Data retention is a funny thing. Atmel will gladly tell you that the flash memory in an ATmega32A will retain its data for 100 years at room temperature. Microchip says its EEPROMs will retain data for over 200 years. And yet, humanity has barely had a good grasp on electricity for that long. Heck, the silicon chip itself was only invented in 1958. EEPROMs and flash storage are altogether younger themselves.
How can these manufacturers make such wild claims when there’s no way they could have tested their parts for such long periods of time? Are they just betting on the fact you won’t be around to chastise them in 2216 when your project suddenly fails due to bit rot.
Well, actually, there’s a very scientific answer. Enter the practice of accelerated wear testing.
Will Rogers once said that veterinarians are the best doctors because their patients can’t tell them where it hurts. I’ve often thought that electronic people have a similar problem. In many cases, what’s wrong with our circuits isn’t visible. Sure, you can visually identify a backward diode, a bad solder joint, or a blown fuse. But you can’t look at a battery and see that it is dead or that a clock signal isn’t reaching some voltage. There are lots of ways to look at what’s really going on, but there is no substitute for a scope. It used to be hard for the average person to own a scope, but these days, it doesn’t require much. If you aren’t shopping for the best tech or you are willing to use it with a PC, oscilloscopes are quite affordable. If you spend even a little, you can now get scopes that are surprisingly capable with features undreamed of in years past. For example, many modern scopes have a dizzying array of triggering options. Do you need them? What do they do? Let’s find out.
I’ll be using a relatively new Rigol DHO924S, but none of the triggering modes are unique to that instrument. Sometimes, they have different names, and, of course, their setup might look different than my pictures, but you should be able to figure it out.
What is Triggering?
In simple terms, an oscilloscope plots time across the X-axis and voltage vertically on the Y-axis. So you can look at two peaks, for example, and measure the distance between them to understand how far apart they are in time. If the signal you are measuring happens repeatedly — like a square or sine wave, for example — it hardly matters which set of peaks you look at. After all, they are all the same for practical purposes.
Pretty square waves all in a row. Channel 2 is 180 degrees out of phase (inverted). But is that all there is?
The problem occurs when you want to see something relative to a particular event. Basic scopes often have level triggering. They “start” when the input voltage goes above or below a certain value. Suppose you are looking at a square wave that goes from 0 V to 5 V. You could trigger at about 2.5 V, and the scope will never start in the middle of a cycle.
Digital scopes tend to capture data before and after the trigger, so the center of the screen will be right on an edge, and you’ll be able to see the square waves on either side. The picture shows two square waves on the screen with the trigger point marked with a T in the top center of the display. You can see the level in the top bar and also marked with a T on the right side of the screen.
What happens if there are no pulses on the trigger source channel? That depends. If you are in auto mode, the scope will eventually get impatient and trigger at random. This lets you see what’s going on, but there’s no reference. If you are in normal mode, though, the scope will either show nothing or show the last thing it displayed. Either way, the green text near the top left corner will read WAIT until the trigger event occurs. Then it will say T’D.
Once upon a time, when the earliest spy satellites were developed, there wasn’t an easy way to send high-quality image data over the air. The satellites would capture images on film and dump out cartridges back to earth with parachutes that would be recovered by military planes.
It all sounds so archaic, so Rube Goldberg, so 1957. And yet, it’s still a viable method for recovering big globs of data from high altitude missions today. Really, you ask? Oh, yes indeed—why, NASA’s gotten back into the habit just recently!
Although the concept of nuclear fission is a simple and straightforward one, the many choices for fuel types, fuel design, reactor configurations, coolant types, neutron moderator or reflector types, etc. make that nuclear fission reactors have blossomed into a wide range of reactor designs, each with their own advantages and disadvantages. The story of the pebble bed reactor (PBR) is among the most interesting here, with its development winding its way from the US Manhattan Project over the Atlantic to Germany’s nuclear power industry during the 1960s, before finding a welcoming home in China’s rapidly growing nuclear power industry.
As a reactor design, PBRs do not use fuel rods like most other nuclear reactors, but rather spherical fuel elements (‘pebbles’) that are inserted at the top of the reactor vessel and extracted at the bottom, allowing for continuous refueling, while helium acts as coolant. With a strong negative temperature coefficient, the design should be extremely safe, while providing high-temperature steam that can be used for applications that otherwise require a coal boiler or gas turbine.
With China recently having put its twin-PBR HTR-PM plant into commercial operation, why is it that it was not the US, Germany or South Africa to first commercialize PBRs, but relative newcomer China?
Of all the rabbit holes we technical types tend to fall down, perhaps the one with the most twists and turns is: time. Some of this is due to the curiously mysterious nature of time itself, but more has to do with the various ways we’ve decided to slice and dice time to suit our needs. Most of those methods are (wisely) based upon the rhythms of nature, but maddeningly, the divisions we decided upon when the most precise instrument we had was our eyes are just a little bit off. And for a true time junkie, “a little bit off” can be a big, big problem.
Luckily, even the most dedicated timekeepers — those of us who feel physically ill when the clock on the stove and the clock on the microwave don’t match — have a place to go that’s a haven of temporal correctness: radio station WWV. Along with sister stations WWVB and WWVH, these stations are the voice of the US National Institutes for Standards and Technology’s Time and Frequency Division, broadcasting the official time for the country over shortwave radio.
Some might say the programming coming from these stations is a bit on the dry side, and it’s true that you can only listen to the seconds slip by for so long before realizing that there are probably better things to do with your day. But the WWV signals pack a surprising amount of information into their signals, some of it only tangentially related to our reckoning of time. This makes these stations and the services they provide essential infrastructure for our technological society, which in turn makes it worth your time to look into just how they do it.
We’ve seen a lot of AI tools lately, and, of course, we know they aren’t really smart, but they sure fool people into thinking they are actually intelligent. Of course, these programs can only pick through their training, and a lot depends on what they are trained on. When you use something like ChatGPT, for example, you assume they trained it on reasonable data. Sure, it might get things wrong anyway, but there’s also the danger that it simply doesn’t know what you are talking about. It would be like calling your company’s help desk and asking where you left your socks — they simply don’t know.
We’ve seen attempts to have AI “read” web pages or documents of your choice and then be able to answer questions about them. The latest is from Google with NotebookLM. It integrates a workspace where you can make notes, ask questions, and provide sources. The sources can be text snippets, documents from Google Drive, or PDF files you upload.
You can’t ask questions until you upload something, and we presume the AI restricts its answers to what’s in the documents you provide. It still won’t be perfect, but at least it won’t just give you bad information from an unknown source. Continue reading “Can Google’s New AI Read Your Datasheets For You?”→
By the end of the decade, NASA’s Artemis program hopes to have placed boots back on the Moon for the first time since 1972. But not for the quick sightseeing jaunts of the Apollo era — the space agency wants to send regular missions made up of international crews down to the lunar surface, where they’ll eventually have permanent living and working facilities.
The goal is to turn the Moon into a scientific outpost, and that requires a payload delivery infrastructure far more capable than the Apollo Lunar Module (LM). NASA asked their commercial partners to design crewed lunar landers that could deliver tens of tons of to the lunar surface, with SpaceX and Blue Origin ultimately being awarded contracts to build and demonstrate their vehicles over the next several years.
Starship and Blue Moon, note scale of astronauts
At a glance, the two landers would appear to have very little in common. The SpaceX Starship is a sleek, towering rocket that looks like something from a 1950s science fiction film; while the Blue Moon lander utilizes a more conventional design that’s reminiscent of a modernized Apollo LM. The dichotomy is intentional. NASA believes there’s a built-in level of operational redundancy provided by the companies using two very different approaches to solve the same goal. Should one of the landers be delayed or found deficient in some way, the other company’s parallel work would be unaffected.
But despite their differences, both landers do utilize one common technology, and it’s a pretty big one. So big, in fact, that neither lander will be able to touch the Moon until it can be perfected. What’s worse is that, to date, it’s an almost entirely unproven technology that’s never been demonstrated at anywhere near the scale required.
