182MHz Oscillations :-0

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
www.hifisonix.com
Joined 2003
Paid Member
I was doing some debugging this week end. Here's what a cascode can do if your layout and cascode base circuit are not taken care of. I have left a small write up on my experience on my web site - last page. :cool:
 

Attachments

  • Cascode_Oscillation.JPG
    Cascode_Oscillation.JPG
    523.3 KB · Views: 396
Last edited:
I was doing some debugging this week end. Here's what a cascode can do if your layout and cascode base circuit are not taken care of. I have left a small write up on my experience on my web site - last page. :cool:

Do not cascode your VAS. Most of the "improvements" dictated by race for lowest THD and based on simulations only in fact lead to VHF local oscillations .... :)
 
www.hifisonix.com
Joined 2003
Paid Member
Hello PMA - the picture above was for the front end LTP cascode . . .

My new design does not cascode the VAS. I agree its not needed unless you are shooting for sub 1ppm - but that is not my aim. What's clear is if you want to cascode, you need a good widebandwidth scope :)
 
Last edited:
BJT input fully balanced with cascode. If you go to my website on the last page (Audio Amplifier Design) its right at the top of the page Audio Amplifier Design and Circuits | hifisonix.com.

I was using a 200MHz analog scope and it looked like noise until I hooked up Rigol digital scope.

That's it ... the BJT input fully balanced with cascode - same problem. I will visit your page. Many designs have similar problems inherent without being disclosed ..

Thanks,
 
Hi Andrew,

Before further discussing your oscillation problem, I've a more fundamental question:
Why connecting the bases of the cascode trannies to a fixed voltage (12V) anyhow?
Wouldn't it be better (and much simpler!) to let them follow the input voltage, as shown below. Moreover, in this case you get the base stoppers (R11&R14) for free.;)

Cheers,
E.
 

Attachments

  • cascode.png
    cascode.png
    10.6 KB · Views: 313
Disabled Account
Joined 2006
First Id like to ask if a little oscillation at 200 MHz is really such a big problem, does it affect sound audibly ?? It is true that these small quirks arent always shown and its not easy for most builders to find with the scopes they have.

Secondly Ill recommend a book, here you can find in depth analises on these problems and solutions if you deem them necessary. There are more problems with other simple circuits than most think than just with cascodes.

http://www.amazon.com/Wideband-Amplifiers-Peter-Staric/dp/0387283404

Excellent book, except the price but worth it if you donrt want these type problems.
 
Edmond,
why do the "tails" of R11 & R12 tie into the nearby degeneration resistors? Why don't the tails swap over to the opposite side degeneration Resistors? This would reference the cascodes to their respective LTP emitters.

Bonsai,
thanks for confirming that. It seemed from the pdf that lead impedance of all 3 components could affect the stability, but I wasn't sure I was reading too much into my interpretation.

Home,
I'll do without the book and let the Forum experts guide me towards nirvana.
 
Last edited:
www.hifisonix.com
Joined 2003
Paid Member
First Id like to ask if a little oscillation at 200 MHz is really such a big problem, does it affect sound audibly ?? It is true that these small quirks arent always shown and its not easy for most builders to find with the scopes they have.

Secondly Ill recommend a book, here you can find in depth analises on these problems and solutions if you deem them necessary. There are more problems with other simple circuits than most think than just with cascodes.

Amazon.com: Wideband Amplifiers (9780387283401): Peter Staric, Erik Margan: Books

Excellent book, except the price but worth it if you donrt want these type problems.

Homemodder, yes, I think this is a problem. Ripple and the audio signal can modulate the oscillation. This comes out the speaker as a shhh sound and also in the case of ripple as harmonics thereof.
 
swapping

Edmond,
why do the "tails" of R11 & R12 tie into the nearby degeneration resistors? Why don't the tails swap over to the opposite side degeneration Resistors? This would reference the cascodes to their respective LTP emitters.
[...]

Hi Andrew,

As far as I can see, you cannot swap them (that is, without adding complexity).
Besides, because of symmetry, the circuit makes no 'objections' if you reference the cascodes to the 'wrong' LTP's.

Cheers,
E.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.