Output Stage Oscillating #2

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

bruno2000

Well-known member
Joined
Dec 29, 2007
Messages
1,570
Location
Atlanta GA
After seeing all the great responses to another Output Oscillating, I have this problem and am stumped.  This is the output stage of the PYE 4060 compressor, except for C*1.  That cap is not in the original schematic.  The transformer was wound by one of the  most respected transformer wizards on the Forum.  Without this cap, the circuit oscillates.  Looks like noise on the scope.  I've tried all kinds of "grounding", but nothing helps much except for the cap.  it's 100nF which seems huge.  Lower values don't touch it.  Put the same cap across TR6, and it oscillates again.  The circuit draws about 160mA  which also seems excessive.  Suggestions?
Thanks!
Best Bruno2000
 
Freq resp (with the cap) is flat to 55kHz with a 1.5 dB bump at 30kHz.  Distortion is about 0.03% at 0 dBu output 20-20k.  Phase looks like it should.
Thanks!
Best,
Bruno2000
 
> circuit draws about 160mA

I calculate 270mA. R30 R32 put say 5V to TR8, say 4V to R40 33r, 133mA each side, plus TR5.
 
PRR said:
> circuit draws about 160mA

I calculate 270mA. R30 R32 put say 5V to TR8, say 4V to R40 33r, 133mA each side, plus TR5.

Thanks for the comment!
That may be one of the problems,  The 160-169 mA is what I measured @ 15.9 VDC.
The circuit in its "stable" form has about 28 dB of gain.
Best,
Bruno2000
 
CJ said:
a dual trace scope would be nice, put one set of leads on the primary and one on the FB wind and look at the phase angle,

Differential scope.  Top trace is fb, bottom is op. (scope gain adjusted)
With extra cap.
Thanks for your help!
Best,
Bruno2000
 
CJ said:
did you try reversing the polarity of the FB loop?

scope patterns look good, maybe take the frequency up and check for shift,

Reversing FB makes HUGE square wave oscillation.
The way I have it wired is the only way it will work.
Thanks!
Best,
Bruno2000
 
try the .1 cap trick, alligator lead from ground to cap, go around the circuit with the cap, a base here, collector there, at which test point does the osc stop...
 
bruno2000 said:
After seeing all the great responses to another Output Oscillating, I have this problem and am stumped.  This is the output stage of the PYE 4060 compressor, except for C*1.  That cap is not in the original schematic.  The transformer was wound by one of the  most respected transformer wizards on the Forum.  Without this cap, the circuit oscillates.  Looks like noise on the scope.  I've tried all kinds of "grounding", but nothing helps much except for the cap.  it's 100nF which seems huge.  Lower values don't touch it.  Put the same cap across TR6, and it oscillates again.  The circuit draws about 160mA  which also seems excessive.  Suggestions?
Thanks!
Best Bruno2000
This is another example of a negative-impedance drive circuit.
In fact the analysis of the circuit is quite complex because of the nested NFB loop, the PFB loop  for the negative-Z drive, and the overall NFB loop from the tertiary winding.
You may have to alter the values of R28/R40 and/or R29/R36. These values are determined for a particular value of the xfmr's primary DCR; if the xfmr has lower DCR, you may expect unstability.
I would start by lifting R29/36.
 
abbey road d enfer said:
This is another example of a negative-impedance drive circuit.
In fact the analysis of the circuit is quite complex because of the nested NFB loop, the PFB loop  for the negative-Z drive, and the overall NFB loop from the tertiary winding.
You may have to alter the values of R28/R40 and/or R29/R36. These values are determined for a particular value of the xfmr's primary DCR; if the xfmr has lower DCR, you may expect unstability.
I would start by lifting R29/36.

Thanks Abby!
I'll plug in some resistor sub boxes and give it a try.  Here is a table of published DCR values (4060) vs the measured values of the transformer I am using.


                                                  4060 Book                        Measured  XF (12" test leads)
+ PRI to CT                        2.16r                                      1.55r
-Pri to CT                            2.12r                                      1.45r

FB                                            54.2r                                      39.22r

OUT                                        10.3r                                        13.31r

Could the less than 1r difference in the primary be the problem?
Thanks for the help and analysis!
Best,
Bruno2000

 
CJ said:
try the .1 cap trick, alligator lead from ground to cap, go around the circuit with the cap, a base here, collector there, at which test point does the osc stop...

I'll try this, too.
Thanks!
Best,
Bruno2000
 
abbey road d enfer said:
Very difficult to say; it's less than 1r, but still 30% out...
But it may be a totally different issue with the stability margin; have you tried increasing C26/C27?

No, but I will this afternoon.
Thank you!
Best,
Bruno2000
 
The problem as I see it, in general terms, is that the positive feedback loop is critically dependent on the parameters of the transformer. This means that, unless you use the actual transformer the circuit was designed for, it is going to be very difficult to ensure stability of the circuit  by trial an error. In addition, the transformer actually being used probably does not need the positive feedback for proper operation so it may as well be dispensed with.

Cheers

Ian
 
bruno2000 said:
No, but I will this afternoon.
Thank you!
Best,
Bruno2000


I hooked up a couple of cap decade boxes across C26/C27 and found that a value of 6nF stops the oscillation without C*1 in the circuit ;o)
That still seems awfully high, but the freq response and phase still good.
Thanks so much for the suggestion.


ruffrecords said:
The problem as I see it, in general terms, is that the positive feedback loop is critically dependent on the parameters of the transformer. This means that, unless you use the actual transformer the circuit was designed for, it is going to be very difficult to ensure stability of the circuit  by trial an error. In addition, the transformer actually being used probably does not need the positive feedback for proper operation so it may as well be dispensed with.

Cheers

Ian

I agree, Ian! And thanks so much for your input!  I'll bet the transformers I have outspec  / outsound the originals, but it's the "spirit" of the circuit that interests me.
The transformers I'm using are a second "best guess" as we have VERY limited info on them.
We'll keep plugging away.
Best,
Bruno2000
Wonder if anyone has a photo of the originals?
 
A little more testing......
Optimal distortion figure is with C26/C27 at 8n2F, however, the distortion gets  much worse at higher outputs (around 20dBu) than the circuit with only the C*1 added. 
Rats!
Best,
Bruno2000
 
FYI: the transformer is based on the Manley Vari Mu OPT,

F/B phase angle looked pretty good,  maybe mess with the layout, move some wires around,
 

Latest posts

Back
Top