Category: Uncategorized

This isn’t free energy.

Harvesting byproducts very slowly can power a microcontroller and sensor wakes from deep sleep just to take a reading and broadcast the results. I’ve looked into it but these are mostly just placeholders for now. If you can charge up a supercapacitor then it can provide power for days, we’d just be filling it up as it was used. These are all big power, small power we are talking about ambient body heat, and the movement of your wrist powering mechanical watches.

I learned using TI, but they didn’t have an easy wireless solution. Their MSP432 has 95uA/MHz and 850nA when in deep sleep forgetting everything. Possibly having one WiFi base with an RF receiver and have sensor nodes using low power RF to talk to it. MPPT maximum point power tracking is a thing to know.

Mechanical: think car or bridge vibrational.
Heat: Doubtful you’re using all 360º of a  fire, Peltier cells are the most basic of this. Apply energy and get one side hot and the other cold, apply hot and cold and you get energy.
Light: Ambient solar

Low power modes can be used in everything so it would be good to make it a best practice sooner rather than later.

Do you use closed-captioning?
It’s not for you, it’s for the deaf. Huge learning of mine, making something accessible makes it better for everyone.

I don’t know where to start, so this is a placeholder for people to comment and for future me to remember.

A good solder joint creates a electrical and mechanical bond. You can test continuity for electrical but mechanical is hard.

I guess you don’t need to see if a spool of components is labeled and loaded into a pick and place machine and the whole board is wave soldered. A solderless breadboard is already a thing and the different lengths of the anode and cathode leads. I can draw circuits. Maybe a conductive putty would work, not too mechanical, maybe something like Sugru or Epoxy that was workable for a set amount of time and turned solid.

A simple tester for if an LED is lit and what color/frequency it is, also look up what kind of oscilloscope display they have for the blind, I Kickstarted a sound representation.

My dentist tells me I grind my teeth, it probably isn’t every night, but we have no way of telling- challenge accepted.

• Grinding teeth, pressure resistor in thin mouth guard
• Breathing, holding a mirror in front of the mouth/nose senses breath, so humidity if a sensor sees changes fast enough
• Pulse & Oximetry, a finger probe but I wonder what kind of belt, or if something close to the temples or neck could get a good reading
Movement, accelerometer, a gyroscope would give what side you sleep on as well if calibrated

One of my sisters wore a mask for monitoring breathing while she slept, it was a large piece of lab equipment a ‘darth vader mask’, there is room for improvement. Seems that most current sensors are external/environmental, that makes them not intrusive at all:
mattress
bedside alarm
pillow (person can move off)
pyjamas (not always the same)

Early stages, does breath (CO2), heat (IR), or a chemical (maybe pheromones) attract them most?

Tough to test variables to isolate would be the: size and amount.

A few traps next to each other would help isolate for the environment. Different concentrations of CO2. Different lure sizes of CO2. Also with IR the blinking and intensity rates, different sizes of panels, simulate movement with patterns, test light diffusers.

There is also the control area, if you get really big I’m sure that Mosquitoes from a longer distance would be attracted so it would look more effective but actually not be, just attracting them from longer distances, you’d have to control but as soon as you put them in a large greenhouse type thing you are loosing the natural CO2 and IR from fauna and the city. So size and effectiveness I’ll take with a grain of salt.

A way to zap them is tempting. I wonder what kind of wind they can sense or fly away from, having an air intake drawing them in when close may be effective, you can create an air ionizer I wonder if a bug could store a large enough charge that the shock from grounding (like us being charged and touching a door handle) could be enough to kill. That may solve them avoiding, if they do, wind. They would be charged up and pulled to discharging/death.

The size, time of life, and the millions they kill with malaria etc. mosquitoes are exempt from compassion.

References: Cubebot, Wall•e

Wall•e is a box that expands to a robot. I don’t think I’d design anything as advanced as a biped, but it would be interesting to see what you could do with space that could be expanded. An umbrella, antenna, parachute, could all be in a small area and expanded out. What about a passive heatsink design that was a solid block but expanded or fanned out for cooling. Shipping of products would be different if all the empty space was taken out, but it could expand. Foam mattresses are recent proof of this.

Reference: Ultrasonic Anemometer, Anemometer, Carl47

This would also tell you the wind direction.

When using a set distance ultrasonic distance sensors sense the air motion. Not a doppler effect, but the same idea. First to find out how accurate would it be using a HC-SR04s. I have 200kHz ultrasonic transducers, I’d need to find out how to drive them and what I’d need to receive.

For multiple ultrasonic sensor considerations
http://maxbotix.com/articles/031.htm

Ideas in security:
1. Physically secure the device. e.g. clear box, you can’t (within reason) remove it.
2. Trigger an alarm if contact is removed. e.g. alarm switch attached, you will get caught.
3. Destroy the product if it is removed. e.g. clothes ink tags, the value of what you are taking is nothing.

There is physiological reasoning behind each. You can’t easily replace the item. You want a person to be aware of an alarm. It would be easy to disguise or don’t want to stop the person.

References: BFO metal detector, Magnetic Field Measurement, Teardown.

First thing, I should be to recreate what has been done, once I have a good understanding of the principles I could customize the design. One coil emits and the other coil receives, there is a larger refection when a metal is present.

Magnetometers are very sensitive MEMS devices, I wonder if you can attach an external antenna to one of those.

References: M&M Color Sorter, Automatic Sorter, faster.

The challenge for me would be speed.

References: Robot Arm, uArm Open Source

I have a few servos, bearings, and a CNC. The plans are for laser cut or 3D printed so I’d modify. It would be interesting to write a library for natural motion, using something like motion curves.