Fortnight ending Saturday 1st August 2026. Read time: about 10 minutes.

This fortnight begins with a genuinely serious story: the withdrawal of STFC funding from e-MERLIN and the resulting threat to the Lovell Telescope’s future as a research instrument at Jodrell Bank. Elsewhere, Psyche has skimmed past Mars, Hubble has found a helium nova firing “cosmic bullets”, a rocky exoplanet may finally have yielded firm evidence of an atmosphere, and New Scientist reckons that the cosmological principle could be dead!?

There is also a guest piece by Dr Daniel Jones on teachers building physics tools with AI, an eclipse reminder, a clearer route into entropy, and a Chladni plate that may explain several unidentified objects in the prep room.

The big UK story is Jodrell Bank. UKRI’s STFC prioritisation outcomes say the council is trying to put itself on a sustainable footing and focus spending on what it sees as the greatest long-term capability and value. In practice, that means funding for the e-MERLIN network at Jodrell Bank is due to end in March 2028. The Institute of Physics has called the decision a “terrible blow”, while the University of Manchester says Jodrell remains open, the public engagement centre is unaffected, and alternative routes to funding are being pursued. That is the optimistic reading. The less optimistic reading is that, if that replacement funding does not arrive, the Lovell Telescope could stop operating as a research instrument.

Why it matters is obvious enough: Jodrell is not just a famous dish in Cheshire. It is part of the UK’s radio astronomy infrastructure, a UNESCO World Heritage Site, and a place with actual scientific, engineering and cultural weight. It has also been a route into physics for generations of pupils who first encountered “space” there and discovered it involved quite a lot of cables. For departments, this is one of those stories worth knowing in detail because students will ask about it, and because it says something broader about what a country thinks science is for.

Source: UKRI, Institute of Physics, The University of Manchester, The Guardian, AP News

NASA’s Psyche mission has released superb images and a timelapse from its Mars gravity assist, with the spacecraft passing around 2,800 miles above the Martian surface and using the encounter to tweak its trajectory towards asteroid Psyche. The fly-by also gave engineers a chance to calibrate instruments, including the magnetometer and multispectral imagers, and to spot both Phobos and Deimos.

Classroom-wise, this is rich: gravity assists, orbital energy changes, phase, imaging, and why “just taking the direct route” is often not how spaceflight works. The Sky at Night write-up is especially useful because the footage is the hook and the physics comes quietly bundled with it. Which is usually the best way.

Source: NASA, Space.com, BBC Sky at Night Magazine

A very good piece of stellar-weirdness this fortnight: astronomers studying V445 Puppis, the only known helium nova in our galaxy, have found fast clumps of gas shooting out at up to 20 million mph. The system was hidden for years by dust after its 2000 eruption, but new observations with Hubble, TESS, the VLT and others have now revealed the underlying binary system and identified unusual high-speed ‘bullets’ within its ejecta. Space.com’s coverage is a good readable summary; the Royal Astronomical Society highlight gives the cleaner scientific framing.

For sixth form, this is a lovely enrichment story because it joins up stellar evolution, accretion, thermonuclear runaway, spectra and the reason Type Ia supernovae matter so much for cosmology. It is also a good reminder that “nova” is not one thing. Physics does hate a tidy noun.

Source: Royal Astronomical Society, Space.com

The exoplanet story worth knowing is LHS 1140 b, where researchers report the first confirmed atmosphere around a rocky planet in another star’s habitable zone. The key signal is helium escaping from the planet, described in a Science paper and summarised clearly by ScienceDaily via the Center for Astrophysics. It does not mean “new Earth found”; it does mean the atmospheric side of habitable-zone exoplanet work has taken a serious step forward.

This is one for spectroscopy, transits, inference and caution. Students often imagine exoplanet science as glorified star-spotting. It is much subtler than that: you infer an atmosphere from what a gas does to light, and from how that signal changes as a planet passes its star. Properly taught, it is an excellent antidote to the idea that astronomy is just taking prettier pictures with more expensive mirrors.

Source: ScienceDaily, Science

Modern cosmology rests on the cosmological principle: viewed on a sufficiently large scale, the universe should be statistically uniform and look broadly the same in every direction. A new analysis discussed by New Scientist challenges that assumption.

Using galaxy data from DESI, researchers reported coherent directional structures stretching across billions of light-years—larger and stronger than those produced in simulations based on the standard cosmological model. If the result holds, cosmologists may need models that allow the universe to remain significantly lumpy and direction-dependent even at enormous scales.

There is an important caveat. A rapid reanalysis of the same data argues that the apparent anomaly disappears when the correct distance scale is used. So the cosmological principle has not necessarily fallen; someone has merely begun checking the foundations rather carefully.

For sixth form, this is an excellent example of the difference between an observation, an assumption and a model—and of science working through disagreement rather than waiting for a tidy final answer.

Source: New Scientist, Nature, arXiv

A strong A-level move this fortnight is to use WIRED’s entropy piece with a short probability-first starter. The article’s central point is that entropy is better understood through the number of microscopic arrangements compatible with a macroscopic state than through the tired “disorder” shorthand. For students doing thermal physics, that gives you a much cleaner route into why some macrostates are overwhelmingly more likely than others.

If you want one explicit classroom sequence, keep it simple. Start with two or three outcomes from a handful of dice or coin tosses, ask which macro-outcome has the most microscopic realisations, and only then connect that counting argument to entropy. The advantage is that you are not asking students to memorise a slogan; you are asking them to see why the slogan was always a bit wobbly. Pairing the article with this week’s Physics Video of the Fortnight below works well, so long as you say out loud that the video’s language comes from information theory and needs joining carefully to thermodynamics rather than dumped on top of it.

Who this is for: strong A-level groups on thermal physics, or GCSE classes you want to stretch without pretending Boltzmann lived in the specification.

Source: WIRED, YouTube

One to watch rather than react to immediately: Schools Week reports that new education secretary Lucy Powell said she wants to “take forward some changes to the curriculum and assessment system”, described current GCSEs as “a lot of pressure”, and linked that discussion to the government’s accepted curriculum-and-assessment-review recommendation for a 10% reduction in exam time. At the moment this is direction-of-travel material, not a timetable change for next term.

For physics teachers, that means the sensible response is not to rewrite a scheme of work in August. It does mean keeping an eye on future consultation detail, especially where exam volume, technical pathways and the shape of 14–16 assessment start intersecting. In other words: not a fire, but definitely smoke.

Source: Schools Week

By Dr Daniel Jones

AI has made it possible for physics teachers with little coding experience to build bespoke simulations and tools that fill genuine gaps in existing classroom resources.

Until November 2025, I had never seriously considered how AI might be used to improve physics education. A post from Rhodri on the Physics Teaching News and Comment (PTNC) forum inspired me, and since then I have been building apps and tools that I hope will improve learning for my students.

These include resources for tackling misconceptions and creating visual representations that would be difficult to draw or convey using a pen and whiteboard alone. Some of the newer tools may also save colleagues a considerable amount of time.

I had very limited coding experience before starting what has become known as “vibe coding”: describing what you want to an AI tool, then repeatedly testing and refining the code it produces.

There are a number of teachers doing similar work. For many of us, the appeal has been the ability to create a resource where no suitable tool previously existed. PhET and oPhysics remain excellent, but there are still important areas that their existing resources do not cover.

Much of the work so far has focused on simulations and apps that introduce, revisit or review practical aspects of KS3, GCSE and A-level physics. More recently, teachers have also begun creating tools for worksheet generation, retrieval practice and resource discovery.

One example is my Doppler-effect simulation. Feedback from colleagues has highlighted how easily it makes a difficult concept visible: students can see wavefronts bunch together ahead of a moving source and spread out behind it, rather than trying to reconstruct the whole process from a static sketch. The app can also show what happens as the source reaches and exceeds the speed of sound, including the formation of a sonic boom. This is exactly the kind of explanation that can be awkward on a whiteboard but immediate in an interactive model.

There are now several AI products that can be used to create physics resources, and all have merit. I began with Gemini because it was included with my Google subscription, and I have also tried Copilot and ChatGPT. For me, however, Claude has been by far the most impressive.

Its ability to create complex animations and visuals is one feature that currently sets it apart. It can also generate code running to tens of thousands of lines; in my own experience, that greater depth has often correlated with more robust and capable apps.

None of this removes the need for careful professional judgement. AI-generated resources must still be checked for physics accuracy, accessibility and classroom suitability. Teachers should also consider data protection carefully, particularly where students might enter information into a tool. AI can produce convincing mistakes remarkably efficiently, so subject expertise remains essential.

I recorded a short introduction to the process using Gemini. The tools have already become easier to use, but the video still demonstrates the basic cycle of prompting, testing and refining an app.

I also wrote about this work in the July 2026 issue of the Association for Science Education’s School Science Review, which contains several articles focused on AI in science education.

The most important point is not that AI can replace the expertise involved in designing a good physics resource. It cannot. Its value is that it lowers the technical barrier, allowing teachers who understand the classroom problem to build, test and improve tools that might previously have required professional coding skills.

If I can help, please get in touch through X, Bluesky or LinkedIn, or email me at [email protected].

Dr Daniel Jones is Head of Physics at Haberdashers’ Monmouth School and created the PTNC physics education forum, which has more than 600 members. His physics videos have received nearly 15 million views, and he now develops a growing collection of AI-assisted apps and resources for physics teaching.

If you have an astronomy club, a school social account, or just one colleague who enjoys sending excellent mass emails in August, the Royal Observatory Greenwich eclipse guide is worth bookmarking now. The UK and Ireland will see a partial eclipse on 12 August, with roughly 90–96% coverage depending on location; Greenwich gives safe-viewing advice and the livestream page is already live. If you want London timings, timeanddate gives first contact at 18:17, maximum at 19:12, and the end at 20:06. Last issue was right to flag it early; this issue’s job is simply to nag.

Source: Royal Observatory Greenwich, Royal Observatory Greenwich Live Stream, timeanddate

If you teach Year 13 and like sending pupils off with something better than “good luck at university”, the IOP Ambassadors scheme is worth passing on. It is for university physics students and involves running physics- and careers-related events on campus, with IOP support and training. It is the sort of thing confident leavers often genuinely like doing, and the sort of thing less confident leavers often assume is for someone else until you explicitly tell them it is not.

Source: Institute of Physics

CLEAPSS has a useful cluster of practical-safety CPD and support live right now. The free introductory video on managing safety in practical science is a sensible thing to sling at non-specialists or new starters before September. If you want something more specific for physics, there is an in-person Radiation Protection Supervisor course in London on 17 September, an online RPS course on 21 September, and an early-career practical science safety course in London on 22 October amongst other dates and locations. None of that is glamorous. All of it is useful. (CLEAPSS login may be required to access links)

Source: CLEAPSS, CLEAPSS Courses, CLEAPSS Courses, CLEAPSS Courses

And if you want a very odd bit of August outreach, a spent Falcon 9 upper stage is predicted to hit the Moon on 5 August at about 06:35 UTC. The observing paper on arXiv explicitly encourages both professional and amateur observations, and EarthSky’s guide gives a quick public-facing overview. The Independent piece is the headline version; the arXiv paper is the one to trust if you actually want the science. It is a nice reminder that “space debris” and “lunar observation” are no longer separate conversations.

Source: arXiv, EarthSky, The Independent

This fortnight’s watch is Entropy (for data science) Clearly Explained!!!. Yes, it comes from the information-theory side rather than school thermal physics, and that is exactly why it is useful. If your students’ idea of entropy is still “messiness but in a smart font”, this gives a far more intuitive route through probability, ‘surprise’, uncertainty and the counting logic underneath the concept. Use it as a bridge, not a replacement, and it does a very decent job. There are so many ‘but what is entropy!?’ videos out there, but this one, for me, was by far the most explanatory and intuitive derivation. Other brands of explanation are available.

Source: YouTube, WIRED

A Chladni plate turns stationary waves into something students can actually see. A metal plate is mounted on a vibration generator and driven by a signal generator; when a resonant frequency is reached, fine sand or salt moves away from the antinodes and gathers along the nodal lines, producing striking two-dimensional patterns.

The CLEAPSS Chladni plate video shows how to assemble and operate the apparatus rather than presenting only the finished result. It is particularly useful for anyone who has found the plate, vibrator and signal generator in three different cupboards and suspected they might once have belonged together.

Source: CLEAPSS

Misconception: Entropy just means disorder.

Better framing: Entropy is about how many microscopic arrangements are compatible with the macroscopic state you observe. “Disorder” is sometimes a passable metaphor, but it is not the idea.

Why students get stuck: Because “messiness” feels intuitive, while counting microstates sounds like statistical mechanics has arrived early and without biscuits.

Source: WIRED

“We live in a society exquisitely dependent on science and technology in which hardly anyone knows anything about science and technology; that is a clear prescription for disaster. I mean, if you'd like to commit National Suicide, here is a relatively slow and painless way to go about it”

— Carl Sagan

Source: Minnesota Public Radio Archive — Education: Strategies for Change

The Physics Staffroom is a human–AI collaboration. AI helps gather and format material, but each issue’s selection, verification, editing, design, and regular features are all done by a humble, human physics teacher! 🤓

You're spending 40 hours a week writing code that AI could do in 10.

While you're grinding through pull requests, 200k+ engineers at OpenAI, Google & Meta are using AI to ship faster.

How?

The Code newsletter teaches them exactly which AI tools to use and how to use them.

Here's what you get:

Sign up and get access to the Ultimate Claude code guide to ship 5X faster.

Join 200K+ engineers

Trade on real-world events you already follow, from elections and inflation to sports, tech, and more. Choose “Yes” or “No” based on what you think will happen and see how your prediction plays out.

Pick your market and start trading what’s next.

Trade $25, Get Up to $500

_{Bonus credit varies from $15 to $500. Terms apply.}