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
XAUDIA
Posts: 12
Joined: Sun May 25, 2008 5:31 pm
Location: York, UK
Contact:

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

Post by XAUDIA »

Thanks guys. I've had another look at the chip schematic in the datasheet & Mediatechnology's post, and that makes sense.
raf wrote:The gain was relay controlled and I did not use Cgain. The only time I could see or hear the servo working was while changing gain. After the servo settled all was good AFAIK.
I think you will be surprised how these sound!
That's good to know. I'm looking forward to hearing them.
raf wrote:Mine had phantom blocking caps, two 20uF 100V polypropylenes.
:shock:
New music at xaudia.com
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 »

If Cgain is used the output Vos, without any correction will be < +/- 5 mV. With Cgain a servo is unnecessary. Adding a servo will actually increase offset. I'm not sure a trimmed TL071 is much better as it can be trimmed, but the drift is high.

The output offset servo does contribute noise: An LF351 or TL071, at lower gains where second stage noise dominates, is marginal there. Quiet enough but barely so. I haven't calculated it myself.

Analog Secrets and Phantom Menace both talk about the input servo as does the demo board.
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 »

That's a temperature drift: 10 uV/deg C Vos and a doubling of Ibias for every 10 deg C. Probably not enough to hurt but it could go up to 1/2 mV for +20 deg rise with a 500K input resistor due to Ibias with a TL071. OP07s are cheap and you save a $3 trimmer.

This is why you wouldn't want to use a TL071 as an input servo (plus also noise). As an output servo it's OK if you don't mind the trim.
User avatar
XAUDIA
Posts: 12
Joined: Sun May 25, 2008 5:31 pm
Location: York, UK
Contact:

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

Post by XAUDIA »

My basic preamps with output-servo'd 151x are up and running, and on a first listen seem pretty uncoloured. More serious tests today!

Thought I'd post up the resistor table calculations I came up with. These are for the 1512 with 12 and 24 step switches. I'm just using a 12 position rotary for now. Basic stuff but it may be of use to someone.
Having listened to it at the highest gain settings (67 and 62 dB using 'real' resistors) I don't think I'll ever need the top two steps so I may recalculate for 2dB steps and just run up to about 57dB gain.

Since I have two channels, my plan is to keep one as it is and add in Waynes additional stages one by one, to see if I can detect any influence on the sound.

Roger, I see you used the Goldpoint switches in your control room box. How are you finding them?

Cheers!
Stewart
New music at xaudia.com
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 for the update Stewart.

You also have +6 dB of gain built in with the 1646. Your 57 dB of preamp gain will actually be 63 which, as you point out, should be plenty.

You might want to have a pair of low value resistors as "stop resistors" one in each leg in series with pin 1 and 8 and have the switch vary resistance in series with those. The highest gain setting on the switch could then be a short. Those two Rs will isolate capacitance in the switch line and by having two of identical value keep things more balanced. It doesn't take much of a Cstray to reduce HF CMR. I think they touch on that in "Analog Secrets" at the THAT website.
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 »

Stewart - Did you get something built?
User avatar
XAUDIA
Posts: 12
Joined: Sun May 25, 2008 5:31 pm
Location: York, UK
Contact:

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

Post by XAUDIA »

Hi Wayne,

I haven't got past my very basic 1512 -> servo'd outpt (OP77) -> 1646 design. I had a hard drive crash this week which has cost me a lot of time, which is why I haven't been back to post. :evil:

I did get to run the basic preamps last Sunday recording some stereo electric guitar tracks, using a pair of Crowley and Tripp Naked Eye ribbon mics. My perception of the preamps in that application was that I didn't even notice them, which I think has to be a good thing & the whole point of this design. I was using 12-24 dB gain on the 1512 (so 18-30 dB overall).

I've just ordered up a pair of the 24 position goldpoint switches to give me better control.

Here are my latest calculations for the Rg, including the nearest 1% resistor values. I've tried to minimise the total error - hope it makes sense.
Thanks for the tip about using a pair of resistors to isolate stray capacitance - I'll give that a go on the next rebuild.

They may have been pointed out before but I found a couple of links to commercial DIY projects using THAT chips:

http://fivefishstudios.com/index.php?op ... &Itemid=59
http://www.seventhcircleaudio.com/T15/T ... _about.htm

Cheers!
Stewart
New music at xaudia.com
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 »

Stewart - Glad it went well with the preamp. Sorry about the hard drive.

The Five Fish and Seventh Circle are both nice designs. Owel posts occasionally at Prodigy and I made a suggestion WRT the Seventh Circle that got incorporated into the design. Don't know if they saw my post at Prodigy about it or just made the change themselves. Both are really clean and nice layouts.

I did a THAT1510 THAT1512 gain calculation spreadsheet some time ago that I just found. It uses an equation I got from EDN that takes "odd" resistor values and converts them into E96 1%. (EDIT: Fixed the OpenOffice problem by using LOG10 rather than LOG.) New rev here:

http://www.ka-electronics.com/THAT1510/ ... n_rev2.xls

I ended up building a 12 step gain control using 1%s calculated by the spreadsheet and they came out exactly - for all practical purposes - right.

The E96 equation looks like this:
=ROUND(10^(INT(LOG10(E7))+(ROUND((LOG10(E7)-INT(LOG10(E7)))*96,0))/96),2-INT(LOG10(E7)))
E7 is the example cell which contains the value of resistance to be normalized to E96 1%.
User avatar
XAUDIA
Posts: 12
Joined: Sun May 25, 2008 5:31 pm
Location: York, UK
Contact:

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

Post by XAUDIA »

That looks like a very useful spreadsheet, and for the record it works with the apple iWorks spreadsheet program too.
Thanks for sharing!
New music at xaudia.com
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 »

Glad it works in iWorks.

I think the problem is an error in the way OpenOffice evaluates the log function. The log function can be in any base. Calc does not see a base specified but should assume, in the absence of one, base e or 10. It doesn't.

I used the log10 function in a cell and it appeared to work.

OK, Calc will evaluate it properly if log10 is used. So will Excel. I think Excel assumes base 10 for the log function; Calc does not and sees a parameter missing in the argument.
Locked