Bob Cordell Interview: Error Correction

Jump to Latest
#1,883
#1,892
Fanuc said:
Forr,

Any chance of getting a copy?

From what I remember it wasn't an article with EE's giving it a rigorous analysis. I think it showed Bob's amp.

I had it somewhere but lost it. EW's go AWOL from the company.

Thanks

Kevin
Click to expand...

Oops! I forgot to include my email.

Kevin.dabson(@)gmail.com - Please remove the brackets ()

Kevin
 
#1,894
I asked for it as well.
 

Attachments

  • e.png
    e.png
    768 bytes · Views: 498
#1,895
Correction

Bob,

In respect of your ''back-of-envelope'' calculations here you state:

We now work our way forward through the bottom set of transistors. Q6 will be off with a reverse base-emitter bias of about 1.4V, since both ends of R10 are sitting at the output voltage level of +16.0.
Click to expand...

This is incorrect, as Q6 only stops conducting when the output swings beyond roughly 25V.

Given below is the corrected prose as i think it ought to have been:

Assume that the amplifier delivers 2A into 8ohms placing the output at +16V.

Now work your way backward through the conducting path comprising the upper set of transistors. Resistor R12 drops 0.66V thereby placing +16.66V at Q9’s emitter.

Transistor Q9’s base is then at +17.26V with Q7’s base at +17.86V. Transistor Q5’s base is then roughly at +18.46V.

The bias network needed all along to have a fixed drop of about 3.6V. This means that the base of Q6 must be at about +14.86V.

We now work our way forward through the bottom set of transistors.

Transistor Q6’s emitter is then at +15.46V, while the top end of R10 sits at the output voltage level of +16.0V.

The base of Q8 is at about +15.46V also, and, ignoring insignificant base current, resistor R10 therefore passes 1.63mA.

Now, the top end of Q8’s emitter resistor, R11, is pulled to +17.26V by Q7, while Q8’s emitter is roughly at +16.06V. The voltage drop across R11 is 1.2V and, being 56ohms, therefore conducts about 21.4mA.

Thus, output transistor Q10 is off with a reverse base-emitter bias of about 0.06V, while all the drivers operate in Class A for the specified output of +16V.

Transistor Q6 only turns off when the output swings beyond +25V, while Q7/Q8 remain in Class A.

If, however, resistors R9/R10 are disconnected from the output rail and shorted together, then Q5/Q6 will also operate in Class A regardless of output voltage swing.
Click to expand...

I suggest running a DC Sweep of the circuit in question; this is less error-prone than ''back-of-envelope'' calculations.
 

Attachments

  • drivers.png
    drivers.png
    66.6 KB · Views: 530
#1,896
Re: Correction

mikeks said:
Bob,

In respect of your ''back-of-envelope'' calculations here you state:



This is incorrect, as Q6 only stops conducting when the output swings beyond roughly 25V.

Given below is the corrected prose as i think it ought to have been:



I suggest running a DC Sweep of the circuit in question; this is less error-prone than ''back-of-envelope'' calculations.
Click to expand...

Good catch, Mike. You're right. I wrongly had it in my mind that the voltage on the base of Q8 was reverse biasing it instead of forward biasing it. I'm a big believer in both back of the envelope and simulation, as the former provides insight and teaching (when stupid mistakes like mine are not made) and the latter provides mindless precision (as long as the netlist is entered properly).

Bob
 
You must log in or register to reply here.
Top Bottom