Timeline for Concept of transistor oscillator
Current License: CC BY-SA 4.0
19 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 15, 2021 at 16:34 | history | edited | nanofarad | CC BY-SA 4.0 |
added 72 characters in body
|
| Apr 15, 2021 at 16:33 | comment | added | nanofarad | @LvW (1) I don't agree that counting the summing junction's phase shift as part of the loop gain is in contrast to the figure. Using the stable amplifier example as a reference: I opened the loop at an arbitrary point, and the transfer function from start to end is the amplifier (positive near DC), feedback network (positive near DC), -1 from summing junction with a negative sign. (3) - see 1; Final comment - my circuit is a simplification of ieeexplore.ieee.org/abstract/document/8449103 that removes inductor coupling; I don't believe the full details are germane or useful to the OP | |
| Apr 15, 2021 at 15:44 | comment | added | LvW | nanofard, (1) Absorbing the inversion of the summing junction is in contrast to your figure (showing the loop opening for demonstrating loop gain definition, (2) OK, agreed, I did not see the footnote, (3) see (1). Final comment: I think, the BODE diagram of this unspecified "microelectromechanical device" is confusing within the context of this discussion. You speak about a Colpitt oscillator using a feedback block with 0 deg phase shift? What about the phase shift of the gain stage? | |
| Apr 15, 2021 at 14:03 | history | edited | nanofarad | CC BY-SA 4.0 |
added 240 characters in body
|
| Apr 15, 2021 at 14:00 | comment | added | nanofarad |
@LvW responding to your other comment: 1. I use a notation where the 180 degree phase shift of the subtractor is absorbed into the loop gain as plotted, much like it is presented in Cadence stb analysis. We ideally want 180 deg at low frequencies for the stable amplifier, and zero for oscillation. 2. See footnote. This is a conceptual simplification because the answer is already too long to discuss loading and bilateral feedback in any strong detail. 3. Adjusting wording, but I continue to absorb the inversion into the loopgain phase.
|
|
| Apr 15, 2021 at 13:57 | comment | added | nanofarad | @LvW The oscillator has an external resonator in the feedback loop, but the resonator is not a crystal, but rather a different (unspecified) microelectromechanical device which can be fabricated and integrated more easily. | |
| Apr 15, 2021 at 11:08 | comment | added | Daniel | But it's interesting that when you reduce the 100 ohm resistor above the collector to 10 ohm it no longer oscillates (and the transistor always remains FA). | |
| Apr 15, 2021 at 11:07 | comment | added | LvW | Daniel, in your Colpitt circuit it is very easy to see the 3rd-order lowpass in the feedback loop: R-C2-L-C1 in ladder structure. The oscillation frequency results when the phase crosses the -180deg-line (which together with the invereting gain stage gives 360 deg). | |
| Apr 15, 2021 at 11:01 | comment | added | Daniel | Thanks @LvW approved the answer in part based on your feedback. By the way I added some diagrams to my question to show what I had in mind (you can click on them to expand them full screen). (Actually one of the images was wrong, but updated just now). | |
| Apr 15, 2021 at 11:00 | vote | accept | Daniel | ||
| Apr 15, 2021 at 10:04 | comment | added | LvW | Daniel, with the exceptions of my small corrections (see my comment) I think the contribution from nanofarad is the best answer to your basic questin. One further comment: The second BODE diagram (with two phase crossings) belongs to an oscillator with a crystal in the feedback loop. This should be mentioned. | |
| Apr 15, 2021 at 9:58 | comment | added | Daniel | @nanofarad and thanks for the effort of drawing diagrams and graphs. | |
| Apr 15, 2021 at 9:50 | comment | added | Daniel | @nanofarad thanks for the detailed answer. | |
| Apr 15, 2021 at 8:49 | comment | added | LvW | I like to mention some errors and contradictions: (1) Three lines above the first BODE-diagram: We want a loop gain of 0 deg (360 deg), NOT 180 deg. (2) We must NOT "break the loop at any point". We must use a node where a small ouput resistance meets a large load resistance. Otherwise, we must mirror the disconnected load. (3) The referenced Colpitt oscillator uses an inverting amplifier. Hence, it is not correct to say "the phase reaches zero degrees". In contrast, the feedback circuit must NOT operate at "resonance". We need -180 deg provided by a 3rd-order lowpass in ladder topology. | |
| Apr 15, 2021 at 4:39 | history | edited | nanofarad | CC BY-SA 4.0 |
added 166 characters in body
|
| Apr 15, 2021 at 0:29 | history | edited | nanofarad | CC BY-SA 4.0 |
added 159 characters in body
|
| Apr 15, 2021 at 0:23 | history | edited | nanofarad | CC BY-SA 4.0 |
added 159 characters in body
|
| Apr 15, 2021 at 0:18 | history | edited | nanofarad | CC BY-SA 4.0 |
added 159 characters in body
|
| Apr 15, 2021 at 0:13 | history | answered | nanofarad | CC BY-SA 4.0 |