John Curl's Blowtorch preamplifier part II

[morinix]

Originally Posted by john curl

Look at the topology of the AD797, for a good example.

Seems like these guys gave it the collage try.
AD797 Clone people! Who can make a better one.

[PMA]

Originally Posted by simon7000

Before I go further let's see if we can get agreement on a really simple circuit.

Pavel, Wavey, Scott, SY and anyone else interested, would you each please post an answer to what is the power dissipated in the resistor in the attached circuit?

5W.

[PMA]

Originally Posted by Steve Eddy

Average would be zero? Far as I'm aware, the average of 10 VRMS is 8.99 V.

se

Over half period of sine. Over one complete period it is zero as Ed has stated.

[RNMarsh]

Originally Posted by PMA

5W.

2.5W ?

[PMA]

No. 5W for 1kHz/10Vrms sine, 15.9uF capacitor and 10 ohm resistor. Please check the Ed's image, it is really a trivial task.

[VladimirK]

PMA is a winner.
Impedance modulus of the cap is 10 Ohms at 1kHz, but its vector direction is 90 deg down from 10 Ohms of the resistor active impedance. Total impedance modulus of the RC chain is SQRT(10**2 + 10**2)
From this we calculate current and power dissipated on R.

[simon7000]

Dick,

The reason why it is not just Xc/(R+Xc) is the phase shift when R = C is 45 degrees. I do want to go over vector math in a bit.

So who wants to show the Cartesian coordinate system with a 10 ohm resistor and equivalent capacitor value?

[PMA]

Ed, please do not. What fascinates me is the lack of basic knowledge even with renowned people. This is really a 'pub threat.

[john curl]

It isn't just that PMA, many of us are 'rusty' and have not done such mental 'manipulation' for many years. When I first started at Ampex in 1967, for example, I was given the task of manipulating Laplace transforms to generate phase compensation circuits for a NASA quality Instrumentation Recorder that operated up to 120ips. Then, I could do it, and it was fun! Today, I can't remember exactly what I did.
It is best not to get too 'smug' as to what it takes to design good analog circuits. This sort of 'engineering' knowledge isn't what is necessary.

[bcarso]

Originally Posted by PMA

Ed, please do not. What fascinates me is the lack of basic knowledge even with renowned people. This is really a 'pub threat.

When, in late 1989, after the previous business had failed, a recruiter had me interview with Harman. Besides a flock of interrogations with various vice presidents and managers, I had to take a written test. It was based on one administered to FAEs and others at Tektronix. Multiple choice (but show your work), pretty basic, with some discrete components, a tiny bit of Boolean algebra and other logic circuitry. The time limit was 30 minutes.

I was ushered to the cafeteria, which was initially quiet. Within five minutes it filled with line workers. I soldiered on. The very cold fish of a Human Resources person came to collect the thing at the end of the period. I could already tell that he disliked me intensely after our little chats.

I aced it, one of a couple who got a perfect score. When I went over it with Brad Plunkett, he asked me if I were sure of one of the Boolean questions, and when I said yes I was pretty confident, he marked that answer on the sheet he had

One of the questions was a cleverly worded one about the voltage at the output of a C-R voltage divider with sinusoidal excitation. Many many people got that wrong, including a VP from Harman-Motive, who wiggled and wriggled to insist that he was right, according to BP.