Author Topic: Sziklai pair  (Read 3821 times)

0 Members and 1 Guest are viewing this topic.

Offline Lucky-LukaTopic starter

  • Supporter
  • ****
  • Posts: 240
  • Country: it
Sziklai pair
« on: December 31, 2019, 09:26:00 pm »
Hi all
I've read that Sziklai is prone to high freq oscillations due to its feedback loop..
I can't figure out where this feedback loop actually is...
It should be obvious but I haven't understood this concept...
Happy new year.
Memento audere semper.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2783
  • Country: au
Re: Sziklai pair
« Reply #1 on: December 31, 2019, 10:30:36 pm »
Hi all
I've read that Sziklai is prone to high freq oscillations due to its feedback loop..
I can't figure out where this feedback loop actually is...
It should be obvious but I haven't understood this concept...
Happy new year.
The feedback loop being referred to is likely to be in the other circuitry controlling the Sziklai pair.
I guess that  Sziklai pair would have a degraded frequency response compared to a single BJT, requiring some extra attention to the loop's frequency compensation.
It is good practice to put a bleeder resistor across B-E of the second transistor to improve turn off response and also frequency response.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 
The following users thanked this post: Lucky-Luka

Offline Lucky-LukaTopic starter

  • Supporter
  • ****
  • Posts: 240
  • Country: it
Re: Sziklai pair
« Reply #2 on: December 31, 2019, 10:35:34 pm »
I think I've read the loop I was talking about is in the picture.
This loop wasn't mentioned talking about the Darlington pair.
Memento audere semper.
 

Online nfmax

  • Super Contributor
  • ***
  • Posts: 1556
  • Country: gb
Re: Sziklai pair
« Reply #3 on: December 31, 2019, 10:48:30 pm »
The feedback loop occurs within the Sziklai pair itself. Assume, for simplicity, that the input voltage is constant. A change in the collector current of Q3, acting across R2, causes the emitter voltage of Q6 to change. Since the input voltage is constant, this device operates in common base mode, so the change in Q6 emitter voltage causes a change in its collector current. But Q6's collector current is Q3's base current, so the change in Q6's collector current is amplified by Q3 and appears at its collector, completing the feedback loop.

At DC the sense of this feedback is negative. Q3's fT provides the dominant pole at the frequency where its beta starts to fall from its DC value. If the loop gain has not dropped to less than 0dB before the additional phase shift from Q6 reaches 90˚, the loop will oscillate. Too fast a device in the Q3 position, relative to Q6, is likely to give trouble.
 
The following users thanked this post: xavier60, Lucky-Luka

Offline Lucky-LukaTopic starter

  • Supporter
  • ****
  • Posts: 240
  • Country: it
Re: Sziklai pair
« Reply #4 on: December 31, 2019, 11:45:00 pm »
Does the Darlington pair have this loop too?
Memento audere semper.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2783
  • Country: au
Re: Sziklai pair
« Reply #5 on: December 31, 2019, 11:51:08 pm »
The feedback loop occurs within the Sziklai pair itself. Assume, for simplicity, that the input voltage is constant. A change in the collector current of Q3, acting across R2, causes the emitter voltage of Q6 to change. Since the input voltage is constant, this device operates in common base mode, so the change in Q6 emitter voltage causes a change in its collector current. But Q6's collector current is Q3's base current, so the change in Q6's collector current is amplified by Q3 and appears at its collector, completing the feedback loop.

At DC the sense of this feedback is negative. Q3's fT provides the dominant pole at the frequency where its beta starts to fall from its DC value. If the loop gain has not dropped to less than 0dB before the additional phase shift from Q6 reaches 90˚, the loop will oscillate. Too fast a device in the Q3 position, relative to Q6, is likely to give trouble.

That explains the reason for this circuit oscillating if the 27pF compensation capacitor isn't fitted.
It's an amplifier circuit I use for video signal.

HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2783
  • Country: au
Re: Sziklai pair
« Reply #6 on: January 01, 2020, 04:29:53 am »
Does the Darlington pair have this loop too?
For a single transistor Emitter follower, there is a very short and direct feedback path.
If the Emitter voltage changes for some reason other than a change of Base voltage, for example a change of load current, there will be a direct change of B-E difference voltage. This change will almost simultaneously alter the C-E current flow in a way that will mostly oppose the original change.
This correction can never be 100% so this feedback will be stable.
I'd say the same would also apply to a Darlington pair used as an Emitter follower.

I know from experience that a single transistor Emitter follower can oscillate due to external influences such as the parasitic inductance of long traces and poor bypassing.   
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6923
  • Country: ca
Re: Sziklai pair
« Reply #7 on: January 01, 2020, 05:57:03 am »
Sziklai pair in audio amplifiers, the BJT beta droop can cause problems. At higher currents the beta (gain) of the pair drop, which can make an amplifier's overall feedback loop unstable. It is also an issue in chip audio amps, where they use an NPN emitter-follower up top and Sziklai pair on the bottom, for the output stage like in LM386.
 

Offline OwO

  • Super Contributor
  • ***
  • Posts: 1250
  • Country: cn
  • RF Engineer.
Re: Sziklai pair
« Reply #8 on: January 01, 2020, 12:31:27 pm »
At higher frequencies even an emitter follower by itself can be unstable ;)
Email: OwOwOwOwO123@outlook.com
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: ca
Re: Sziklai pair
« Reply #9 on: January 01, 2020, 10:28:56 pm »
The emitter follower can indeed oscillate but it requires a capacitive load.  I was surprised when I first heard about it because emitter followers were supposed to be good drivers because they were unconditionally stable.  Here's a couple of papers on the subject:

https://www.eevblog.com/forum/beginners/noise-on-emitter-follower/?action=dlattach;attach=348962
http://audioworkshop.org/downloads/AMPLIFIERS_OSCILLATION_BJT_CIRCUITS.pdf

I understand that the capacitance on the output is transformed into an inductance in series with the base and if not sufficiently damped, it will oscillate. 

I think the Sziklai pair can be considered as an emitter follower first stage with a complementary common emitter amplifier booster stage. I wouldn't be surprised to see it oscillate and perhaps even more so than the Darlington pair.
« Last Edit: January 01, 2020, 11:52:40 pm by duak »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: Sziklai pair
« Reply #10 on: January 02, 2020, 12:38:50 am »
The feedback loop occurs within the Sziklai pair itself. Assume, for simplicity, that the input voltage is constant. A change in the collector current of Q3, acting across R2, causes the emitter voltage of Q6 to change. Since the input voltage is constant, this device operates in common base mode, so the change in Q6 emitter voltage causes a change in its collector current. But Q6's collector current is Q3's base current, so the change in Q6's collector current is amplified by Q3 and appears at its collector, completing the feedback loop.

At DC the sense of this feedback is negative. Q3's fT provides the dominant pole at the frequency where its beta starts to fall from its DC value. If the loop gain has not dropped to less than 0dB before the additional phase shift from Q6 reaches 90˚, the loop will oscillate. Too fast a device in the Q3 position, relative to Q6, is likely to give trouble.

That explains the reason for this circuit oscillating if the 27pF compensation capacitor isn't fitted.
It's an amplifier circuit I use for video signal.

Exactly although the compensation can also be done with a shunt capacitor to ground as shown below for the positive side output on a Tektronix PS503A dual supply.  I am inclined to think that any solution which does not use a series RC pair is guesswork.

The difficulty with this type of problem is knowing if there is sufficient phase and gain margin available when the two transistors have their worst case characteristics with the worst case load and operating point.  Notice that in this example, the series 0.6 ohm current shunt and 47 microfarad output capacitor help buffer the output load of the power supply in a measured way and are essential for stability.
« Last Edit: January 02, 2020, 12:43:58 am by David Hess »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf