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.
Locked
User avatar
mediatechnology
Posts: 5444
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. Looks like the 4050s may be the ones which are not transformer output.
Pair of 4050s
0.2mV/0.7mV
The Vos is <<1 mV. Not a smoking gun. Anyone else want to measure some phantom-powered mics?

Anybody got a TLM103?
Bruno2000
Posts: 38
Joined: Fri Jan 04, 2008 5:32 pm

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

Post by Bruno2000 »

We have TLM 103s and TLM 149s, and some Schoeps. What exactly are you measuring?
The large print giveth, and the small print taketh away - Tom Waits
User avatar
mediatechnology
Posts: 5444
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 »

Wow, cool - thanks!

What we're looking for is the offset developed by the microphone when phantom is applied through the customary 6K81 pull-ups. It cannot be loaded by a transformer but an active input, or just the pullup resistors as load, will tell us how symmetric the current draw is.

The DC potential between pins 2 & 3 (probably millivolts) is what we're looking for but knowing the actual phantom voltage relative to ground couldn't hurt either as we can learn more about the current draw.

Roger, was there an input transformer involved when you made the 4050 measurements? An input transformer would also force the offset between pins 2 & 3 to be equalized. It's easy to forget that it's there and I forgot to mention that.

Wayne
User avatar
mediatechnology
Posts: 5444
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 »

Glad I thought to ask. When you get a chance you might try it with a couple of 6K81s pulled up to 48V.

If you don't have an easily obtainable 48V supply you can get 45V from the IP27-18 by stacking all three or use the dual 20V and a 9V battery in series.

I'm curious about Bruno's TLM103s....
Bruno2000
Posts: 38
Joined: Fri Jan 04, 2008 5:32 pm

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

Post by Bruno2000 »

I'm doing measurements now, but it seems to take quite a while (15 min so far) to have the voltage settle. (AT 4033 transformerless right now) I’m using our SSK 4K transformerless mic pres. Does this seem right?

EDIT: That SSK is a new top-secret console design :o)
Last edited by Bruno2000 on Thu May 07, 2009 12:14 pm, edited 1 time in total.
The large print giveth, and the small print taketh away - Tom Waits
User avatar
mediatechnology
Posts: 5444
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 found that with some cap-coupled inputs it can indeed take awhile to settle, particularly the common mode voltage.

If the preamp has been off awhile it takes ages for the capacitors to re-form.

What sort of differential (pin 2 to 3) offsets are you seeing?
Bruno2000
Posts: 38
Joined: Fri Jan 04, 2008 5:32 pm

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

Post by Bruno2000 »

OK, here’s what I measured, although unexpected:
AT 4033 serial 39XX DCR=~740K offset=6.8mV
AT 4033 serial 90XX DCR=20K offset=0.3mV
AT 4033 serial 95XX DCR=22K offset=0.7mV
TLM103#1 DCR=9K5 offset=0.8mV
TLM103#2 DCR=9K4 offset=0.9mv
Schoeps CMC5 body DCR>2M offset=15.7mV (wow, checked it 3 times)
TLM149 DCR=3K9 offset=0.8mV
Hope this helps!
The large print giveth, and the small print taketh away - Tom Waits
User avatar
mediatechnology
Posts: 5444
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 »

OK, here’s what I measured, although unexpected:
AT 4033 serial 39XX DCR=~740K offset=6.8mV
AT 4033 serial 90XX DCR=20K offset=0.3mV
AT 4033 serial 95XX DCR=22K offset=0.7mV
TLM103#1 DCR=9K5 offset=0.8mV
TLM103#2 DCR=9K4 offset=0.9mv
Schoeps CMC5 body DCR>2M offset=15.7mV (wow, checked it 3 times)
TLM149 DCR=3K9 offset=0.8mV
Hope this helps!
Awesome! All easily within the servo correction range with plenty of room to spare. I'm suprised many of them weren't higher.

Any one else here want to follow Roger's and Bruno's lead?

Thanks for doing this!
Bruno2000
Posts: 38
Joined: Fri Jan 04, 2008 5:32 pm

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

Post by Bruno2000 »

What happens (don’t laugh/cry, this really does happen) when someone plugs an un-balanced device into the servo’s input with the phantom power on, thus grounding the – servo input?
The large print giveth, and the small print taketh away - Tom Waits
User avatar
mediatechnology
Posts: 5444
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 »

No problemo and a great question.

IIRC correctly I measured the peak current into the 1510 it was around 8 mA vs. the >>3A when input caps are present. I tried just about every way I could to blow up a 1510. I failed, it didn't.

That particular fault, when input caps are used (one input pulled up, the other grounded, then the second one grounded) is the worst-case fault for a 1510. An amp or more easily flows through the 1510 input transistors and back out the other input. When Rgain is 5-10 Ohms the current is significant. You'll notice on the 1510 data sheet (the new one) two added diodes to parallel the reverse polarity Vbe diodes internal to the 1510. With the added diodes and 10R input resistors the 1510 will survive. But the currents are just huge. Sparks can be drawn from the stored charge.

In the DC-couped topology, a voltage divider is formed by three resistors: The 6K81, the Rterm (we'll call it 2K) and the second 6K81. With one input grounded, the "bottom" 6K81 is shunted to ground forming a divider of 6K81/2K. 48V at the top of the 6K81 becomes ~10.9V at the 1510 input.

In that fault situation, the servo has to ramp from whatever potential it was before, say 35V with a mic connected, to the newly-developed common mode voltage of 10.9V/2. The common mode servo tau is around 50 ms. During that time the inputs may be negative relative to Vcc and Vee. With a 35V Vcm, Vee is about +20V. That's when the steering diodes - the bridge to supplies from tip and ring - begin to conduct.

In my tests I ran the 1510 without any external protection diodes just to see if I could blow it up. I couldn't.

As to the servo and its' phantom tolerance the inputs to it have very large-value resistors. The input is an integrator. I used 100K X2 and 470 nF. The peak phantom currents into the input can't be more than 500 uA and there is some stored charge on the 470 nF. There probably ought to be a couple of 1N4148s as steering diodes to the OP07's flying rails as there are on the low-voltage "dry" side of the buffer that follows the 1510 stage.
Locked