Really cool! BTW since you're here, have you checked https://github.com/malloydata/malloy/ and their visualizations? it seems it could be a nice fit for ObservableHQ
kazinator
2 days ago
[ - ]
"Escher's art, Smith Chart and Hyperbolic Geometry"
"A conceptual relation between Circle Limit IV, an artistic creation by M.C. Escher, and Smith Chart, geographical aid for microwave engineering created by P.H. Smith, was established. The basis of Escher's art and Smith chart can both be traced back to invariance of the cross ration of four complex numbers under Möbius transformation on the domain of complex numbers. The Smith chart can be used as an aid for constructing Escher-like drawings that display periodic mosaic patterns and at the same time convey the perception of infinite progression within a unit circle."
I've used Smith charts many times on my NanoVNA analysing antennas, but despite being mathematically inclined I never thought of the complex plane mapping involved
f(z) = (z − 1)/(z + 1)
The Smith chart is useful to electronic engineers because a given VSWR (the thing you try to minimise to get a good antenna) becomes a circle about the center of the Smith chart
VSWR= (1 + | Γ |)/(1 - | Γ |)
So to make your antenna better, get the plot closer to the center. Whether it is above the line or below tells you whether the antenna is inductive or capacitive and hence which kind of loading to add.
Not knowing enough about Smith Charts, do the intermediary frames have any meaning in terms of actual practical problems or calculations?
It seems like the chart itself would be a member of an entire family of chart types that would go through the entire 3D rotation with several versions that were "nearly tangent circles" and eventually concentric circles at the center of the graph, along with some of the intermediary frames shown in the animation.
kazinator
1 day ago
[ - ]
The intermediary frames have an interpretation as being congruent to a scaling. Near the outset of the animation when the imaginary axis slightly bends, it has already become a circle. So it is congruent to a large Smith chart.
btkramer9
2 days ago
[ - ]
I loved smith charts during my microwave classes in college. I've always felt like there should be a fun game based on the mechanics of using a smith chart.
uzby
2 days ago
[ - ]
This software is so good it is as enjoyable as any game, it taught me to love Smith Charts
I felt the same way. At some point shortly after graduation, I found a heavy plastic Smith chart that was designed to be written on with a grease pencil, and erased with a cloth when you're done.
dmd
2 days ago
[ - ]
I’m just amazed there isn’t a nearly impenetrable Greg Egan book based on them.
jordigh
2 days ago
[ - ]
Btw, as all Möbius transforms, a Smith chart can be understood by looking at the complex plane as a sphere:
Smith charts look really cool, and they can be really useful for modeling, but for me I can't see the phasors when I use a Smith chart. Myself and all the EEs I work with use bode plots to see mag and phase response across frequency. I know some sick f EEs that just want raw IQ pairs. But Smith charts look sci fi so they're good too.
munchler
2 days ago
[ - ]
> The Smith chart from electrical engineering is the image of a Cartesian grid under the function f(z) = (z − 1)/(z + 1). More specifically, it’s the image of a grid in the right half-plane.
Well, I'm already confused. The domain is complex numbers where the real part is >= 0? I think it would be helpful to make that clear from the get-go. When I see "Cartesian grid", I think R^2, not C.
crdrost
2 days ago
[ - ]
So you can continue it beyond that circle if you like, it just happens to be the case that the thing electrical engineers and others dealing with waveguides are plotting in the source space, is impedance. (Well, a ratio of impedances—a load impedance divided by a transmission line impedance.) The real part of impedance is resistance, and negative resistance is very uncommon. The area outside the unit circle so mapped, I think also corresponds to reflected amplitude ≤ transmitted amplitude, with the center of the diagram being a perfectly matched impedance, and no reflection.
My prof even said to me "If you'd pay attention, you'd understand this!". I was the only one who was brave enough to ask questions about it!
OldSchool
2 days ago
[ - ]
Heh, this reminds me, don't worry:
One of my extremely intelligent roommates in the 80s switched from EE to CS, seemingly due to Smith charts and Electromagnetics coursework.
He went on to make a large fortune in software.
thraxil
2 days ago
[ - ]
I switched from EE to CS (well, "Computer Engineering" technically) in the late 90s. Not specifically due to Smith charts, but that's relatable. For me it was just realizing that I was procrastinating on doing my EE problem sets, which just started to seem like endless grinding of differential equations, by playing around with whatever we were doing in the couple CS classes I had. I wouldn't say I've made "a large fortune" in software, but it's kept me gainfully employed for a few decades so I think it worked out.
gg82
2 days ago
[ - ]
Sounds like it was the right decision then.
hammock
2 days ago
[ - ]
Is this related to Cardioid patterns for directional microphones?
buescher
1 day ago
[ - ]
No. The RF equivalent of that would be an antenna pattern. The smith chart is a particular mapping of complex impedance that allows for fairly straightforward calculation and visualization of things like conjugate matches. It's related to the impedance plots for speakers, but audio engineers generally don't use it. RF engineers will plot impedance vs frequency on it though. If you come across curlicue plots on it, those are typically frequency sweeps i.e. they plot impedance vs frequency. You might have to look for the frequency legend or endpoints.
dieselerator
2 days ago
[ - ]
No. Sorry, that is not a helpful comparison.
raldi
2 days ago
[ - ]
Why would you use one?
s20n
2 days ago
[ - ]
I remember using these for impedance matching back when I was in college. Basically when you connect two transmission lines (like coax cables), you need to match their impedances so the signal does not bounce back. (Ik this is a gross oversimplification but yeah)
How to make a Smith chart
(johndcook.com)
165 points
by: tzury
2 days ago
33 comments
mbostock
2 days ago
[ - ]
I made an interactive implementation here: https://observablehq.com/@mbostock/smith-chart
aeontech
2 days ago
[ - ]
that's beautiful!
flimflamm
2 days ago
[ - ]
Really really nice. Thanks for this!
oulipo2
2 days ago
[ - ]
Really cool! BTW since you're here, have you checked https://github.com/malloydata/malloy/ and their visualizations? it seems it could be a nice fit for ObservableHQ
kazinator
2 days ago
[ - ]
"Escher's art, Smith Chart and Hyperbolic Geometry"
https://www.researchgate.net/publication/3427377_Escher's_ar...
"A conceptual relation between Circle Limit IV, an artistic creation by M.C. Escher, and Smith Chart, geographical aid for microwave engineering created by P.H. Smith, was established. The basis of Escher's art and Smith chart can both be traced back to invariance of the cross ration of four complex numbers under Möbius transformation on the domain of complex numbers. The Smith chart can be used as an aid for constructing Escher-like drawings that display periodic mosaic patterns and at the same time convey the perception of infinite progression within a unit circle."
myself248
2 days ago
[ - ]
This is one of my favorite explanations of how to use a Smith chart in practice: https://www.cypress.com/file/136236/download
Also this one: https://www.ti.com/lit/an/swra046a/swra046a.pdf
nickcw
2 days ago
[ - ]
I've used Smith charts many times on my NanoVNA analysing antennas, but despite being mathematically inclined I never thought of the complex plane mapping involved
f(z) = (z − 1)/(z + 1)
The Smith chart is useful to electronic engineers because a given VSWR (the thing you try to minimise to get a good antenna) becomes a circle about the center of the Smith chart
VSWR= (1 + | Γ |)/(1 - | Γ |)
So to make your antenna better, get the plot closer to the center. Whether it is above the line or below tells you whether the antenna is inductive or capacitive and hence which kind of loading to add.
RossBencina
2 days ago
[ - ]
> f(z) = (z − 1)/(z + 1)
Also known as the Bilinear Transform https://en.wikipedia.org/wiki/Bilinear_transform
Used for mapping between s-plane and z-plane when discretising using the trapezoidal rule.
kazinator
2 days ago
[ - ]
This animation is helpful:
https://en.wikipedia.org/wiki/Smith_chart#/media/File:Animat...
araes
1 day ago
[ - ]
Not knowing enough about Smith Charts, do the intermediary frames have any meaning in terms of actual practical problems or calculations?
It seems like the chart itself would be a member of an entire family of chart types that would go through the entire 3D rotation with several versions that were "nearly tangent circles" and eventually concentric circles at the center of the graph, along with some of the intermediary frames shown in the animation.
kazinator
1 day ago
[ - ]
The intermediary frames have an interpretation as being congruent to a scaling. Near the outset of the animation when the imaginary axis slightly bends, it has already become a circle. So it is congruent to a large Smith chart.
btkramer9
2 days ago
[ - ]
I loved smith charts during my microwave classes in college. I've always felt like there should be a fun game based on the mechanics of using a smith chart.
uzby
2 days ago
[ - ]
This software is so good it is as enjoyable as any game, it taught me to love Smith Charts
https://www.ae6ty.com/smith_charts/
seoulbigchris
11 hours ago
[ - ]
Oh, found a picture of one of these plastic ones flikr:
https://www.flickr.com/photos/jonsphotos/218609214/in/album-...
seoulbigchris
11 hours ago
[ - ]
If you liked the Smith Chart, you'd love using the Spirule to solve root locus problems.
https://en.wikipedia.org/wiki/Root_locus_analysis
seoulbigchris
11 hours ago
[ - ]
I felt the same way. At some point shortly after graduation, I found a heavy plastic Smith chart that was designed to be written on with a grease pencil, and erased with a cloth when you're done.
dmd
2 days ago
[ - ]
I’m just amazed there isn’t a nearly impenetrable Greg Egan book based on them.
jordigh
2 days ago
[ - ]
Btw, as all Möbius transforms, a Smith chart can be understood by looking at the complex plane as a sphere:
https://www.youtube.com/watch?v=0z1fIsUNhO4
mtneglZ
1 day ago
[ - ]
Smith charts look really cool, and they can be really useful for modeling, but for me I can't see the phasors when I use a Smith chart. Myself and all the EEs I work with use bode plots to see mag and phase response across frequency. I know some sick f EEs that just want raw IQ pairs. But Smith charts look sci fi so they're good too.
munchler
2 days ago
[ - ]
> The Smith chart from electrical engineering is the image of a Cartesian grid under the function f(z) = (z − 1)/(z + 1). More specifically, it’s the image of a grid in the right half-plane.
Well, I'm already confused. The domain is complex numbers where the real part is >= 0? I think it would be helpful to make that clear from the get-go. When I see "Cartesian grid", I think R^2, not C.
crdrost
2 days ago
[ - ]
So you can continue it beyond that circle if you like, it just happens to be the case that the thing electrical engineers and others dealing with waveguides are plotting in the source space, is impedance. (Well, a ratio of impedances—a load impedance divided by a transmission line impedance.) The real part of impedance is resistance, and negative resistance is very uncommon. The area outside the unit circle so mapped, I think also corresponds to reflected amplitude ≤ transmitted amplitude, with the center of the diagram being a perfectly matched impedance, and no reflection.
vismit2000
2 days ago
[ - ]
Smith Chart 101: https://www.youtube.com/watch?v=vDU5XnvZXwc
_whiteCaps_
2 days ago
[ - ]
I have PTSD from learning Smith charts in school.
My prof even said to me "If you'd pay attention, you'd understand this!". I was the only one who was brave enough to ask questions about it!
OldSchool
2 days ago
[ - ]
Heh, this reminds me, don't worry:
One of my extremely intelligent roommates in the 80s switched from EE to CS, seemingly due to Smith charts and Electromagnetics coursework.
He went on to make a large fortune in software.
thraxil
2 days ago
[ - ]
I switched from EE to CS (well, "Computer Engineering" technically) in the late 90s. Not specifically due to Smith charts, but that's relatable. For me it was just realizing that I was procrastinating on doing my EE problem sets, which just started to seem like endless grinding of differential equations, by playing around with whatever we were doing in the couple CS classes I had. I wouldn't say I've made "a large fortune" in software, but it's kept me gainfully employed for a few decades so I think it worked out.
gg82
2 days ago
[ - ]
Sounds like it was the right decision then.
hammock
2 days ago
[ - ]
Is this related to Cardioid patterns for directional microphones?
buescher
1 day ago
[ - ]
No. The RF equivalent of that would be an antenna pattern. The smith chart is a particular mapping of complex impedance that allows for fairly straightforward calculation and visualization of things like conjugate matches. It's related to the impedance plots for speakers, but audio engineers generally don't use it. RF engineers will plot impedance vs frequency on it though. If you come across curlicue plots on it, those are typically frequency sweeps i.e. they plot impedance vs frequency. You might have to look for the frequency legend or endpoints.
dieselerator
2 days ago
[ - ]
No. Sorry, that is not a helpful comparison.
raldi
2 days ago
[ - ]
Why would you use one?
s20n
2 days ago
[ - ]
I remember using these for impedance matching back when I was in college. Basically when you connect two transmission lines (like coax cables), you need to match their impedances so the signal does not bounce back. (Ik this is a gross oversimplification but yeah)
dchasson
2 days ago
[ - ]
Terrifying. My nightmares will return.
bmiekre
2 days ago
[ - ]
These are wormhole graphs, right?