Hacker News new | past | comments | ask | show | jobs | submit login
3D-printed antenna radiation patterns in color [video] (youtube.com)
68 points by blueintegral on May 28, 2017 | hide | past | favorite | 19 comments



Is it that much more "enlightening" to 3D print vs just rotate the model on a computer screen? I can rotate a rendered model and get a pretty solid idea of the shape. I don't have to hold a physical model.


You inherently get a better sense of the pattern since you can use stereoscopic vision rather than parallax+movement. It could give you a better sense of what a complicated shape looks like. When you look at something complex enough[1] it helps to get as much information as possible. The eyes alone are not optimal for grokking it, even when the shape is moving around[2].

[1] https://images4.sw-cdn.net/product/picture/710x528_583780_19...

[2] https://www.youtube.com/watch?v=K9YAAS0cZ_w

also all hail king Quilez, master wizard


Ok, but why not use VR goggles?


Part of it is the sense of touch, weight, inertia and angular momentum. Like knowing how full a bottle is when you shake it.


Because VR still sucks, and even good VR still messes with sense of proportion. Easier for most just to handle the object. But for complex or impractical to print items, it is a fantastic idea used by most modern engineering companies.


And his physical models are now rendered to our computer screens just like a virtual 3D model would be. Just with a bit more noise and shading.


I like how you think. 2D ---> 3D ---> 2D. Nothing has changed and yet we feel the latter is more "comprehendible"!


Hard to say.

Is it much more enlightening for a doctor to hold a model of a medical scan in their hand, or rotate it on a screen?

There's potentially a lot of information to be interpreted in a 3D model, being able to use our natural systems to interpret the model seems like it would make it easier to understand. I imagine VR would be a step from looking at a model on a flat screen.

Having the physical model goes a step further, allowing you to experience it tactiley as well as visually. I suspect that the ability to manipulate the model with your hands while looking at it would make it easier to understand its features as well.

For simple antenna models I suspect not much is gained by having the printed model, they just look pretty cool. If you were to model more complicated antennae, or include interference from surrounding components, the resulting physical models have the potential to make understanding exactly what is going on much easier.


I think this is more of a vanity/luxury than actually TRULY useful thing. Many complex geometries are designed on CAD - from Airplane assemblies to intricate VSLI circuits.

Physical model is nice...but its utility is not justified for the expense of 3D printing.

Probably in industrial design and architecture. I don't come across a 3D graph where I go "Gee! I wish I had a physical model. I just can't understand this thing".

Don't get me wrong, it is nice. But I am debating how much better?


Yeah I agree with you, the value in this case is probably pretty small.

I think it comes down to two things, how useful it is to understand the detail of the geometry, and how much easier the printed model makes it to understand.

It's hard to estimate how much being already familiar with the geometries in question helps as well. If you've studied them a lot then a physical model probably adds little. If you're trying to explain to a designer why they need to move a component it probably helps a lot.

A physical model also has an extremely intuitive interface. The value probably comes more when introducing someone to the geometry for the first time, rather than for the person making the print (though I can still imagine situations where it would be worthwhile).


There is value in the 2d -> 3d print -> 2d because the realworld lighting is much better in showing us aspects of the shape than say a raw CAD render.


Maybe for some other phenomena, but for antenna radiation it's not that complex, it just looks cool.


This makes me wonder: is it possible to submerge the antenna in a special fluid or powder and send some modulated current through it, so that said fluid hardens in form of a radiation pattern? We could then MR scan it and see the real pattern in 3D. It would be super interesting to compare the result to a simulation. Any ideas on what that fluid should be?


Do a helical or "quadfilar" haha throw some words in

Like that color, I bet you could print a pretty convincing peach


I am more impressed at that ability to blend colors on a print, but like you would on an ink-jet printer.


This is done with an inkjet process :)


These 3D prints are amazingly perfect compared to what I've been able to achieve. With what methods/hardware can I achieve something similar when it comes to shape/texture?

What about the colours?


Nice, but antenna patterns are typically shown on a log scale. It would probably be better if he'd 3d printed on a linear scale, which would give a better physical idea of what the antenna is doing.


Most antenna patterns are plotted at the 3dB roll off. While this is logarithmic, it's also "half power".

https://electronics.stackexchange.com/questions/6959/what-is...




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: