Op Amps: Diff Amp Phase Splitter

[@hieuo3073]

what is software simulation ?.Thank you brooo

[@ElectronicswithProfessorFiore]

TINA-TI

[@DarkSceptile99]

Hi, due to matching error in discrete devices and the difficulties in keeping the two bjt at the very same temperature, I believe it is a bit unadvisable building a differential pair out of discrete devices. This could be easily done while designing integrated circuit, since equal junction temperature is guaranteed and matching error is greatly readuced placing the bjt close as possibile. But truth to be told I never tried that in practice. I'm curious whether it is possible to build a working differential pair out of discrete component. Thank you!

[@ElectronicswithProfessorFiore]

Most definitely. In fact, at one time, diff amps were made using tubes (and called "long-tailed pairs"). It is true that you won't get the sort of matching that you'd get using integrated devices these days, but you can most certainly make diff amps using discretes, and get good performance.

[@chrisw.3077]

Maybe you familiarize yourself with the dual transistor.

[@DarkSceptile99]

Thank you for the tip! I didn't know there were ICs as "Matched dual transitors", that is definitely super cool!

[@theminertom11551]

I have a question and I hope that this is the forum to ask this. Prof Fiore, if I connected Cout1 between the two collectors of the diff pair, can we derive a a frequency response analysis? I would expect that if the voltage at the input of T1 was higher than AC GND at T2, then the phase would shift to "phase lead". On the other hand, if the input of T1 was below AC GND then the path would shift and the T2 path would introduce a phase lag. I have seen this circuit work as a phase shift network for a Colpits type oscillator in a clock recovery network. In any case, it would make an interesting analysis.

[@ElectronicswithProfessorFiore]

I'm not quite sure what you're proposing/asking, but in general whenever you insert a reactive element into a signal path (cap or inductor), there's going to be a phase shift, and whether or not it's lag or lead will depend on the configuration (a cap could be either). Part of my uncertainly is when you describe the signal at T1 being higher/below AC GND at T2. Half the time it's above and half the time it's below. Are you suggesting that the frequency response would be different for the positive vs. negative half cycles? Ultimately, I suggest simulating the altered circuit and see what it does.

[@theminertom11551]

@@ElectronicswithProfessorFiore Professor, thank you for your reply. I really mean it! It is not easy these days to find good advice concerning interesting analog circuits, these days. In terms of full disclosure, someone gave me this circuit about 30 years ago at a company that I worked at and he is long dead, as are a lot of the older analog people. Would it be possible to send you a jpg of the circuit in question? I can't insert one within the confines of a youtube comment section. All that I am looking for is a way to determine, analytically, a "phase response" (vector diagram or other) of a diff pair with a capacitor between the two collectors. Every 10 years or so I do a search for current data on this type of a circuit and I have never found any. If either end of the cap would be connected to AC ground, the analysis would be easy, but neither is. This is, in fact, an interesting circuit.

[@ElectronicswithProfessorFiore]

@@theminertom11551 Simulation should be fairly straight-forward. There are multiple free simulators out there. I use TINA-TI but I've heard from folks using Linear Tech's LTSpice or MicroCap.

[@rudygomez6996]

cool