https://jaycorp.us/instrument-platforms.html

Skip to content (https://jaycorp.us/instrument-platforms.html#main)
jaycorp.us (https://jaycorp.us/) / Research

Beyond radio: programmable instruments, ranked

Published and checked 18 October 2026 · Plaintext (https://jaycorp.us/instrument-platforms.txt) · Changelog (https://jaycorp.us/changelog.html)

The best way to get “HackRF-grade capabilities in new mediums” is to combine a programmable acquisition platform with the right transducer. Start where a small experiment can reveal something real, then add more ambitious sensing once the measurements are trustworthy.

Editorial guide · checked 18 October 2026. S+ through F ranks practical value for a self-hosted, programmable research bench—not a measured performance benchmark. The proposed buying order is advice, not a purchase. Prices are listed USD prices before tax, shipping and unspecified accessories; stock can change. Unknown prices remain unknown. None of these new instruments has been installed or tested here.

Companions: what the existing HackRF and RSP1 can do (https://jaycorp.us/sdr-tempest-research.html) · consciousness, biometrics and evidence (https://bada.im/signals-brains-and-evidence.html) · historical foundations at AOL (https://aol.pm/foundations.html).

Contents
• Tier list (https://jaycorp.us/instrument-platforms.html#tiers)
• Electronics and protocols (https://jaycorp.us/instrument-platforms.html#electronics)
• Acoustics (https://jaycorp.us/instrument-platforms.html#acoustics)
• Optical and thermal (https://jaycorp.us/instrument-platforms.html#optics)
• Motion and radar (https://jaycorp.us/instrument-platforms.html#motion)
• Bioelectric signals (https://jaycorp.us/instrument-platforms.html#bioelectric)
• Wilder, real instruments (https://jaycorp.us/instrument-platforms.html#experimental)
• Speculation (https://jaycorp.us/instrument-platforms.html#speculation)
• Order of purchase (https://jaycorp.us/instrument-platforms.html#order)

The quick tier list

Five considerations drive these judgments: useful questions for this project, access to samples and programming, reproducibility, total cost including accessories, and maturity of the supporting evidence. Unknown cost or availability reduces confidence. A specialized B-tier instrument can be the best tool for its particular job. Increasing experimental ambition is a separate axis from quality.

Editorial priorities; read the model and accessory caveats below

Tier | Platforms or choices | Why this placement |

S+ | Analog Discovery 3 (https://jaycorp.us/instrument-platforms.html#ad3) | The strongest general-purpose addition: observe and generate signals, then reuse the same bench across many transducers. |

S | Glasgow (https://jaycorp.us/instrument-platforms.html#glasgow); AudioMoth (https://jaycorp.us/instrument-platforms.html#audiomoth); OPT101 plus an acquisition platform (https://jaycorp.us/instrument-platforms.html#opt101) | High exploratory value with inspectable interfaces. Glasgow revision uncertainty and unknown prices need resolution before buying. |

A | STEMlab 125-14 (https://jaycorp.us/instrument-platforms.html#red-pitaya); MCHStreamer (https://jaycorp.us/instrument-platforms.html#audio-bridge); global-shutter camera (https://jaycorp.us/instrument-platforms.html#camera); AS7341 (https://jaycorp.us/instrument-platforms.html#spectral); MLX90640 (https://jaycorp.us/instrument-platforms.html#thermal); ICM-20948 (https://jaycorp.us/instrument-platforms.html#imu) | Useful, programmable capabilities with more specific use cases or additional integration work. |

B | TI radar evaluation boards (https://jaycorp.us/instrument-platforms.html#radar); OpenBCI Cyton (https://jaycorp.us/instrument-platforms.html#cyton); Polar H10 reference (https://jaycorp.us/instrument-platforms.html#h10); seismic and particle research builds (https://jaycorp.us/instrument-platforms.html#experimental) | Credible specialist tools. Setup, interpretation, validation or total cost dominate the decision. |

C | A synchronized instrument backpack | A real integration project with substantial clock, calibration and provenance work. Build it after its individual instruments work. |

D | Buying a sensor collection before defining an experiment | A weak purchasing strategy, not a judgment that the sensors are defective. |

E | Promoting a graph directly into a claim about identity, emotion or causation | A measurement may be real while its interpretation is unsupported. |

F | Purchasing a purported portal, light-body or universal thought scanner without validated evidence | No adequate evidence for the promised capability. A compelling story cannot substitute for an independently testable instrument. |

A receiver measures; a generator creates a signal. Some platforms do both, and some need a separate sensor or actuator. Radar remains radio-frequency sensing, even though its output is motion rather than communication. A microphone, photodiode or electrode introduces a genuinely different physical interface.

1. Electronics and protocols: the most reusable foundation

S+ · Digilent Analog Discovery 3

Best first addition: automated tests of owned, low-voltage circuits and transducers. AD3 (https://digilent.com/shop/analog-discovery-3/) lists two 14-bit, 125 MS/s scope inputs, two waveform outputs, sixteen digital I/O and WaveForms SDK access. Published 30+ MHz scope and 12 MHz output bandwidth require the BNC adapter.

Budget: $379 plus a $22 adapter, or $401 before probes and other extras. Access: vendor software and SDK; do not assume an open FPGA development platform. Keep this bench to low-voltage electronics: no direct mains or body connections.

The persuasive case is reuse. One acquisition-and-generation platform can support a light sensor today, a filter response tomorrow and repeatable protocol tests later. Write down the voltage limits and required probes before deciding that the base price is the total cost.

S · Glasgow Interface Explorer

Best use: exploring owned digital interfaces with open hardware and programmable protocol applets (https://glasgow-embedded.org/en/intro.html). Its Python and Amaranth approach makes the interface itself a project. Need: suitable leads, level-compatible targets and the right revision.

Revision gate: current documentation (https://glasgow-embedded.org/en/index.html) labels revD pre-launch and says its built-in cycle-accurate analyzer was removed pending rework. Do not treat every documented feature as shipping. Price and availability (https://glasgow-embedded.org/en/purchase.html) remain unverified here. This S rating is conditional on matching the chosen hardware and firmware to the intended experiment.

A · Red Pitaya STEMlab 125-14, original model

Best use: an instrument that can become part of a larger software/FPGA experiment. The original STEMlab 125-14 (https://redpitaya.com/stemlab-125-14/) provides two 125 MS/s, 14-bit inputs and two outputs, network control and programmable FPGA/software interfaces. Need: probes, cabling, suitable signal conditioning and any application-specific expansion.

Price: current complete-kit cost unknown. These are original-model specifications, not Gen2 specifications. Open software does not imply that every hardware design is open: the separate Gen2 documentation (https://redpitaya.readthedocs.io/en/latest/developerGuide/hardware/GEN2/125-14_Gen2/top.html) distinguishes its software from non-open hardware schematics. Choose an exact model before comparing clocks, connectors or capabilities.

2. Acoustics: a spectrum you can hear and measure

S · AudioMoth

Best use: passive acoustic and ultrasound observations with exportable recordings. AudioMoth (https://www.openacousticdevices.info/audiomoth) records raw WAV at sample rates up to 384 ksample/s. It is a receiver; that rate does not guarantee flat microphone bandwidth or calibrated sound-pressure measurements.

Need: batteries, microSD and an appropriate enclosure; price unknown. The project’s openness policy (https://www.openacousticdevices.info/open-source) uses MIT software/firmware licensing but delays the newest hardware revision’s release until a successor. The live interface (https://www.openacousticdevices.info/live) is another workflow to check against the chosen revision. Start with environmental or owned-device sounds that do not capture private conversations.

A · miniDSP MCHStreamer

Best use: building a custom audio acquisition/generation chain. The $105 digital USB audio bridge (https://www.minidsp.com/products/usb-audio-interface/mchstreamer) supports formats and rates up to 384 kHz, with channel count dependent on format and firmware. Need: compatible ADCs/DACs, microphones or other transducers, power and wiring. It is not an analogue microphone input by itself.

A calibrated microphone path is a better starting point for an acoustic claim than guessing from a radio waterfall. For neuroscience questions, the binaural-beat evidence review (https://bada.im/signals-brains-and-evidence.html#beats) distinguishes an auditory response, a subjective experience and a clinical claim. No sound-exposure treatment is proposed here.

3. Optical and thermal: intensity, images and coarse spectra

S · TI OPT101 plus the bench instrument

Best use: light intensity over time—LED modulation, flicker and timing. OPT101 (https://www.ti.com/product/OPT101) combines a photodiode and amplifier; its default bandwidth is 14 kHz (https://www.ti.com/lit/ds/symlink/opt101.pdf). It produces an analogue signal for an ADC or scope.

Need: power, wiring, a stable fixture and acquisition hardware; total price unknown. Begin with modest LEDs. It measures neither optical phase nor a spectrum, and more scope samples cannot restore light variations beyond the sensor’s response.

A · Raspberry Pi Global Shutter Camera

Best use: controlled motion imaging without rolling-shutter distortion. The manufacturer specifies RAW10 and up to 1456 × 1088 at 60 Hz (https://www.raspberrypi.com/products/raspberry-pi-global-shutter-camera/); camera software (https://www.raspberrypi.com/documentation/computers/camera_software.html) exposes programmable capture. A short minimum exposure is not the reciprocal frame rate.

Budget: Adafruit listed $50, out of stock (https://www.adafruit.com/product/5702) at the check. Add a compatible Pi, lens, suitable cable, power and lighting. The sensor price is not a complete camera system.

A · Adafruit AS7341 breakout

Best use: compare controlled light sources across broad bands. $18.95, listed in stock (https://www.adafruit.com/product/4698); eight overlapping visible bands plus clear/NIR/flicker functions, with libraries and schematics. Need: host, cable, calibration and controlled illumination. This is a coarse spectral sensor, not a high-resolution spectrometer or automatic chemical identifier.

A · Adafruit MLX90640 breakout, 110° field of view

Best use: thermal patterns on owned equipment. $74.95, listed out of stock (https://www.adafruit.com/product/4469); a 32 × 24 array with Python/Arduino access and practical 16 Hz operation described by the seller. Need: host, cables and attention to emissivity and reflections. Image interpolation adds no measured pixels; subpage refresh and complete-frame rate are not interchangeable.

4. Motion, fields and radar: useful, but different measurements

A · Adafruit ICM-20948 breakout

Best use: orientation and motion experiments with raw accelerometer, gyroscope and magnetometer readings over I²C/SPI. $19.95, listed in stock (https://www.adafruit.com/product/4554). Need: host, wiring, mounting and calibration. Integrating acceleration drifts; metal and currents affect magnetic readings. This is not a brain-field or gravity-anomaly detector.

B · TI IWR6843ISK or IWRL6432BOOST

Best use: a separately planned motion/radar experiment. IWR6843ISK (https://www.ti.com/tool/IWR6843ISK) is the 60–64 GHz, four-RX/three-TX option discussed in the companion article. IWRL6432BOOST (https://www.ti.com/tool/IWRL6432BOOST) is a different 57–64 GHz, three-RX/two-TX board. Both remain RF instruments; they do not add a new physical medium.

Access and cost: USB demo outputs are not raw ADC data; raw capture requires a DCA1000EVM and a compatible capture workflow. Board and complete-system prices are unverified. Separate algorithms, calibration and validation are needed for a physiological claim. Neither board is a ready-made identity authenticator. The cardiac feasibility analysis (https://jaycorp.us/sdr-tempest-research.html#cardiac) explains why the existing HackRF/RSP1 pair is not a drop-in substitute.

5. Bioelectric signals: a separate, consented research branch

B · OpenBCI Cyton

Best use: passive EEG, EMG or ECG research with appropriate electrodes. The eight-channel board lists $1,249 (https://shop.openbci.com/products/cyton-biosensing-board-8-channel), with 24-bit channels at 250 Hz. GUI and raw-data workflows (https://docs.openbci.com/Software/OpenBCISoftware/GUIDocs/) support analysis. Need: electrodes, leads and a suitable reference/acquisition plan; board cost alone is incomplete.

This is non-medical research equipment, not a stimulator or universal thought decoder. Follow manufacturer battery and safety instructions (https://docs.openbci.com/Cyton/CytonSpecs/); do not connect the general-purpose bench directly to a person. No acquisition or setup is authorized by this guide.

B · Polar H10 as a cardiac reference

Best use: a contact reference in a future consented timing study. The official SDK documentation (https://github.com/polarofficial/polar-ble-sdk/blob/master/documentation/products/PolarH10.md) exposes ECG and beat-interval data. Need: chest strap, compatible host and synchronization with the other instrument; current total price unknown. Contact ECG and radar motion differ physically, and neither becomes a secure login merely by producing a waveform.

For actual authentication, use the separate presence, identity and intent proposal (https://bada.im/signals-brains-and-evidence.html#authentication): credential proof, local verification, explicit action approval and a usable recovery route. Physiological collection remains optional and separate from account access.

6. Wilder, but real: vibrations, particles and combined instruments

B · Raspberry Shake and Shake & Boom research

Best use: passive ground-motion and, with the appropriate model, infrasound studies. The manufacturer distinguishes geophone channels from pressure channels (https://raspberryshake.org/news/things-you-might-find-with-your-raspberry-shake-pt1/). A quiet installation, mounting, power, storage and environmental context matter.

Current model selection, acquisition interface and complete-kit cost need a separate check. Treat this as a research shortlist, not a buy-now recommendation. A large transient can be a nearby ordinary event; corroborate it with timing and independent observations before identifying a cause.

B · Cosmic Pi research build

Best use: passive particle-event detection and coincidence/statistical analysis. Cosmic Pi (https://cosmicpi.org/) and its CERN-hosted project outline (https://indico.cern.ch/event/656460/contributions/2718491/attachments/1678858/2696480/Cosmic_Pi_outline.pdf) describe a cosmic-ray detector project. Detector assembly, electronics, timestamped data and background characterization are part of the work.

Current kit availability, supported revision and total cost are unverified. This is a research build using naturally occurring events, with no radioactive source acquisition proposed. Particle counts are not a portal detector.

C · A synchronized instrument backpack

Combine an acoustic channel, a light channel, motion, temperature and an existing RF receiver. The interesting output is a common event timeline with raw evidence and uncertainty. Start by showing that the channels can reliably recognize the same known event. Clock drift, sensor latency and automatic gain can manufacture apparent leads and lags.

The C rating reflects integration work, not a lack of real physics. A large stack of sensors does not automatically become a better instrument. Preserve each sensor’s provenance and make “unknown” a valid result.

7. Where ambition needs evidence

Light-body, dimensional-root and portal language can be discussed as spiritual or literary framing. The companion essay attributes the Drunvalo-linked tradition (https://bada.im/signals-brains-and-evidence.html#light-bodies) without treating it as a measured RF identity. None of the instruments here validates a literal portal, arbitrary thought insertion or a universal consciousness scanner.

The F tier applies to spending on those promised capabilities without adequate evidence. It does not apply to a specialized instrument merely because its physics is unfamiliar. A proposed extraordinary detector should first name a measurable quantity, a calibration method, a blinded control and an independent replication. Buying more sensitivity cannot repair an undefined claim.

A proposed order of purchase

• Use what is already present. Keep the existing radios with their current owners; plan offline analyses and a small, owned-source measurement before acquiring another RF platform.
• Add one general bench. AD3 is the first candidate for broad utility. Resolve probes, adapter and total cost. Choose STEMlab instead only when its FPGA/network role solves a defined requirement; there is no need to buy both immediately.
• Add one new physical interface. Choose OPT101 for light timing or AudioMoth for passive acoustics. A repeatable experiment with raw exports is the acceptance test.
• Add specialist imaging or motion. Pick the camera, thermal array, spectral sensor or IMU that answers the next question. Check stock and the full host/accessory budget.
• Make a separate decision about radar or human data. Require a testable objective, consent and retention plan where appropriate, a reference measurement and a realistic validation budget. Keep real authentication on credentials.
• Earn the ambitious build. Attempt a combined backpack, seismic installation or particle detector after the simpler instruments produce reliable data.
Before approving a purchase, ask for the exact model/revision, base price, all accessories, working software path, first experiment and failure criterion. The best platform makes an honest negative result easy to obtain and explain. That is the capability worth expanding across mediums.

Original analysis with linked sources. Research proposals are explicitly labeled; no deployment or clinical validation is implied. Disclaimer (https://jaycorp.us/disclaimer.html) · jaycorp.us (https://jaycorp.us/) · Changelog (https://jaycorp.us/changelog.html)