Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
one issue is that the microstrip line will heat up with ohmic losses. So a thicker copper thickness helps by reducing the line resistance.
Also there is a trade off. If you use a thick substrate the line can be wider, but there is more thermal resistance in the dielctric layer since it is now...
they operate in the "near field"--meaning the fields die off very rapidly with distance, and follow a different attenuation/distance formula. That is one difference between them and a "normal" antenna.
Perusing the fallout of the Nvidia GPU/AI conference today, i notice that Synopsis announced it purchased ANSYS, and is working to include AI LLMs into its software. Wonder if that includes HFSS?
I HOPE they keep supporting HFSS!
in many cases, a tapered line will give your virtually the same response as a series of quarter wave transmission line impedance transformers.
and there are plenty of papers out there that tell you the optimal impedance of each section. Look up the book by Matthei, Young, and Jones.
tapers...
not a realistic filter.
what is your input and output impedance requirements. You are driving it with a voltage source, instead of a signal generator with a source impedance. also how are you analyzing it with a load? is the load infinite impedance? 50 ohms?
Usually when i see a ripple like...
i am not following your circuit.
if you differentiate a positive going square wave leading edge, you get a positive going spike. If you differentiate the negative going square wave leading edge, you get a negative going spike. If you delay the first and recombine the two you get two spikes...
i would put a 3 dB attenuator at the amp output. then i would reconfigure the capacitor to add an inductor to make a 13 MHz resonator with a high Q. Then i would drive that resonator. Be careful, there may be very high voltages across the cap
yes i have a couple of those HP8662/3s still. Fractional N synthesizers before you could BUY a fractional N chip! HP was a great innovator!
Yes, the whispering gallery mode saphire resonators were the pinnacle, some smart australian company patented their use.
He was one of those rare genius type engineers, who thought it was their duty to bring along the next crop of engineers to take his place. a great guy, easy to get to know.
the genius of those noise degeneration synthesizers was, instead of finding the worlds highest Q resonators and somehow...
a little off topic, but i just learned that Zvi Galani, who pioneered a number of ultra low phase noise sythensizer designs, including this class of YIG Resonator frequency-discriminator phase noise reduction circuits, has just passed away peacefully. He holds a number of patents in this area...
it could be a LOT of things. but you are "Tuning" with the FM coil, and keeping a constant current in the TUNE coil?
Maybe you are overdriving the FM coil's current limit, and it is causing weird effects, like yig sphere heating?
I would typically drive the TUNE coild with a low noise D/A...
THIS would be a lot more "constant impedance" than your first circuit:
You might want potentiometers that are more like 75 to 110 ohms....
Optimise the impedance for the nominal attenuation value you want, such as perhaps 5 dB.
Usually you change the LOAD and SOURCE impedances to 500 ohms, then the graphical outputs plot it that way. Each program has a different way to do it though, you have to read the instructions
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
