Author Topic: Weird attenuator circuit - why / how does this work?  (Read 3372 times)

0 Members and 1 Guest are viewing this topic.

Offline TinkeringSteveTopic starter

  • Frequent Contributor
  • **
  • Posts: 298
Weird attenuator circuit - why / how does this work?
« on: July 31, 2015, 05:59:48 pm »
http://postimg.org/image/459i6l03p

Why / how does this circuit work?In my modest level of understanding of BJTs, this shouldn't work :-D
The two BJTs emitters connected together, but not to anything else? How can any current flow there?

complete circuit:
http://postimg.org/image/6arbcsgqz/

as LTspice file here:
http://s000.tinyupload.com/index.php?file_id=45691190684901314743

Doing AC analysis, looking at the "AudioOut" shows that the feedback path does what it's supposed to do.
The frequency response gets a peak around the cutoff, the higher the voltage at the control input, the higher the peak.

(Only the 2x BJT attenuation part is from the manual as said in the spice file, I just cobbled together a fixed 4 pole LPF before that to see the familiar response)
« Last Edit: July 31, 2015, 07:13:47 pm by TinkeringSteve »
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Weird attenuator circuit - why / how does this work?
« Reply #1 on: July 31, 2015, 06:06:15 pm »
I don't have a clue what any of the off-sheet connectors are.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline TinkeringSteveTopic starter

  • Frequent Contributor
  • **
  • Posts: 298
Re: Weird attenuator circuit - why / how does this work?
« Reply #2 on: July 31, 2015, 07:14:14 pm »
I don't have a clue what any of the off-sheet connectors are.

Tim

I added an image with the complete circuit
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Weird attenuator circuit - why / how does this work?
« Reply #3 on: August 01, 2015, 12:43:13 am »
Hmm, okay...

Not sure where this comes from, or what the purpose was; it appears to be a rather steep 4kHz low pass filter, but it's a very impractical way to do it.  The filter effect also varies with feedback, and will oscillate if it goes high enough.

Does Pot1 (or -a, -b) do anything or are they fixed, too?

As for the transistors, it looks like DC bias into the base will forward bias both C-B junctions (give or take which end is higher or lower in voltage).  This will connect some impedance (the incremental impedance r_e ~= Vth / Ib) between the output and feedback nodes.  The values as shown give something like 100kohms.

It's a poor method, and the AC Analysis will not tell you why, because it assumes a small-signal linearization around a DC operating point.  Transient Analysis will give more insight, such as nonlinearity and output DC offset, but this will still not tell you about the variation in performance over temperature and manufacturing tolerance.

A much better method to control gain is to use a proper gain-control device, such as a digital potentiometer (assuming you have something else digital handy to drive it), or a multiplying DAC; or among old fashioned analog solutions, an OTA (operational transconductance amplifier -- output amplitude depends on temperature, however) or PGA (programmable-gain amplifier, basically an OTA plus support circuitry).

Likewise, the filter should be improved by using a typical 3rd or 4th order active filter design, and not being placed within a variable gain feedback loop.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline TinkeringSteveTopic starter

  • Frequent Contributor
  • **
  • Posts: 298
Re: Weird attenuator circuit - why / how does this work?
« Reply #4 on: August 01, 2015, 02:22:55 am »
the 4 identical simple filters I just slapped on there, I only wanted to test the feedback part of it all, and give it vaguely similar surroundings to the original.
It is supposed to be for a voltage controllable filter, which usually do use such identical simple filter sections and control them in parallel, but it was quicker for me to do what I did there.

The feedback and even self oscillation are intentional effetcs for this kind of stuff.

I was just curious to try this out, even though I'm a hobbyist with very fractioned knowledge, even to me this looked crude, yet fascinating because cheap, low parts count, etc.
This 2 transistor feedback thing was used in Roland's budget polyphonic synth JX-3P. Its VCA uses one NPN + 1 opamp - that "budget" :-D
It did use some IC for the filter, with external caps and all, but the feedback circuit, to get controllable amount of resonance, was done the way like depicted here.
And although primitive, it was, nontheless, used in a commercial product which sold quite okay, end even sounded quite okay.


Thanks for the detailed explanations, and hints what tests to do.
That will be some homework, as I only know some of the things you mentioned, and the depth of understanding may not be enough.

The pot was just a quick way to make a resistor divider that I could manually change.
Just today I discovered the step param command in ltspice ;-)

But yeah, I probably will try to use on OTA - as soon as I get the LM13700 subcircuit working in ltspice...

I also though about using something like a CD4051 analog mux + 8 resistors to build a DIY digi pot, would that be advisable?
It can do only 8 steps if A,B,W connectors are needed as I see it, but that could be enough for this task.
All this analog synth stuff uses quite high voltages, and usual digi pots don't, those which do get quite expensive.
« Last Edit: August 01, 2015, 02:30:30 am by TinkeringSteve »
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Weird attenuator circuit - why / how does this work?
« Reply #5 on: August 01, 2015, 03:48:15 am »
Yeah, you can half-ass a low-bit (or no-bit :-DD ) DAC doing that.  It depends on having many different, accurate, resistor values though.  Or a CD4066 arranged as an R-2R switch, which gets you the same number of bits per chip (1-to-8 mux has A0-A2; CD4066 is a triple switch) but uses identical resistors (in singles and pairs).

On the upside, if you roll your own, you can build a log DAC, which was handy for telecom encoding, and would make a handier volume control (to get a log response), but might not be as useful for synth.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf