John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Jump to Latest
#18,341
john curl said:
What is the problem with the 10 ohm resistors, Mark? Am I missing something? The 10 ohm resistors are a crude current limiter for the fast, but very fragile driver transistors (RCA).
Click to expand...

In my opinion there is nothing wrong with them, in fact I think they're a good idea. In addition to the benefit you mention, those resistors also tend to reduce instability / oscillation when an emitter follower drives another emitter follower. I call attention to the resistors because they are unusual and because I believe they are worth thinking about.
 
#18,345
Mark Johnson said:
In my opinion there is nothing wrong with them, in fact I think they're a good idea. In addition to the benefit you mention, those resistors also tend to reduce instability / oscillation when an emitter follower drives another emitter follower. I call attention to the resistors because they are unusual and because I believe they are worth thinking about.
Click to expand...
I use to put a serial resistance between the two stages of my diamonds. Using them, in input stages of my amps, to set a low pass filter, independent from the input's source impedance. It kills too those "nervous" behaviors.
 
#18,346
Mark Johnson said:
In my opinion there is nothing wrong with them, in fact I think they're a good idea. In addition to the benefit you mention, those resistors also tend to reduce instability / oscillation when an emitter follower drives another emitter follower. I call attention to the resistors because they are unusual and because I believe they are worth thinking about.
Click to expand...

Good insurance policy. You cannot build an EF3 without them.

This is also discussed in Bob Cordell's book.

Another nasty one I've experienced is cascodes - the 2N5551/5401 often used must have a base stopper c. 470 to 1k and small loop areas in the base<>emitter circuit otherwise you will get problems. I've had two instances where they oscillated at >150MHz 😱
 
#18,347
Bonsai said:
Good insurance policy. You cannot build an EF3 without them.

This is also discussed in Bob Cordell's book.

Another nasty one I've experienced is cascodes - the 2N5551/5401 often used must have a base stopper c. 470 to 1k and small loop areas in the base<>emitter circuit otherwise you will get problems. I've had two instances where they oscillated at >150MHz 😱
Click to expand...

Thanks for the info Bonsai, I cascoded the same transistors on the input of a Hafler 500 because of an article in TAA and since then the amp would seem to run hotter that before. The fan would be on a higher speed after a while. Now I see that it might have been oscillating at a very high frequency. I need to put the amp on the bench and check this out sometime.
 
#18,351
Tournesol said:
I was not asking about the industrial process but the design. Not the manufacture, but the R&D offices.
I had visited once those strange sterile white rooms with martians walking in it. Scary.
Click to expand...

Design is basically simulation based only, including macro cells, already characterized. The device/macro cells models - you don’t want to go there, it’s essentially at the core of a process development, a circuit designer is not necessary aware of the model internal details.

Layout and interconnection routing is a quasi mechanical process that is largely irrelevant today (at least for analog design).

There’s always a review process before triggering the reticle/mask manufacturing, which is an 1M+ cost point of no return.
 
Last edited:
#18,358
No, I was the computer operator. IBM 7094, 1401, 1620 from 1963-1967 '-) It would be interesting do be shown the development methods for IC's as it has changed over the decades. Still, IC's are limited by their power dissipation no matter what.
 

Attachments

  • Picture 34.jpg
    Picture 34.jpg
    176.5 KB · Views: 216
  • Picture 43.png
    Picture 43.png
    404.4 KB · Views: 221
Last edited:
#18,359
Mark Johnson said:
Maybe it might be a good idea to propose a TRADE: you do X for me, and in repayment, I will do Y for you.

So far your "proposal" has Y==Zero. Not a fair and balanced trade as far as I can see.
Click to expand...
If this message is for me, I just have a question. When somebody ask-you "what time it is ?" did you ask for a $ ?
Strange way to consider friendly exchanges on a forum. We are not in a market place.
 
Status
Not open for further replies.
Top Bottom