DesignCon 2024
Conventions of this type are about riding the latest wave of advancing technology. These “interesting times” include a new acronym for printed circuit board geometry. We use Ultra-HDI (UHDI) when the conductors are too fine to call it “High Density Interconnect.” So, UHDI is the new kid on the tradeshow block. The significant difference is that HDI is based on etching processes while UHDI deposits the conductors on the dielectric much the same as the case with packaging substrates.
The key limitation, as far as I can tell, is that the fine lines do not go on outer layers or other layers where vias start or end. Take a six-layer board with a blind via from Layer 2 to Layer 5. The additive layers would be 3 and 4. It’s about what can be plated up once the circuit pattern is in place. The ultra-fine lines are not currently candidates for additional plating. As a result, we wouldn’t get to have micro-vias from layer 2 to 3 or 3 to 4 while using the modified semi-additive process (MSAP) on those routing layers. Ironically, that means you use the bigger core via to access MSAP layers.
Figure 1. Image Credit: Author - The ceiling is the limit of my home base any time the conventions hit town.
Artificial Intelligence is Also in the Spotlight
No self-respecting convention taking place in 2024 is going to ignore Artificial Intelligence. This is another pervasive theme at DesignCon. The tech under this banner relies on next generation memory, higher bandwidth and more thorough verification among other topics. This leap in compute power can’t be held back.
Whether it’s all hype or the next pivot point in the global economy, we’ll find out over the course of this formative time. Personally, I jumped into machine learning back in 2017 in the service of lidar sensors that would generate a “point-cloud”. The 360-degree range finder performed simultaneous location and mapping or SLAM for short. Synthesizing the immersive views that I referred to as the point cloud is only a means of letting us see what the sensor is sensing with its spinning lasers and detectors.
Engineer of the Year 2024
Shout-out to Scott McMorrow, Casey Morrison, Yurly, Shlepnev, Bert Simonovich and Ransom Stephens. These were the finalists for DesignCon 24 Engineer of the Year with Casey taking home this year’s prize. Kudos to all of the engineers who are helping us move our technology forward. Any and all of these folks are worth following for the content they share.
Let’s Talk About the James Webb Space Telescope
Imagine if, instead of mils or microns, our daily conversations were spoken with “lightyears” as a common unit of measure. Take it much, much further. How many lightyears? How about all of them? Let’s look at the very first lightyear and break it down.
In today’s memegen world:
Q: “How far into distant space and time?”
A: “Yes!”
What’s left of that early light has doppler shifted into the infrared band - over the entire length of the space/time continuum. We don’t see infrared as light but we feel it near a campfire or out in the sunshine. This project, however, would be far more difficult than noticing a match burning from 93 million miles away. You don’t have to be in the telescope business to know that it would be a nearly impossible assignment to capture that primordial energy.
Most of us non-astronomers in attendance learned about a different type of orbit. The Lagrange points are described well in the NASA doc regarding them as follows:
“Lagrange Points are positions in space where the gravitational forces of a two body system like the Sun and the Earth produce enhanced regions of attraction and repulsion. These can be used by spacecraft to reduce fuel consumption needed to remain in position.”
That is the most digestible description I can find. The historical mission to fix the Hubble Space Telescope would be out of the question where the JWST orbits. If you listed all of the things that had to go right to get on orbit, the length of the list would give pause. If one thing goes wrong, the project is lost. Somehow, they nailed it on the first and only try.
Figure 2. Image Credit: Author - Jonathan Arenberg discussing the challenges of creating the most visionary telescope yet.
Inventing New Ways to Test
Of course, we all would want to test these activities extensively. There were tests that couldn't be performed on Earth since our gravity is so strong. Fundamental things like aligning the mirrors and unfolding the heat shield could not be tried in advance. Operating temperature is a big deal that far from home.
What I recall about the required temperature was 40 Kelvin. When they use Kelvin with two digits, that’s cold beyond what we can imagine let alone endure. So, you had to deploy it somewhere that is cold, dark and caught in the gravitational eddy between two celestial bodies. That was the L2 Lagrange point if it was going to be anywhere.
Figure 3. Image Credit: Science.NASA.gov - I have a hard time wrapping my head around the gravity that animates the Sun-Earth-Moon interrelationship. This L2 orbit is “out there”.
When he detailed the complexity of deploying the shield, he mentioned that it was constructed of mylar, something we call coverlay in the flex circuit business. They used five layers of Kapton, with four of them being the usual 25 micron dielectric with aluminum instead of copper clad. The outermost layer is 50 microns thick with a special coating for high durability and reflectivity. I wonder if all of that demand caused a lead time issue.
DesignCon has a history of showcasing the vendors that supply connectors. That may even be what is meant by the “Con” part of the name rather than convention. I kid, but the likes of Samtec, Molex and Lotes were there and will be again in 2025. A lot of test equipment vendors also make the annual pilgrimage to Santa Clara. If you’re going to build out a data center, this is one of the events you should attend. Even if you’re not pushing the envelope, it’s good to know what is on the horizon. They can count on my attendance and I hope to see you there too.