Choosing the right microSD by use case — a quick selection guide
HomeResourcesChoosing the right microSD by use case — a quick selection guide
Guides · Selection

Choosing the right microSD by use case — a quick selection guide

By Kalstor 7 min read
Key takeaways
  • Match the exact card to measured workload and host requirements; a category label alone does not establish service life or compatibility.
  • Decide in order: host support, aggregate write load, retention, supplier endurance evidence, speed class, temperature and real-device tests.
  • Speed class is a floor, not a guess: V10/U1 = 10 MB/s, V30/U3 = 30 MB/s minimum sustained write. Pick the class that clears your worst-case stream with headroom.
  • For vehicle or harsh use, require the exact operating range and configuration-control terms for the quoted SKU; do not infer them from the word industrial.

There's no universal "best microSD." A phone, action camera and circular recorder generate different write patterns, but category names do not predict a fixed lifetime. Selection begins with the host, measured workload and evidence for the exact card.

Here's the short version as a table, then the order to decide in.

The selection table

Use caseWrite patternRecommended gradeSpeed classTemperature
Security camera / CCTVContinuous, 24/7 loopHigh-enduranceV10–V30Standard or wide
Dashcam (1080p)Continuous, in-vehicleHigh-enduranceU1 / V10+Wide (−25 to 85 °C)
Dashcam (4K / dual-channel)Continuous, high bitrateHigh-endurance (industrial for fleets)U3 / V30Wide
Body-camHeavy, rugged, burstyHigh-endurance / industrialU3 / V30Wide
IP camera / IPCContinuousHigh-enduranceV10–V30Standard
Industrial / IoT / embeddedMixed, long service lifeConfiguration-controlled candidatePer hostExact SKU range
Vehicle equipmentHeat, cold and power transitionsVehicle-use candidate with documented evidencePer hostMeasured environment versus SKU
Phone / tablet / generalLight, occasionalConsumerU1 / V10 (V30 for 4K capture)Standard

How to read it — decide in this order

  1. Host support first. Use the device manual and compatibility list to establish capacity, filesystem, bus mode and minimum speed class.
  2. Measure the stream. Aggregate bitrate and duty cycle determine retention and annual host writes; the retention calculator shows the arithmetic.
  3. Check exact-SKU evidence. Compare rated hours or endurance, operating temperature and warranty under their published conditions. Do not transfer a number from another capacity or family.
  4. Qualify in the host. Verify full addressability, repeated wraparound, peak scenes, temperature and credible power interruptions before release.

Two traps the table can't fix

A grade is only as good as the card you actually receive. Two things still bite buyers:

  • Faked capacity. A reprogrammed controller reports the right size and fails when written past its real limit. Always write-verify a sample — see why a 128GB card shows 119GB.
  • Wrong supplier. In a shortage, re-graded and counterfeit stock floods the channel. Run the supplier checklist before a volume order.

Bottom line

Pick from host evidence and measured workload, not a category name. Kalstor can review the application and confirm a proposed SKU, current commercial terms and available evidence for an RFQ; availability and fixed-configuration scope are not implied by this guide.

FAQ

What microSD should I use for a security camera or dashcam?
Start with the camera manual and compatibility list, then measure aggregate bitrate and temperature. Compare exact-SKU endurance evidence and test repeated overwrite in the real host. V30 or a wide-temperature grade is required only when the host and environment call for it.
Is a higher speed class (V30) always better?
Only up to what you need. Speed class is the minimum sustained write speed; you want enough to clear your camera's stream with headroom, not the highest number on the shelf. A V30 card with low endurance still wears out fast — endurance and speed are separate choices.
How big a card should I buy?
Size capacity for the required retention window, then compare the resulting annual host-write load with the exact SKU's documented endurance conditions. More capacity can reduce rewrite frequency, but it does not create a universal life multiplier.
Sourcing in volume?

We publish measured usable capacity and welcome trial-batch verification — automotive-grade, direct from the source factory.