This has to be a great method to get real understanding of transmission lines, distributed components, matching networks and all that sweet and lovely microwave knowledge
Stubs can be added, and sliced across to add capacitance with reactance. Sliding stubs along the stripline, to find a good match, then tuned by sllicing across the stub, which adds capacitance, and you can continue the process to impedance match, and bandwidth.
If you check out the EEVBLOG channel, the guy likes to tear down scopes. I think it might have been an spectrum analyzer that he tore apart that had an RF module that was loaded with circuits like that. They actually had capacitor shaped (like the schematic symbol) traces that actually were RF capacitors. Without breaking your trace, you might be able to build a couple of capacitors in parallel with your 50 ohm xmission line.
Hi, you could add a short circuit and measure the impedance at some frequency and then measure the impedance with the open circuit at the same frequency. In case of short circuit your Zin_sc=jZoTan(Bl) and for open circuit Zin_oc=-jZo CoTan(Bl) Then Zin_oc *Zin_sc =Zo^2 or Zo=SQRT(Zin_oc *Zin_sc) So Multiply short circuit impedance by open circuit impedance then take square root of it , that is the impedance of your transmission line. Also that "Funny circuit ie. cross connected line" will gives you a tuned circuit at some frequency. They use it at microwave as a LC tank. The bow tie is called "tapered line" is for broadband matching of transmission lines. A good reference; though it is old: Lines, Waves and Antennas The Transmission of Electric Energy, 2nd Edition by Robert Grover Brown, Robert A. Sharp, William Lewis Hugues, Robert E. Post John Wiley & Sons copyright 1961, 1973
With distributed elements, it seems that anything along the path is an inductor (like right-hand rule), and anything perpendicular to the path is a capacitor. I think the bow ties are popular because it allows the capacitor to join the circuit--avoiding impedance mismatches which cause reflections. It seems that abrupt trasitions are bad for PCBs, as far as signal integrity is concerned. And still, everything you don't want to be is resistor and an antenna : )
Small copper squares slid along the stripline, using a toothpick...sliding along stripline to find a match, then try this over again until the entire line matches 50 Ohms at the operating frequency.
What happens if you tack solder the bow ties to the main stripline? I assume that their only contact is through coupling and the adhesive is acting as an insulator? I will admit I know or remember nothing from my stripline course in college. I will see if I saved the book
Thanks for the strip line Z demo. I can't wait to buy the mini VNA... Fred go read about the Nano.. YES Please play IMASI Guy... Thanks for sticking to the little VNA...wow it's capabilities are amazing. I know you showed some earlier comparisons to the reference HP big boy..Big Boy can be used for verification..& comparison For some reference...I am thinking of Motorola RF databook...at least they show some near micro wave amplifiers ...and dimensions of some strip lines ...and they R & C component part numbers dimensions... Their stripline dimensions could be ran through the calculator And every transistor usually has Smith Chart shown displaying collector and emitter Z... It be nice if ...sort of like advanced IMSAI Guy graduate class..perhaps you found a small signal transistor... Safe ...lower power & low Vcc Make a 8 dB gain amplifer 10 mW...padded output with your home made attenuators and triming to prevent fried front end Nano... Demo if you can match or measure what OEM claim.. It look like you already made some nice 50 ohm attenuators already built ... I was always amazed by RF.... The scattered parameters seemed like more black magic...it will be good to revisit... I remember Al...our RF class teach had some generator fed to 1.5 foot long waveguide and terminated with open short and matches...a study in VSWR... I need to go look up the index of IMASI Guy already published lessons and demos.. Like the cat eye O-scope...BER ..sort of tells use something about matching source and load How about a ECL circuit design...build strip line...send a digital stream...see if you run w/o error near their max claimed max toggle frequency... I remember pops playing with some TTL and CMOS on our kitchen table 1976 era while he worked on PLL for Lo in the home build CW HF...his complaint being the TTL wouldn't toggle much behind 5 MHz and supposedly should run up to 20 MHz per the little written Radio Shack data package... now I can see he probably wasn't driving them properly... I can see you working with Gunn diodes maybe a simple parabolic dish X transmitter...reciever...do a field test..to get out of work shop a tiny... Have a neighbor 2 blocks down your street recieve this line of site microwave signal then demodulate HD video....you be in the home lab giving a lesson..transmitting....neighbor allows space for down converter...then your lesson of that day transmitted recodered video is posted to us RUclips fans.. I think the images would be just as good and clear as your present camera set up...look at it is like a system build lesson...visit Hartley laws Oh Happy pre 4th of July Kind regards & thanks again I know there will great things coming...may $ & happyness rain down the from the sky on IMASI Guy Fred
The center of the first circle was the Zo of TL. You were right, the wide TL had a Zo < 50 Ohms. When you made it too narrow, Zo > 50 Ohms, which was the Center of that circle.
I used to think the center of the circle was the Zo but my good friend did the math and got: Zo = sqrt( ( Rmiddle*3*50^2 - 50^3 ) / ( 3*50 - Rmiddle ) )
@@IMSAIGuy the center is just an arbitrary reference point. 50 Ohms is a common Zo, so it is often used as the center, but the Zo of your microstrip can be anything. In a microstrip RF filter, you would select the Zo of the different segments to get the frequency response you want (lengths too of course).
What frequency range did you make the stripline for? What is the dielectric constant of the board material? FR4 glass epoxy is normally not used for microwave stripline construction. Rogers RT Duroid is a 'standard' that I became accustomed to use. Just my 2 cents.
I was wondering about the tape. I am assuming that it is copper tape with adhesive on the backside. Would the adhesive act as an insulator? So if you stuck one piece of tape on top of another, would you have to solder them together to make a connection?
I don’t claim to know much about the magic of RF, but this video makes me wonder what would happen if you simply took a piece of 50ohm coax and sliced it longways and basically unwrapped it (leaving the coax wrapper and center conductor intact) and glued the entire filleted piece of coax onto any non-reactive surface so as to make the insides of the cable accessible for insertion of elements.
IMSAI Guy I’m an ee student struggling to understand why when I terminate my microstrip with a matched load i get predictable results but when I just connect both ports to VNA I get reflection and weird curls on the polar/smith plot (the microstrip ~should be~ impedance matched). We’re also supposed to calculate the width and length of the microstrip somehow based on frequency but I thought the width of the microstrip was set?
@@eastonbryan4171 the section of microstrip should act as a transmission line. so to test your setup. put a short piece of coax in place of your microstrip and measure it. it is a known good transmission line. if you then put your microstrip in and it looks bad then it is bad. the width of the strip is a variable as is many other mechanical dimensions. there are many programs to calculate the correct dimensions. I don't how your professor wants you to do this
This has to be a great method to get real understanding of transmission lines, distributed components, matching networks and all that sweet and lovely microwave knowledge
This is *great* for prototyping microstrips without etching a PCB. Thanks!
Stubs can be added, and sliced across to add capacitance with reactance.
Sliding stubs along the stripline, to find a good match, then tuned by sllicing across the stub, which adds capacitance, and you can continue the process to impedance match, and bandwidth.
If you check out the EEVBLOG channel, the guy likes to tear down scopes. I think it might have been an spectrum analyzer that he tore apart that had an RF module that was loaded with circuits like that. They actually had capacitor shaped (like the schematic symbol) traces that actually were RF capacitors. Without breaking your trace, you might be able to build a couple of capacitors in parallel with your 50 ohm xmission line.
the guy at 'the signal path' show lots of that stuff
Hi, you could add a short circuit and measure the impedance at some frequency and then measure the impedance with the open circuit at the same frequency.
In case of short circuit your
Zin_sc=jZoTan(Bl)
and for open circuit
Zin_oc=-jZo CoTan(Bl)
Then
Zin_oc *Zin_sc =Zo^2
or
Zo=SQRT(Zin_oc *Zin_sc)
So Multiply short circuit impedance by open circuit impedance then take square root of it , that is the impedance of your transmission line.
Also that "Funny circuit ie. cross connected line" will gives you a tuned circuit at some frequency. They use it at microwave as a LC tank.
The bow tie is called "tapered line" is for broadband matching of transmission lines.
A good reference; though it is old:
Lines, Waves and Antennas The Transmission of Electric Energy, 2nd Edition
by Robert Grover Brown, Robert A. Sharp, William Lewis Hugues, Robert E. Post
John Wiley & Sons
copyright 1961, 1973
The "bowtie" is called a radial stub. Its meant to have a radiused end. It's just a wideband 1/4 wave stub.
OK, that makes sense
Capacitance ... the bow ties are used to tune a filter for broadband.
With distributed elements, it seems that anything along the path is an inductor (like right-hand rule), and anything perpendicular to the path is a capacitor. I think the bow ties are popular because it allows the capacitor to join the circuit--avoiding impedance mismatches which cause reflections. It seems that abrupt trasitions are bad for PCBs, as far as signal integrity is concerned. And still, everything you don't want to be is resistor and an antenna : )
Small copper squares slid along the stripline, using a toothpick...sliding along stripline to find a match, then try this over again until the entire line matches 50 Ohms at the operating frequency.
What happens if you tack solder the bow ties to the main stripline? I assume that their only contact is through coupling and the adhesive is acting as an insulator? I will admit I know or remember nothing from my stripline course in college. I will see if I saved the book
The copper tape has conductive adhesive. you can test with a multimeter
@@IMSAIGuy that's good to know. Let me know where you picked it up from. It might be fun to play with here
@@mjrtude go to amazon and search for 'copper tape conductive adhesive'
Thanks for the strip line Z demo.
I can't wait to buy the mini VNA...
Fred go read about the Nano..
YES Please play IMASI Guy...
Thanks for sticking to the little VNA...wow it's capabilities are amazing. I know you showed some earlier comparisons to the reference HP big boy..Big Boy can be used for verification..& comparison
For some reference...I am thinking of Motorola RF databook...at least they show some near micro wave amplifiers ...and dimensions of some strip lines ...and they R & C component part numbers dimensions...
Their stripline dimensions could be ran through the calculator
And every transistor usually has Smith Chart shown displaying collector and emitter Z...
It be nice if ...sort of like advanced IMSAI Guy graduate class..perhaps you found a small signal transistor...
Safe ...lower power & low Vcc
Make a 8 dB gain amplifer 10 mW...padded output with your home made attenuators and triming to prevent fried front end Nano...
Demo if you can match or measure what OEM claim..
It look like you already made some nice 50 ohm attenuators already
built ...
I was always amazed by RF....
The scattered parameters seemed like more black magic...it will be good to revisit...
I remember Al...our RF class teach had some generator fed to
1.5 foot long waveguide and terminated with open short and matches...a study in VSWR...
I need to go look up the index of IMASI Guy already published lessons and demos..
Like the cat eye O-scope...BER ..sort of tells use something about matching source and load
How about a ECL circuit design...build strip line...send a digital stream...see if you run w/o error near their max claimed max toggle frequency...
I remember pops playing with some TTL and CMOS on our kitchen table 1976 era while he worked on PLL for Lo in the home build CW HF...his complaint being the TTL wouldn't toggle much behind 5 MHz and supposedly should run up to 20 MHz per the little written Radio Shack data package... now I can see he probably wasn't driving them properly...
I can see you working with Gunn diodes maybe a simple parabolic dish X transmitter...reciever...do a field test..to get out of work shop a tiny...
Have a neighbor 2 blocks down your street recieve this line of site microwave signal then demodulate HD video....you be in the home lab giving a lesson..transmitting....neighbor allows space for down converter...then your lesson of that day transmitted recodered video is posted to us RUclips fans..
I think the images would be just as good and clear as your present camera set up...look at it is like a system build lesson...visit Hartley laws
Oh Happy pre 4th of July
Kind regards & thanks again
I know there will great things coming...may $ & happyness rain down the from the sky on IMASI Guy
Fred
The center of the first circle was the Zo of TL. You were right, the wide TL had a Zo < 50 Ohms. When you made it too narrow, Zo > 50 Ohms, which was the Center of that circle.
I used to think the center of the circle was the Zo but my good friend did the math and got:
Zo = sqrt( ( Rmiddle*3*50^2 - 50^3 ) / ( 3*50 - Rmiddle ) )
@@IMSAIGuy the center is just an arbitrary reference point. 50 Ohms is a common Zo, so it is often used as the center, but the Zo of your microstrip can be anything. In a microstrip RF filter, you would select the Zo of the different segments to get the frequency response you want (lengths too of course).
really great stuff, i enjoy your videos
What frequency range did you make the stripline for? What is the dielectric constant of the board material?
FR4 glass epoxy is normally not used for microwave stripline construction. Rogers RT Duroid is a 'standard' that I became accustomed to use. Just my 2 cents.
it's
I was wondering about the tape. I am assuming that it is copper tape with adhesive on the backside. Would the adhesive act as an insulator? So if you stuck one piece of tape on top of another, would you have to solder them together to make a connection?
copper tape comes with conductive and non-conductive adhesive versions. that said, I always solder them together.
How Can I Make One Of Them HF Filters For 80 Meters To 2 Meters?
I have a question. How do we measure the resistors by using this board?
Thanks!
you don't. this is a transmission line
I don’t claim to know much about the magic of RF, but this video makes me wonder what would happen if you simply took a piece of 50ohm coax and sliced it longways and basically unwrapped it (leaving the coax wrapper and center conductor intact) and glued the entire filleted piece of coax onto any non-reactive surface so as to make the insides of the cable accessible for insertion of elements.
it would not be 50 ohms
Info on the program used to calculate the design?
www.hp.woodshot.com/
You should mount stiching vias to better (for high frequency) connect gnd from top and bottom. Signal backward way would be quicker.
you are describing a co-planar waveguide. hard to prototype. this is called microstrip waveguide
@@IMSAIGuy for microstrip works fine too
@@mariuszwozniak6918 I don't think you can call it microstrip then to be picky
www.pasternack.com/t-calculator-microstrip.aspx
@@mariuszwozniak6918 chemandy.com/calculators/coplanar-waveguide-with-ground-calculator.htm
Very cool !!!
It's a simple unbalanced transmission line. So what causes the re-entrant loops in the Smith impedance trace?
loops are a phase shift due to increasing electrical length as a function of frequency
@@IMSAIGuy explain this to me like I'm 5
@@eastonbryan4171 that knife is too sharp for you
IMSAI Guy I’m an ee student struggling to understand why when I terminate my microstrip with a matched load i get predictable results but when I just connect both ports to VNA I get reflection and weird curls on the polar/smith plot (the microstrip ~should be~ impedance matched). We’re also supposed to calculate the width and length of the microstrip somehow based on frequency but I thought the width of the microstrip was set?
@@eastonbryan4171 the section of microstrip should act as a transmission line. so to test your setup. put a short piece of coax in place of your microstrip and measure it. it is a known good transmission line. if you then put your microstrip in and it looks bad then it is bad. the width of the strip is a variable as is many other mechanical dimensions. there are many programs to calculate the correct dimensions. I don't how your professor wants you to do this
You built an antenna and the loss is radiated
I confirm
You can build a stepped impedance filter or any other filter. I use sonnet em to model them
any freeware like sonnet