New topology I have revealed only a part. Stability is very important thing. My topology is built on only two stages. Each stage of the phase rotated by 90 °. Two stages can not turn around phase 180 °, and have been gain. Engagement can never swing.
Involvement is very fast. Any error in the output, the antenna, any error in the input circuit, the antenna. There is danger of an additional phase shift. Arises interlinked and involvement can oscillate. Spatial arrangement of my topology is very important.We discussed the voltage dependence.
I returned to the original topic. We discuss the frequency dependence.
Me used equivalent circuit of the transistor.
The decrease in gain by 10%. It can be deducted on RC about charts. You can see tension and angle.
Topology Federmann not much different involvement. Federmann topology is different working conditions and working points. Topology Federmann may look like Starter, but the opposite is true. Topology Federmann also want to address not in this thread.
* = +3dB
PLS, Federmann, +3dB is +3dB everywhere, why are you used *** symbols ??
.
Federmann, why don't you rather build your amplifier?? You are wasting too much time here. Your amplifier soldering and construction is as slow as a snail. Show us finally a working sample with full power BW measurements, slew rate measurements (into load, not into 1Mohm) and distortion measurements. After that, we can make some listening comparisons. Instead of that, you are wasting much time here and repeating infinitely same images and reasoning, based on pseudo-theories.
Back on topic ? OK: your amp have not two stages, but two stages + source follower. Mosfets are very fast devices and they can very easy to oscilate, especially by long driver trace. It you can read at every design manuals and I see on your PCB, that you quite forget on it...Which is your answer ?
Same images can be measured oscilloscope. There faces thread. It is necessary to establish boundaries. That's why I founded the thread. Remember when I said John Hidden, that is delayed measure Lissajouss shape. He would get a similar result.
I do the same charts for many years measured. At home I have no equipment and can not continue. Therefore, only work on PC.
Remember that every theory must be proven in practices, otherwise it is useless. Models do not cover real world, they are just models and nothing more.
You have not given yet ANY proof neither on "Influence of the delay amplifiers for listening characteristics", nor on your declarations on Vdiff phase angle at 100KHz.
You have not given yet ANY proof neither on "Influence of the delay amplifiers for listening characteristics", nor on your declarations on Vdiff phase angle at 100KHz.
Federmann, probably for the last time, I'll try to give you an advice. Just today, today afternoon, I was evaluating a modification of my 8-years old mosfet power follower for the guys in another thread here on diyaudio. The follower device is IRF350, supplied by CCS. Without gate stopper placed directly at the IRF gate, that circuit would oscillate near to clipping at hundreds of MHz. No simulator software would reveal that!!
And please do not bother me with your school-like plots. I know them better than well.
Bye, there is no help for you,
And please do not bother me with your school-like plots. I know them better than well.
Bye, there is no help for you,
Last edited:
Yes Vasek, that's very important, but even more - MOSFET simulator models are quite incomplete and do not fit to real devices well. This is something Federmann has no idea about. And no wonder - he does not build amplifiers, he, just he is the best example of an armchair designer
Many times I wrote that this thread is not about topology Federmann, many times I wrote that I did not disclose everything.
For many years, I stopped by engagement, measured and counted on paper. For many years I know exactly what can be neglected, and what not. SPICE-compatible is my assistant not my master. Never a really, trial and error. Measurement always match simulations.
This thread is the border of what I published in charts. Let's stay on topic please. Let's talk about borders, derived from the α phase.
For many years, I stopped by engagement, measured and counted on paper. For many years I know exactly what can be neglected, and what not. SPICE-compatible is my assistant not my master. Never a really, trial and error. Measurement always match simulations.
This thread is the border of what I published in charts. Let's stay on topic please. Let's talk about borders, derived from the α phase.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.