Why

We train students in the intricacy of solid state physics, analog circuit design, digital circuit design, circuit implementation, project management and team work, but I think we sometimes forget to tell them about the value of a problem.

In academia we have extremely bright people working on fantastic technical solutions and innovations. In the absence of real problems, academics will invent theoretical problems, and sometimes, the problems they invent should not be solved.

Inventing problems

One of my pet peeves is the horrible efficiency of ambient energy harvesting, the idea that one can harvest the ambient 2.4 GHz radio waves that are all around us. A normal phone transmitting Bluetooh will transmit at 10 dBm (10 mW), or maybe up to 20 dBm (100 mW) in some instances. Assuming an antenna that transmits equally in all directions we loose about 40 dB on the first meter. After the first meter we loose between 6 dB to 12 dB per doubling of distance, sometimes even more. If I place an energy harvester 1 meter way from my transmitter, I can maximum get -20 dBm (10 $\mu$W), but normally I would get -30 dBm (1 $\mu$W). That’s not a lot of energy. An invented problem is to make a more efficient energy harvester of that insignificant energy.

People do work on more efficient harvesters, take Design of Sub-Gigahertz Reconfigurable RF Energy Harvester From −22 to 4 dBm With 99.8% Peak MPPT Power Efficiency. The publication is in Journal of Solid State Circuits, which means the work is excellente, and they have measurements of real silicon. I’m sure the technical details, the circuits cool, and the people smart.

I question the reason to do the work from the get go. The minimum input power is -22 dBm, which means the transmitter needs to be pretty close (although the frequency is approx 1 GHz, which makes the path loss in the first meter a bit less).

Best case, the system efficiency (Power harvested divided by Power transmitted) is 1/1000, or 0.1 %, a horrible efficiency. So why work on the problem in the first place?

The power of real problems

There are real problems in this world. In industry, we are lucky, problems come and find us. Customers will sometimes tell us what they are struggling with. Our circuits will fail, and we have to find inventive solutions. Specifications can be impossible, which can drive innovations of new solutions.

In case of energy harvesting, I think the real problem is the system efficiency, so that’s what should be attacked first. Is it possible to have beamforming in order to improve the energy transfer between transmitter and reciever? Can a beamforming antenna scan the 3 dimensional space in search of IoT devices, and direct energy efficiently to the reciever? Like a laser? That could make the system efficiency acceptable.

The power of a real problem is that you know it’s valuable. I think Elon’s way of thinking is a good example. For example:

  1. Humans have to become a multiplanatery species if we are to survive long term
  2. Mars is our closest option.
  3. To make travel to Mars low cost, and efficient we need vertical take of and landing rockets
  4. Solve VTOL rockets.

There is a clear link from the ultimate goal “Saving the human race” to the “Solve VTOL rockets”

A collection of real problems

Maybe I’ll add more with time

Sub 100 mV oscillators (as low as possible)

If there is a temperature differential in a system, we can use thermoelectric generators to harvest energy.

Thermoelectric generators can provide significant current, but the voltage is low. See Energy Soruces for details.

In order to drive a circuit we need to boost the voltage up multiple hundred millivolts.

To boost a voltage, we need a clock.

Design a “as low as possible” oscillator of around a kHz