New: 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: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

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

Post by mediatechnology »

Thanks for checking in. Have been wondering how you were doing.
We did discuss in an email varying the top of the phantom resistors similar to what Paul Frindle did.
You are correct that if it was the only servo it would be no different then servo'ing the bases.

What I was proposing was dual servo loops.

One which samples the offset at the input terminals of the preamp and applied correction at the top of the phantom resistors.
That servo would only serve to correct mic's DC offsets and would be very slow.
This servo is the bridge balancer.
SInce the sample point is at the input terminals the servo is not subject to the system gain and the servo Fc is fixed.

With the mic seeing coarse correction a second servo would be used to correct input transistor Vbe errors.
Your internal servo approach, with Fc fixed, is the one I used in the MC preamp and it would be ideal to correct Vbe errors.

What caused me to think about this is going back to the original Neumann spec of 4% max current imbalance.
When I analyzed what 4% could produce and wrote about the dual servo I had forgotten about how much input termination reduces that 4% spread.
olafmatt
Posts: 44
Joined: Wed Oct 22, 2014 11:33 am
Location: Europe

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

Post by olafmatt »

Ah, dual servo loops, now I understand. Will have to think about this, whether it makes any difference in reality.

What I realized is that if you have a servo that is correcting the offset across Rg, then this correction will also show up on the transistor's bases. Not in a totally linear way (because Vbe varies with current), but if we consider Vbe more or less constant, than any change we do "from the back" to the voltage at the emitters (i.e. your second servo) will also be seen on the bases.
But I do agree that if this is only correcting the DC offset due to Vbe mismatch, we're talking about microvolts here. But watch out for interaction between those two servos then. If the first servo is slow, there will at some point be DC offset across Rg (and on the output) that the second servo will then try to correct for. - Maybe we should start a new thread about interaction between DC servos... because I can picture many funky things to happen if you connect two fully DC-coupled circuits together that each use a DC servo internally.

I guess we just agree to disagree on whether it is a good idea to fix "microphone problems" (DC offset due to unequal phantom current draw) from within the preamp.

One other thing I was thinking about is whether your floating frontend helps to reduce distortion caused by common-mode signals. I've been using cascodec input stages for a while now, to keep the voltage across the input transistors constant and thus reduce distortion caused by common-mode signals modulating the input capacitance.
If a truely floating frontend would "ride" on the common-mode offset, it would have a similar effect. But IIRC, your common-mode servo that floats the frontend has some time-constant in there that basically means we're floting the fontend to some average voltage. This might help against distortion caused by microphones that temporarily draw more current (i.e. for loud signals), causing phantom voltage level on the input of the preamp to drop (if the rate of change is slower than the servo).
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

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

Post by mediatechnology »

I started counting and there are three servos if we count the common mode servo operating in feed-forward.
Instead of a "twin servo" (already taken) it's a triple servo. LOL.

I think the time constant of the servo that corrects microphone bridge imbalance could be made very, very slow since it only has to operate at turn-on, during stabilization, and after that, it is only correcting long-term DC drift. JR has pointed out I could use a micro for that and only have it re-calibrate if it went way out of bounds.
I guess we just agree to disagree on whether it is a good idea to fix "microphone problems" (DC offset due to unequal phantom current draw) from within the preamp.
If the microphone is directly-coupled to the bases is there any other choice?

I've observed CM modulation due to mic current increases at higher SPLs.

Off-topic but:
I should point out for readers that the AC reference for realizing CMRR is still 0V.
That's why I carry CM signals across the galvanic barrier and realize CMR on the "dry" side.
That "gotch-ya" is why I realized early that I couldn't do this using an all-in-one THAT1510.

Image
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

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

Post by JR. »

A micro could set some DPOTs and then shut itself off, unless you wake it up again... or not.

JR

PS: I still like the idea of flying a ADC up to phantom voltage, but tweaky customers who might appreciate this approach want a better ADC than I have handy.
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

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

Post by mediatechnology »

Funny JR I was thinking of DPOTs too and having the DPOTs trim the input error as closely as possible and then, once trimmed,not reactivate until the input error grew outside a predefined window.
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

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

Post by JR. »

The DPOT could hold a setting as long as power is not removed. I haven't looked lately but there might even be one with flash memory so it can keep a setting with power removed.

Might be good enough to just run a wake up routine, but might be better to let everything warm up and stabilize first.

or not...

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
terkio
Posts: 304
Joined: Thu Feb 07, 2019 3:17 am
Contact:

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

Post by terkio »

mediatechnology wrote: Sat Oct 05, 2019 5:43 pm Funny JR I was thinking of DPOTs too and having the DPOTs trim the input error as closely as possible and then, once trimmed,not reactivate until the input error grew outside a predefined window.
A calibration procedure ?
Initiated by the user and a warning led to tell when out of window.
I thought of this for servo loops that could be used to adjust, then frozen to keep the adjustment.
I see such a technic for a mike preamp: Half way between balancing with a trimmer and tracking with a servo.
olafmatt
Posts: 44
Joined: Wed Oct 22, 2014 11:33 am
Location: Europe

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

Post by olafmatt »

I think I've seen XLR chassis sockets with a switch that gets engaged once an XLR plug gets plugged in. That way a PIC (or some other logic) could detect when a mic gets connected and then run a callibration for a while and then store that value until the mic gets diconnected... but that way you wouldn't notice a user changing the mic on the other end of the mic cable. So it would probably also need to re-callibrate after phantom power gets toggled off and on again. And you still wouldn't catch mics that don't need phantom power.

This somehow brings us back to the original problem: Since the ciurcuit has DC gain equal to AC gain, 10mV of offset in the input can result in 10V of offset on the output (at 60dB gain). My gutt feeling tells me that as long as we have DC gain in there we have to get the correction on the input just right, and not just approximately. Otherwise the proposed second servo would need to do way more than just correcting the mismatch caused by Vbe differences.
User avatar
terkio
Posts: 304
Joined: Thu Feb 07, 2019 3:17 am
Contact:

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

Post by terkio »

Calibration to take care of Vbe mismatch.
I understand, it would be nice to have it automatic, transparent to the user.
However you want the system to calibrate when all has been set and temperature stable and you do not want the system to go into calibration while recording.
I doubt it can be user transparent. I think it should stay under control and responsibility to the user with a push button to launch calibration and a warning led to tell it's done or need a recalibrate.
Perhaps, this is not nice from a commercial point of view.
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

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

Post by mediatechnology »

The original version of this preamp used injection at the bases so it had lots of range to correct both input offsets and delta-Vbe offsets.
What I originally showed might not have had enough range, but plenty more was available.

There are two problems using base injection that we have all collectively figured out:

1) The servo Fc varies with preamp signal gain.
2) The servo response can be undesirably second order and peaking when it potentially interacts with the mic's output coupling caps.*

* Transformer output phantom-powered mics and emitter follower outputs aren't going to generate enough offset to be a problem. The capacitor-coupled output is the one that is subject to large (4% max?) current imbalances.

The prevention of a DC component between the input stage emitters is the goal.

Pinning the emitter differential DC to 0V to avoid DC gain and clicks during switching and letting the rest of the chips fall where they may is probably good enough.
If there's a small mV level of DC offset at the preamp output or the input transistor bases push the DC balance of the microphone a little bit that's probably OK.

Olafmatt's servo that I used in the MC preamp works great at controlling the Vbe offset but I'm not sure it has enough range to also correct the larger offsets the mic presents.
Post Reply