That's logical for them to show up in discrete designs especially around RF. The Jensen opamp was my first exposure to inductor degeneration of a LTP. In reading the journals there was discussion about reducing the transconductance with resistors, or using JFETs, but i guess most of the design effort was related to making ICs so inductors were not a go to component. I was not familiar with the Burwen cite... So thank you and WJ.
Jack Sondermeyer who I worked with at Peavey was an application engineer for RCA back around then... I could ask him about early transistor design but I don't know if he is still alive... He had a stroke over ten years ago, and didn't do email or phone calls after that.
Reg Williamson the big dog tube amp guy talked about slew rate and NF stability criteria in connection with his design for WWII radar receivers.
My dad's WE notes are all tube stuff from 30s. Lots of inductors but mostly greek to me. I would expect tubes to be lower transconductance (like fets) so not in need of input stage LTP degeneration. Not that a circuit designer back then would ever throw that many active parts into making a single gain stage.
JR
Cancel the "cancel culture", do not support mob hatred.
mediatechnology wrote:Scott: Your re-posted sch looks fine though you'll probably want to add a Cc and Cfb around the 5534
That's one of those things I'm always too lazy to draw in when scratching on a napkin, as I think of it as a given. But thanks for reminding me.
One additional thing you can do is bypass the 5534 differentially with a single C rather than C3/C4.
That way the Qs don't have to slew current to charge them.
I've never done that - hmmmm... What math does one use to determine how much C when doing it differentially? I did read some of the "balanced output" thread where it was mentioned.
That's one of those things I'm always too lazy to draw in when scratching on a napkin, as I think of it as a given. But thanks for reminding me.
Yeah me too. None of my circuits ever show power connected or bypass.
I would just start with a 100 nF from Vcc to Vee. In the input-capacitorless preamp I also use electrolytics. Since there's a constant 30V (or so) across the capacitor when connected differentially there's not a lot of current that needs to be slewed. The cap itself is bootstrapped at both ends.
I built the 5534/BD139/BD140 circuit.
With +/-31V supplies (the max on that particular PSU) I was able to get ~ +24 dBu unloaded.
I set the dividers to provide 30V across the 5534.
The total Vce Sat is large so the circuit requires high power supply voltage for the peak actual output delivered.
Raising the supply voltage of the 5532 to 40V I can get 40V P-P (+25 dBu and change) with +/-31V supplies.
The bias resistor values were 4K7/10K.
I suspect that with +/-36V rails MCI was able to get maybe a +26 to +27 dBu clip point.
I recall that this circuit was primarily used in the 500-series EQ to replace the discontinued HV Harris "2002."