In this 'ring of three' which is gain-set?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

k brown

Well-known member
Joined
Jul 7, 2021
Messages
1,250
Location
California
Unlike most I've seen, this appears to have two FB loops. The 15k in parallel with the R in the 'outside' loop I added to reduce gain from ~40dB, to a more manageable ~30dB; but is that the right one to change - should gain be set by changing the 10k R in the 'inner' loop?

If these are two FB loops, would this be done to improve linearily/distortion, or some other reason?

Also, is the 200uF cap bypassing the 1.5k to gnd be functioning like a cathode bypass in a tube gain stage? - it increases gain but adds a bit of distortion?; would this circuit be cleaner without it? Less gain is actually what I'm after. Optimally, a switch between 20dB and 30dB.

Or should I just leave well enough alone, and just put a pot on the input ; - )

Thanks
 

Attachments

  • DSCN0763.JPG
    DSCN0763.JPG
    127.7 KB · Views: 28
some of the old masters could give you a better answer but consider 1 loop is dc coupled back to t1 emitter and the other is ac coupled--they are in parallels
 
Unlike most I've seen, this appears to have two FB loops. The 15k in parallel with the R in the 'outside' loop I added to reduce gain from ~40dB, to a more manageable ~30dB; but is that the right one to change - should gain be set by changing the 10k R in the 'inner' loop?
As shabtek mentioned, there are a DC and an AC FB loop. So yes, putting this 15k resistor in parallels with the 68k is the way to go.
However, you can't go much further, because the FB loop impedance would load the output too much (the current is limited by the 10k emitter resistor). It also put more DC on the output.
If these are two FB loops, would this be done to improve linearily/distortion
That's the main reason for NFB. Along with improving DC stability.
, or some other reason?
Who knows what the designer's motivations are? Personally, I think it's a little too complicated and does not tick all the boxes.
Also, is the 200uF cap bypassing the 1.5k to gnd be functioning like a cathode bypass in a tube gain stage?
Yes.
- it increases gain but adds a bit of distortion?; would this circuit be cleaner without it?
It does increase the distortion of this stage, but also increases its gain. Since the system is under global NFB, the amount of distortion is actually reduced.
This is a debatable subject. Several designers advocate local nested NFB loops, so each individual stage has less distortion and the global NFB "has less work to do", which results in a less unpleasnt distortion spectrum.. The debate is not closed.
Less gain is actually what I'm after. Optimally, a switch between 20dB and 30dB.
You may try increasing the 470r emitter resistor in the 1st stage, but I m not sure you can triple it (that's what you would need for a 10dB gain reduction) and maintain proper DC operating conditions.
Time for turning on the soldering iron.
 
Thanks so much for all that - probably best to leave FB where it is, remove the bypass cap, and simply pad the input to reduce overall gain.

Thanks again.
 
Interesting - removing the 200uF bypass renders the unit inoperative (?); so it went back in.

Getting what I wanted was easier than I expected; finally got around to checking this against circuit of known gain, and with 15k parallel, it was much closer to 20dB than 30. So only needed to switch between a higher value, the 15k and no added R to get gain switchable ~20/30/40dB - just what I was after.

Thanks for the help!
 
Last edited:
It's a humble little Lafayette Radio home HiFi mic/tape/phono preamp. Probably from the mid sixties. Given the low parts count, the tape and phono EQs are probably terrible, but the mic gain is fairly clean and a bit 'soft' sounding. It's called "Stereo Mixer/Preamplifier"; odd because it doesn't mix anything - just L/R levels and EQ selector sw.
 

Attachments

  • Screen Shot 2021-07-10 at 12.20.02 PM.png
    Screen Shot 2021-07-10 at 12.20.02 PM.png
    142.9 KB · Views: 16
I had something similar back in the 60s. It was a PCB with a little mains transformer that you mounted under the record deck. Only two transistors in each channel. Output impedance was so high it had to feed a tube amplifier
 

Latest posts

Back
Top