Discrete Opamp Open Design

[kgrlee]

Originally Posted by PHEONIX

Do you have a copy of Dimitri's "Enhanced cascode Article".

I don't Pheonix. I'm new to diyaudio. A link would be appreciated or a PM. Thanks

[aparatusonitus]

Send a pm with your email and I'll pass the article.

[linuxguru]

Originally Posted by luvdunhill

Interesting, I'd like to play with this one as well, if that would be okay. I could probably even do a two sided version in a SIP-8 dual opamp format, similar to my layout. I could volunteer to look at higher output bias version.

Sure, you're welcome to try out anything. I'm guessing that a high-bias version that's able to drive low-Z loads will require through-hole TO-92 or TO-126 output devices, and the board size is a constraining factor. Presently, I'm at 14mm x 14mm using the Rohm IMX2/T2 duals, and the 2sk209 singles. I was able to route everything on one layer without vias, but had to remove pads 1 and 5 from the DIP8 footprint to create space.

It is conceivable that a 2-layer SMD DIP8 dual opamp is within reach, but it will be slightly larger because pads 1 and 5 are required.

[luvdunhill]

one last question. I'd like to optimize an incarnation for use in Salen-Key filters. Is there a good rule of thumb in terms of load that approximates a wide variety of filter topologies? I was thinking that something in the 30K range would be normal (with a spike where you'd expect it) and was using 4.7K in my simulations. I thumbed through Self's book on the subject last night and didn't see anything in the opamp section where he goes through a bunch of current parts and discusses their relative merit. I have a feeling this answer is more difficult than what I'm asking for, but go easy on me

[bobaruni]

Originally Posted by luvdunhill

Does anyone have a recommended complementary SOT223 device?

I'm using 2SC5964 and 2SA2125 as they are the highest power rated devices in this package I have found so far and also have a huge Ft and decent Hfe but I don't know how advantageous any of this is. See some others from my list (all SOT223)


2STN1360 2STN2360 60v 3A 1.6W 130MHz 80 400
BCP55 BCP52 60v 1A 2W 150MHz 100 250 linear to 100mA, designed for AF
BCP56 BCP53 80v 1A 2W 150MHz 100 250 linear to 100mA, designed for AF
BCX55 BCX52 60v 1A 1W 150MHz 100 250 linear to 100mA, designed for AF
BCX56 BCX53 80v 1A 1W 150MHz 100 250 linear to 100mA, designed for AF
FTZ692B FTZ792A 70v 2A 2W 160MHz 300 800 linear to over 100mA
FTZ651 FTZ751 60v 3A 2W 140MHz 100 300 linear between 10-200mA

[dchisholm]

Originally Posted by luvdunhill

Output THD @ 20kHz (according to LTSpice)

I'm sure you adjusted resistor values so you compared the transistor pairs at the same collector idle currents.

I still don't have a good intuitive feel for the parameter interactions in the cross-coupled output stage of SW-OPA (see atch image)
.
I recall reading somewhere that the output drivers (Q14 and Q58 in Post #1568) should be faster than the actual output devices (Q56 and Q69) for frequency stability. But for lowest distortion (and perhaps thermal stability) the seat of my pants says they should be the same as the output devices.

Rather than looking for output devices that minimize distortion, should we be looking for pairs that work best in this topology, i.e. consider the compound-connected Q56/Q58 (and Q14/Q69) as single devices to be optimized?

Dale

[linuxguru]

Originally Posted by bobaruni

BCX56 BCX53 80v 1A 1W 150MHz 100 250 linear to 100mA,

This pair is also available in SOT89, smaller than SOT223, and with metal tab.

[scott wurcer]

Originally Posted by dchisholm

I'm sure you adjusted resistor values so you compared the transistor pairs at the same collector idle currents.

I still don't have a good intuitive feel for the parameter interactions in the cross-coupled output stage of SW-OPA (see atch image)

I recall reading somewhere that the output drivers (Q14 and Q58 in Post #1568) should be faster than the actual output devices (Q56 and Q69) for frequency stability. But for lowest distortion (and perhaps thermal stability) the seat of my pants says they should be the same as the output devices.

Rather than looking for output devices that minimize distortion, should we be looking for pairs that work best in this topology, i.e. consider the compound-connected Q56/Q58 (and Q14/Q69) as single devices to be optimized?

Dale

It's really just a diamond that bootstraps itself. Any stability issues would mainly be with Q14 and Q58 looking back into the emitters of the output devices which can look inductive as you approach Fbeta. The output stage breadboard had no problems, and at 8mA bias and 600 Ohm load still only had .01% distortion.

The Sziklai pair has feedback and in my experience is much more prone to oscillate, but the added gain typically reduces the distortion by a larger factor.

[bcarso]

Originally Posted by scott wurcer

The Sziklai pair has feedback and in my experience is much more prone to oscillate, but the added gain typically reduces the distortion by a larger factor.

My experience as well with Sziklai. And the poorer ability to turn off an output device translates into a dynamic asymmetry that is somewhat bothersome at high frequencies. Bryston has used a triple compound that they have managed to make work, but it looks very temperamental in simulation.

[bcarso]

Originally Posted by luvdunhill

one last question. I'd like to optimize an incarnation for use in Salen-Key filters. Is there a good rule of thumb in terms of load that approximates a wide variety of filter topologies? I was thinking that something in the 30K range would be normal (with a spike where you'd expect it) and was using 4.7K in my simulations. I thumbed through Self's book on the subject last night and didn't see anything in the opamp section where he goes through a bunch of current parts and discusses their relative merit. I have a feeling this answer is more difficult than what I'm asking for, but go easy on me

Look at signal-to-noise at a given frequency to determine how sensitive it will be to impedances, and then make the determination about how the filter components will load the amp output. Then fold in the loading thereafter.

Every situation is different, and I know of no rule-of-thumb. Lower Z is always lower noise, but depending on where the circuit is in the system you may be well-dominated by noise in the input signal already, so there may be no reason for heroic measures in the filter.