FI EFFICIENCY · DEVICE RESEARCH
Use less energy, and you can choose where the benefit goes: more time between charges, a smaller battery or less cooling. FI Efficiency explores those choices with computer models.
COMPUTER-MODEL RESULTS
A sensor simulation and a design for storing recordings.
SAME SENSOR · SAME BATTERY
29.94%About 13 hours between charges, up from 10.
See how it worksSENSOR SIMULATION
23.04%Fewer readings, with the same quality in the model.
See the sensor studyRECORDER DESIGN
19.97%Keep all 30 days—including the original data.
See the recording designOur sensor example gains almost three hours per charge by taking fewer measurements. The calculation also counts the energy needed to choose them.
The studies begin with two jobs: taking measurements and keeping records.
One sensor study uses 23 readings instead of 48, at the same modeled answer quality.
Both recorder designs keep 30 days of recordings and the complete input history. One sets aside less storage.
The sensor saving can go toward more time between charges or a smaller battery and cooling assembly.
CALCULATIONS & TRADEOFFS
Open a study for the calculations and design choices. Our latest research update covers medical sensing, energy recovery, power-supply heat and shared radio hardware.
Complete comparisons and sources Latest research updateThe best saved sensing case uses 23 readings instead of 48 at the same modeled quality and job rate. Readings account for 50% of the original energy; planning adds 3% of that original total. After planning, the saving is 23.0416667%.
In the 10 W sensor model, average input falls to 7.6958333 W. With the same 100 Wh of usable battery energy, runtime rises from 10 hours to 12.9940444 hours: 29.940444% longer between charges. This assumes the same operating profile, conversion efficiency and battery cutoff.
Exact sensor inputs and resultsKeeping the original 10-hour runtime lets the model use a smaller battery instead. The 1 kg design starts with a 250 g battery and 150 g of cooling. Scaling both with power gives 9.22% less device mass. Battery energy density and the cooling-scaling rule stay fixed, all other parts stay in place and the model adds no parts.
The cooling calculation assumes the design heat load falls by the same fraction, with the same ambient temperature, surface temperature and heat transfer per area. The modeled sizes assume peak loads, structure and layout allow those smaller parts.
Battery and cooling calculationsConventional exact planners match the selected reading counts. In three equal-noise cases, no readings are avoided and planning raises energy use by 3%. The same battery then lasts 9.71 hours instead of 10. One other case saves 1.17% energy; eight selected cases save 3.25–23.04%.
All 12 sensing casesCombined recorder and input-history storage falls from 1,453,784,064 to 1,163,480,256 bytes, a 19.97% reduction. Both designs keep all records and the complete input history for the same 30-day task.
The recorder reduces its routine restart scan by postponing the full corruption check. The design assumes one writer and no changes to storage that was checked earlier.
Recording design and recovery tradeoffIn the complete radio reference-job study, nominal savings disappeared under some tested conditions. The comparison includes ties with ordinary designs, slower completion and energy losses.
Radio results and comparisonsShare what you’re building and what you want to improve.
