class A amplifier

Oct 13, 2008 134 Replies

I wouldn't bet on that.

I've fixed 3 'runaways' including one in marine radar. I could tell you a story or two about that. I turned a turkey into a significant commercial success. There was a restructuring going on at the time. The Technical Director was being promoted to Managing Director, his Technical Manager was becoming Technical Director and quite a few of the R&D staff suggested I should apply for the open Technical Manager post (staff of about 40).

I was VERY flattered.

Graham

So prove it. If, in fact, discrete analog design is one of your strong points, you\'d know that if you\'re introducing some nuance into a previously proven design and, therefore, don\'t know about the nitty-gritty of the combo and/or you don\'t know how to simulate it properly, Then you need to build a few a of them, test them over the expected changes in supply voltage, ambient temperature, and load impedance, and report your findings. JF

Well you beat me there by a factor of 3 or so.

Graham

concepts like using

client's copyright.

clot handed them over

The buck stops where? JF

concepts like > using

client's > copyright.

could see loads of

If frogs had wings... JF

being

Don't let that go to your head, over here, that's one method of getting rid of some one.

formatting link
"

concepts like >> >

client's >> >

could see >loads of

GGF. The MD and I were discussing dismisalls today. Trouble is, I just took on another contract on Monday. His timing is poor. Poor chap, he's always had the wool pulled over his eyes by idiots who never deserved the positions they had.

Graham

Ok so I can start from Fig 1A

formatting link
with an op amp instead of the entry stage ? And with a 2N3055 and a MJ2955 ?

Olivier

Then you

I disagree with this. If one can obtain a decent BSim3 model of the lateral fets, then an empirical test is pretty much irrelevant. The design of SOC analogue ics, with 10'000s analogue devices, entirely in simulation, that work, is routine. In fact, bench work is not even feasible. Indeed, using worst case corner tools (

formatting link
and Monte Carlo simulation allows one to far better ensure that products will work reliably in production than a few bench tests, with nominal devices, can possible hope to achieve.

Kevin Aylward

formatting link
formatting link
formatting link
-SuperSpice

Other question Graham ... Suppose I work in class A with a current of 1A in the two output transistors. If the current of one transistor goes to 1.5 A the other one should go to 0.5 A and 1A goes through the load. If I forget to connect the load (and I am sure I will), what is going to happen ? Thanks

Olivier

concepts like >> >

the client's >> >

could see >loads of

another contract on

eyes by idiots who

So you think that it would be OK for you to pull the wool over his eyes because you\'re more deserving of the position? JF

I\'d agree had you not snipped a little early. Here\'s the whole thing: " If, in fact, discrete analog design is one of your strong points, you\'d know that if you\'re introducing some nuance into a previously proven design and, therefore, don\'t know about the nitty-gritty of the combo and/or you don\'t know how to simulate it properly, Then you need to build a few a of them, test them over the expected changes in supply voltage, ambient temperature, and load impedance, and report your findings." JF

Ok, I'll buy that. I didn't really notice the full "or you don't know how to simulate" bit. The sentence was just a tad too long for me to take it all in.

Kevin Aylward

formatting link

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required