 and have always used base stopper resistors which double as phantom current limiters.
I found that by splitting the input series resistors I could greatly improve performance by isolating both the large tip/ring capacitor (1nF) and the capacitance of the phantom power diode protection bridge.
Naturally I ditched the Xicon and went with a "non-inductive" capacitor but I'm glad I found the issue using the "bad" inductive ones.
More on that later...
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by JR. »

Nice... I notice the plots stop short of 10kHz, is that related to sampling at 48 kHz?

The performance at low gain may be of interest for directly feeding an A/D convertor. I guess the inclination is to boost up then pad down again at the ADC but I could imagine flying the A/D up with the preamp while the boost/pad decision will be driven by best result. Note: It may not be crazy to convert both + and - stems from preamp, and perform the differential to SE conversion inside the digital domain.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by mediatechnology »

Nice... I notice the plots stop short of 10kHz, is that related to sampling at 48 kHz?
Thanks JR.

Visual Analyser stops short of performing FFTs beyond 20 kHz as the FFT size is increased (I assume to reduce CPU load.)
At 16Kpts it opens back up to ~22 kHz; at 32Kpts it is ~18 kHz.

I wished I had a Soundcard that had great 96 kHz SR.
The EMU0404's is pretty decent but for whatever reason its filters don't open up an octave as you would expect.
The Roland Quad capture does, but it has a very narrow level "sweet spot" where it has better THD than the EMU.
I've been thinking about doing a thread about FFT software, Soundcards and instrumentation interfaces...
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by JR. »

I only know a very little about the "fast" Fourier transform from trying to teach myself DSP... I plan to really dig into it with my next generation (dsp based) drum tuner, to read drum resonances, I plan to experiment with sampling and windowing tricks to get both fast crude and slower precise data.... I know with conventional FFT approaches it's an either or decision, but I suspect I can do tricks where I grab the samples once and re crunch some of the same data first fast and later after I've accumulated more, slower with more resolution.

Sorry for the veer...

Another area I would love to see is a practical interface between computers or smart phones, to hardware. I am fluent in the embedded digital controlled analog side, but still deaf/dumb about communicating with popular computing devices.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by mediatechnology »

John you really should download Visual Analyzer and play with it. It's free but is a Windows application.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by JR. »

mediatechnology wrote:John you really should download Visual Analyzer and play with it. It's free but is a Windows application.
That will just depress me even more, every time I think about redesigning my old TS-1. :oops:

I looked at several PC based cheap SA several years ago when I started investigating drum tuning and found them pretty much useless for my purposes, back then. Not to feed this veer, but I have developed a unique (I think) method of synchronous rectification where I can make 5 one octave apart spaced notes and read the levels coming back from all five simultaneously, with enough resolution for fast scans. This is not perfect and I get leakage from loud octaves into quieter octaves, while artifacts are down around -20 dB, so good enough for my rough scans, but still susceptible to false ghost resonance readings.

I suspect (hope) I can do some similar magic playing games with the low level FFT code implementation, on a DSP microprocessor while i don't know what I don't know yet. I did one textbook FFT implementation in a MICROCHIP DSPic development platform but it was crude and too low resolution for any effective tuning use. I didn't dig deeper into it at the time.

------
In case i wasn't clear, what I was looking for was some way to come up with a generic interface so a computer or smart phone could control a piece of audio equipment. I should probably piggy back on some recording environment controller standard. My goal is a hardware audio box without expensive knobs and control interface, but with multiple layers of control access via cheap computer interface

Best I see from hardware without needing a bunch of unique driver software is supporting my own IP stack in the micro to support a standard browser interface. But I haven't seriously pursued this either. I think this is what the new cheap digital mixers are using.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by mediatechnology »

I'm posting some pics of the Input-Capacitorless Mic Preamp Protoboard using the THAT1570 in a fully-balanced topology.

This is a huge improvement over the THAT1510-based version because the common mode servo is not used as an audio reference - it only brackets the power supplies.
The preamp IC and circuitry immediately surrounding it has no "ground" connection other than the input network.
Common mode rejection is realized in a non-floating, cross-coupled, double-balanced stage after the preamp IC's output coupling capacitors.
The THAT1510 cannot be used this way because it realizes common mode rejection internally.

The three electrolytic capacitors that you see poking up are not in the audio signal path. They are bypass only.
All signal path capacitors are film.

The Preamp:

Image
Fully Balanced Input Capacitorless DC-Coupled Preamp Using a THAT1570.

Lay of the land from left to right:

Input network. TO-92 transistor is 48V sub regulator/capacitance multiplier for phantom.
Orange capacitor is CM servo LPF.
Resistor network to the right is phantom pull-up and diode protection bridge to rails.
Green sub-board is THAT1570 adapter board.
8 pin DIP package is OP2277 "differential Deboo" servo.
8 pin to the right of the servo is an NE5532 (or LME49860) dual op amp.
Section A is the CM DC servo providing "flying rails." Section B is an option AC CM drive for the Rgain switch shield. (proto only)
The TO-220 complementary power devices are the flying rail bootstrap transistors. The heat sinks are over sized.

Above from right to left:

Above the TO-220 devices are a DB-104 protection bridge for the non-floating stage.
The 4 Wima film caps provide DC-blocking and an optional 40 Hz 2 pole -12dB/octave HP filter.
The F4 switch - only two poles are used - are for the HPF.
The 8 pin DIP to the right of the film caps are an NE5532/LME49860 instrumentation amp configured to work as both a +10 dB post-amp (ribbon mode or variable 0-10dB gain trim) and a 40 Hz 12dB/octave Sallen-Key HPF.
The two DIPS to the left of the INA are cross-coupled THAT1246s to realize common mode rejection.
The 1246s will drive current-boosted THAT1646s which are not shown.

The Power Supply:

Image
Fully Balanced Input Capacitorless DC-Coupled Preamp Power Supply Providing +66V, and +/-18V.

The power supply turned out to be very simple.
A single 36VCT transformer is used. (Thanks Roger.)
There are two DB-104 bridges.
The big filters are for the non-floating +/-18V rails.
Two of the black capacitors float a second bridge rectifier for the ~85V DC bulk supply.
The third black capacitor is the 85V filter.
The preamp has exceptional low-line performance <<90VAC.
An LM317/LM337 pair regulate the +/-18V rails.
A TL783 provides a +66V DC supply for the floating preamp supplies.

The way I look at it, two capacitors are removed from the input of a conventional preamp and are moved to the power supply. Otherwise the component count in the PSU is about the same: There's +/-18V and a HV supply with either a conventional preamp, or this one.

The only gizmo required to eliminate the input coupling capacitors of a conventional preamp are the CM servo which is a single op amp, two power transistors and a few passives.

The other stuff is something you need anyway: The differential Deboo is required in a capacitor-based preamp if you want to eliminate Cgain.
emrr
Posts: 570
Joined: Sun Mar 02, 2008 10:27 pm
Location: NC, USA
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by emrr »

I like it.
Best,

Doug Williams
Electromagnetic Radiation Recorders
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by mediatechnology »

Thanks Doug. I just need to buy a bunch of epoxy to hold it together for shipping. :lol:

The gain range is ~ +9 to +66 (or +19 to +76 with +10 dB ribbon boost in.)

Do you think it needs a 20 dB pad?
emrr
Posts: 570
Joined: Sun Mar 02, 2008 10:27 pm
Location: NC, USA
Contact:

Re: A Direct-Coupled Input-Capacitorless Active Mic Preamp

Post by emrr »

I really can' t imagine what it would need a pad for, unless you want to consider those who take DI signals off the output transformers of Marshall amps.

The one time I used an epoxy to repot an ancient power output transformer, I had a nosebleed for 4 hours. I try to avoid the stuff!
Best,

Doug Williams
Electromagnetic Radiation Recorders
Locked