Magnets is a big topic—what part are you most interested in?

Here are a few directions you can choose from (pick one or tell me your own):

1. How magnets work (physics basics)

• Why do magnets attract/repel? What is a magnetic field?

2. Types of magnets

• Permanent vs electromagnets vs temporary magnets
• Neodymium vs ferrite vs Alnico—what’s the difference?

3. Electromagnetism & electricity

• How current creates magnetism (coils/solenoids)
• How generators and motors work

4. Everyday applications

• Speakers, MRI, maglev trains, fridge magnets, credit card strips

5. Fun puzzles & “weird” behaviors

• Why can you levitate a magnet over a superconductor?
• Why do magnets get weaker with heat or time?

6. Building projects

• Want to make an electromagnet, a simple motor, or a compass?

Tell me which option(s) you want—or answer one quick question:
Are you aiming to understand magnets conceptually, solve a homework problem, or build something practical?

How much does the surface magnetic field strength of a specific neodymium magnet grade (e.g., N52, nickel-plated, 20 mm × 5 mm disc) measurably decrease after 1,000 controlled drop impacts (e.g., dropped from 1.0 m onto a steel plate) compared to an identical magnet stored undisturbed—when measured with the same gaussmeter at a fixed distance before and after?

To investigate “How much does an N52 neodymium disc’s surface field drop after 1,000 controlled drops?” you want (1) repeatable impacts and (2) repeatable field measurements. Here’s a concrete, investigatable methodology.

1) What you’d measure (primary outcome)

Magnetic field strength (B) at a fixed location relative to the magnet, before vs after impacts.

Best is to collect:

• B-field at a fixed standoff (e.g., 2.00 mm) on the magnet’s centerline using a Hall probe + a hard spacer.
• Optionally, a B vs distance curve (e.g., 0.5, 1, 2, 3, 5, 10 mm) to see if the whole dipole moment changed or if you’re seeing measurement artifacts.
• Optionally, a 2D “surface field map” on a grid (e.g., 5×5 points at fixed 2 mm height) to detect partial/local demagnetization.

Record also:

• Magnet temperature at each measurement session (NdFeB is temperature-sensitive).
• Visual/mechanical damage (chips/cracks) and mass (loss of material can mimic demagnetization).

2) Samples & controls (so you can trust the result)

• Buy at least 10 magnets from the same seller, same listed grade, same dimensions, ideally same production lot.
• Randomly assign:
  • Drop group: n≈5–10
  • Control group (no drops): n≈5–10 stored under identical conditions

Why: neodymium magnets vary, and measurement noise can be comparable to small changes.

3) Measurement apparatus (make “same distance” truly the same)

Gaussmeter / probe setup

• Use a Hall-effect gaussmeter with a transverse probe (common) or axial probe depending on your geometry.
• Build a nonmagnetic measurement jig:
  • A cradle that holds the magnet in the same orientation every time.
  • A rigid standoff spacer (e.g., 2.00 mm acrylic/Delrin shim) that sets probe-to-magnet distance.
  • A hard stop so the probe touches the spacer in the same way each reading.

Environment control

• Measure in a location away from steel benches, motors, speakers, etc.
• Always zero/calibrate per instrument instructions before a session.
• Keep measurement temperature stable (e.g., 20–22°C); let magnets equilibrate for 30–60 minutes if they warmed from handling.

Measurement protocol (repeatability)

For each magnet, each session:

1. Place magnet in jig (same pole facing probe).
2. Take N repeated readings (e.g., 10) lifting/reseating the probe each time.
3. Save mean and standard deviation.
4. Repeat for the other pole and/or multiple distances if doing a curve.

This tells you your measurement noise floor, which matters for interpreting “real” flux loss.

4) Impact (drop) apparatus (make every drop comparable)

Hand-dropping is usually too variable. Use a guide.

Drop rig

• A vertical drop tube (polycarbonate/PVC) slightly larger than the magnet so it doesn’t tumble.
• A release gate at a fixed height (e.g., 1.000 m measured from magnet bottom to impact surface).
• Impact surface:
  • Your stated condition: steel plate. Use a thick plate (e.g., 1/2" or thicker) mounted on a rigid base so it doesn’t bounce much.

Orientation & repeatability

Pick one impact condition and keep it fixed, for example:

• Flat face down (most consistent) with same pole facing down each time.
• Alternatively, edge impact (harsher) but much harder to control.

Drop schedule

• Do impacts in blocks so you can see if loss happens early then stabilizes:
  • Measure at: 0, 10, 50, 100, 250, 500, 750, 1000 drops
• Between blocks, allow magnets to return to the same temperature before measurement.

Handling damage vs magnetization loss

Because NdFeB is brittle, many “strength losses” after drops are actually:

• Chipping/cracking → geometry changes → field at a point changes So document after each block:
• Photos under consistent lighting
• Mass to 0.01 g if possible
• Note cracks (a cracked magnet may also partially demagnetize due to internal stress and domain changes, but geometry changes are still a confound).

5) Data you’d need and how you’d analyze it

Data table (minimum)

For each magnet:

• ID, group (drop/control)
• Initial B at 2.00 mm (mean of repeats)
• B at each checkpoint
• Temperature at measurement
• Damage score (none / chip / crack / shattered), mass

Analysis

• Compute percent change per magnet:
  [ %\Delta B = 100 \times \frac{B_{1000} - B_{0}}{B_{0}} ]
• Compare drop vs control:
  • Mean ± 95% CI of %ΔB
  • A simple t-test (or nonparametric test) on %ΔB
• Plot B vs number of drops to see whether it’s gradual or step-like (often damage-driven changes are step-like).

6) Extensions that make the result more informative (optional)

• Different impact surfaces (steel vs granite vs wood) to separate “shock” from “damage.”
• Different grades/coercivities (N52 vs N42SH etc.) to see if higher coercivity correlates with better retention.
• Measure a far-field point (larger distance, e.g., 20–50 mm) to better reflect dipole moment and reduce sensitivity to tiny chips.

If you tell me what tools you actually have (specific gaussmeter model? calipers? scale? ability to build a jig?), I can adapt this into a step-by-step protocol with tolerances and a data sheet template.

If I got a clear, quantified answer (e.g., “after 1,000 1 m drops onto steel, this N52 disc loses X% at 2 mm standoff, with/without visible damage”), a few things would change.

Decisions that would be affected

• Whether I’d trust bare neodymium magnets in impact-prone designs (tools, fixtures, latch mechanisms, DIY robotics). If the loss is negligible unless the magnet chips, I’d prioritize mechanical protection over magnetic overdesign. If there’s meaningful loss even without visible damage, I’d choose a different magnet/material or architecture.
• How I’d spec magnets: I might stop choosing “strongest grade” (N52) by default and instead pick higher-coercivity/high-temp grades (SH/UH/EH), or switch to SmCo if impact stability matters more than peak strength.
• Packaging/handling: it would inform whether “drop risk” is mostly a breakage problem (solve with coatings, rubber potting, housings) or a demagnetization problem (solve with grade choice and operating point margin).

Assumptions confirmed or overturned

• Assumption A (common): “Permanent magnets don’t really demagnetize from mechanical shock; they just crack.”
  A clear result would confirm this if the field stays constant for intact magnets and only drops when damaged.
  It would be overturned if intact magnets show consistent measurable flux loss beyond noise.
• Assumption B: “N52 is best for everything.”
  If N52 shows poorer retention under shock than lower-Br but higher-coercivity options, it would overturn the idea that maximum initial strength equals best real-world performance.
• Assumption C: “A point gauss reading reflects magnet health.”
  If results show that small chips change the reading a lot without much change in overall dipole moment, it would reinforce that measurement geometry matters and that “strength loss” can be an artifact of where you measure.

Who else would care (and why)

• Product designers / mechanical engineers using magnets in consumer goods (closures, mounts, wearables): they need to know if they must design for field aging under drops or just prevent chipping.
• Manufacturers/suppliers and QA teams: they could use the result to define incoming inspection, packaging requirements, or whether a shock/vibration qualification is meaningful.
• DIY builders/makers: helps decide whether to buy extras, pot magnets in epoxy/rubber, or avoid magnets in high-impact spots.
• Safety and reliability people (industrial fixtures, magnet-held sensors): if field loss is real, it affects factor-of-safety and maintenance intervals.

Net: a clear answer would let me treat “drops” as either primarily a mechanical durability problem (protect the magnet) or a magnetic performance degradation problem (choose a different magnet/grade and design with margin), and it would change how conservatively I design and measure magnet-based assemblies.

Next-week actions to make real progress

1. Lock the exact test definition (30 minutes)

• Pick one magnet spec (e.g., 20 mm × 5 mm N52 disc, NiCuNi coating).
• Pick one impact condition (e.g., 1.00 m drop, flat-face down onto 12 mm steel plate).
• Pick one measurement condition (e.g., B-field at 2.00 mm standoff on centerline, north pole facing probe). Write these down so you don’t drift later.

2. Get the minimum equipment lined up (same day)

• Check access to a gaussmeter + Hall probe (borrow from a lab/makerspace, or buy).
• Gather/build:
  • Nonmagnetic jig to hold the magnet (wood/acrylic/3D print)
  • A 2.00 mm spacer (acrylic shim/stacked feeler gauges)
  • Digital calipers (for spacer verification)
  • A kitchen scale (0.01–0.1 g resolution is fine)
  • Steel impact plate (thick, flat), rigid backing

3. Buy enough magnets for a control group (order today)

• Order at least 10 identical magnets (so you can do drop vs control).
• If budget allows, order an extra 2–4 as spares (NdFeB shatters sometimes).

4. Build the two fixtures (1–2 evenings)

• Measurement fixture: magnet cradle + fixed probe positioning + hard stop via spacer.
• Drop fixture: simple drop tube with a repeatable release (even a slit tube + pin works). Goal this week: “repeatable enough” beats “pretty.”

5. Do a measurement repeatability study before you drop anything (1 session) Using any magnet you already have (doesn’t have to be the N52 batch):

• Take 30 readings with full reseating each time at your chosen standoff.
• Compute the mean and standard deviation (even in a spreadsheet). This tells you your noise floor (e.g., if σ is 5–10 G, you’ll know what % change is detectable).

6. Create a simple data sheet + spreadsheet (30–60 minutes) Columns like:

• Magnet ID, group, checkpoint drops (0/10/50/…/1000)
• B (mean of 10), B std dev, temperature
• Mass, damage notes, photo filename This prevents losing track later.

7. Pilot run with 1 magnet (end of week) Once the batch arrives:

• Measure baseline for one magnet (and one control magnet).
• Do 10–50 drops only.
• Re-measure and inspect for chips/cracks; weigh it; take photos. This verifies the rig works and that the plate/drop method doesn’t instantly shatter everything.

8. Adjust protocol based on pilot (same weekend)

• If magnets chip too easily: consider a less severe surface (or accept that as a real-world finding, but then your question becomes “field loss + breakage rate”).
• If measurements drift: improve jig stability, reduce nearby steel, standardize temperature/handling.

If you tell me what you already have access to (gaussmeter? 3D printer? steel plate? budget for magnets), I can turn this into a one-week checklist with exact quantities and a ready-to-use spreadsheet layout.

Answer to Question 1 (Claude Opus 4.5)

If the pilot shows that most bare N52 discs chip/crack/shatter long before 1,000 drops onto steel, I’d treat that as a real answer—just not to the narrow “pure demagnetization” framing. It would answer a more important practical version of the question:

• In this use-case, the dominant failure mode is mechanical integrity, not gradual magnetic fading.
• Therefore the engineering decision becomes: don’t deploy bare sintered NdFeB in that shock environment (or redesign so the magnet never takes that shock).

But I wouldn’t stop there, because it’s very plausible the “magnetic degradation” question still matters conditional on survival (e.g., magnets inside housings, potted magnets, rubber-coated pot magnets). So I’d explicitly split it into two experiments that disentangle the phenomena:

1. Real-world / “as-used” survivability test (keeps the harsh steel impact):

   • Outcome: distribution of time-to-failure (chip, crack, shatter) + field change up to failure.
   • Interpretation: tells designers what actually happens if they don’t protect the magnet.

2. Isolated magnetic-stability-under-shock test (redesigned to prevent chipping):

   • Same magnet grade and measurement method, but impacts delivered in a way that greatly reduces brittle fracture while still imposing repeatable acceleration/shock, e.g.
     • magnet potted in epoxy or rubber-encapsulated,
     • or dropped onto a softer but characterized anvil (urethane pad of known durometer),
     • or mounted in a rigid holder so it can’t edge-impact/tumble.
   • Outcome: field loss vs drops when geometry remains intact.

Would the redesigned version still answer the practical question I originally cared about? Yes—because the practical question is rarely “a naked disc in free fall onto steel.” In most products, magnets are mounted/encapsulated. So I’d frame results as:

• “Bare magnet: fails mechanically at ~N drops; field change before failure is ___.”
• “Protected magnet: survives to 1,000 drops; field change is ___.”

That turns the “tension” into a useful design map rather than a contradiction.

Answer to Question 2 (Gemini 3 Pro)

If I observe a measurable field drop with no visible surface damage, my next step would be a follow-up that tests whether the loss is reversible demagnetization vs irreversible structural change (micro-cracking / internal flux-path disruption).

Follow-up experiment: remagnetization recovery test (most decisive, practical)

Goal: See whether re-saturating the magnet restores the original field.

Method:

1. Keep an undropped control magnet from the same batch.
2. Measure both magnets in the same jig (baseline B at 2.00 mm, plus optionally B vs distance).
3. Apply the same shock regimen to the test magnet until the field drop appears.
4. Then use a pulse magnetizer (or access a magnet supplier/lab that can re-magnetize) to apply a saturating magnetizing pulse in the original magnetization direction to:
   • (a) the dropped magnet
   • (b) the control magnet (as a sanity check that your magnetizer procedure is consistent)
5. Re-measure both magnets using the identical jig.

Interpretation:

• If the dropped magnet’s field returns to (near) baseline after re-magnetization (within your measurement noise and typical part-to-part scatter):
  → the loss was largely reversible demagnetization (domains partially knocked off alignment / operating point shift).
• If the dropped magnet’s field does not recover (while the control behaves normally):
  → strongly suggests irreversible change, often consistent with internal micro-cracks or internal structural damage that alters flux continuity even if the exterior looks fine.

Add-on checks to strengthen the conclusion (still doable)

• Far-field dipole moment proxy: measure B at a larger distance (e.g., 30–50 mm).
  • Geometry/local surface effects matter less; persistent reduction after remag points to internal disruption.
• Field mapping at fixed height: a small 2D scan can reveal “lumpy” flux patterns consistent with internal fracture planes.
• Non-destructive internal inspection (if available): micro-CT, ultrasound, or even dye penetrant after coating removal (CT is the cleanest, but not always accessible).

Net: the re-magnetization recovery test is the cleanest practical discriminator. It answers “did shock merely move domains (recoverable)?” vs “did shock damage the magnet’s internal structure (non-recoverable)?”, which directly informs whether you should solve the problem with higher coercivity / stabilization or with mechanical protection / different material.

1. What’s the “failure curve” for bare NdFeB under repeated impacts?
   Not just average flux loss—what is the distribution of drops-to-first-chip / first-crack / shatter, and how does that correlate with any measured field change up to failure?

2. How much of the apparent “strength loss” after impact is just geometry change vs true loss of magnetic moment?
   In other words: if I measure both near-field (2 mm) and far-field (e.g., 50 mm), do they diverge in a way that diagnoses chipping/local field distortion?

3. Can internal micro-cracking cause significant flux loss without visible damage, and how common is that?
   What fraction of “intact-looking” magnets show non-recoverable loss consistent with internal structural degradation?

4. Is impact-induced loss (if any) reversible by re-magnetization, and what does recovery look like quantitatively?
   e.g., after remagnetizing, do dropped magnets return to within 1–2% of baseline, or is there a consistent permanent deficit?

5. How do protective strategies trade off “survival” vs “magnetic performance”?
   For the same drop regimen, how do epoxy potting, rubber coating, steel cups (pot magnets), or plastic housings change (a) breakage rate and (b) measured field at a fixed standoff?

6. Does magnet grade choice change shock robustness in a way that matters more than initial Br?
   Specifically, do higher-coercivity grades (SH/UH/EH) show measurably better retention under shock when mechanically protected, even if their initial surface field is lower?

7. Is there a meaningful “stabilization” procedure for shock (analogous to thermal stabilization)?
   For example: does subjecting magnets to a controlled pre-shock / vibration regimen “settle” them so subsequent shocks cause less flux change?

8. What is the right practical spec to report for engineers: flux loss, dipole moment loss, or functional holding force loss?
   Which metric best predicts real device performance after drops, and how do you translate gaussmeter readings into that metric with minimal ambiguity?