Wilf Rigter's E-Field Sensors

From BEAM Robotics Wiki

Master BEAMer, Wilf Rigter, introduced two simple yet effective E-field detectors that can easily be incorporated into almost any BEAM robot. Either should work equally well; but the 2N7000 version may be easier to get parts for.

NOTE: The human body is a good absorber of stray RF fields, so these sensors should be usable as people-detectors.

[edit] Wilf's Description & Comments

The following is derived from Message #5789, alt-beam email list, and is Wilf Rigter's description of these simple E-field sensor circuits:

Begin Quote 1

The mosfet circuit has the advantage of an adjustable operating point which can also be used to adjust sensitivity. The temperature and +V / gain and set point stability is not great but it's simple.

The CMOS opamp design has good stability and is capable of much higher gain by replacing the 1M feedback resistor with 100K or even 10K.

I found that these designs provide a usable output when sensing my hand 24 inch away from the 8 inch antenna. In the process of experimenting

I discovered a very interesting coupled oscillator design which uses the CMOS opamp as a square wave oscillator tuned to 80Hz but which is locked to the 60 Hz E-field. The summing node is the noninverting input which is used for hysteresis feedback and the antenna input. When I approach the 8 inch antenna, from 48 inch away the frequency suddenly jumps to 120Hz and then approaching closer switches to successively higher harmonics of 60 Hz.

End Quote

In a later post (alt-beam message #5805) Wilf said...

Begin Quote 2

I expected a few questions about the function of the two back to back Si diodes. They function as a very high resistance bias resistor since their leakage current is very small at room temperature but is much larger than the mosfet gate leakage current. The voltage at the wiper of the pot will leak to the gate and set the operating point of the FET.

The alternative to Si diodes is to use a >1000M resistor which is not commonly available. To effectively use such high resistances, stray leakage currents to other parts of the circuit also has to be controlled, ideally using a Teflon standoff as the junction point for the FET gate and antenna (and Si diode if used).

E-fields are invisible complex , fluctuating electrostatic fields and would be interesting to map in 3D.

The predominant fluctuating (ac) E-field comes from the household 50/60Hz power circuit. The ac sensitivity of the sensor is proportional to the stray capacitance to GND and the AC signal at the output signal may actually go down as you bring your hand closer to the antenna coupling stray capacitance between your body and GND. If were are closely coupled to the power circuit (danger!) the signal would of course increase. The dc field is the difference in static electron charges but this is not as simple as it seems since charges (and E-fields) are displaced (disturbed) by attraction or repulsion of other charges. Charges are also carried by gas ions so the picture is quite muddled.. Since the leakage and bias currents "leak" the induced gate charge, the detector is not stable at detecting true dc E-field but can detect slow changes in voltage as E-fields change.

One thing you should consider is that the sensor detects the charge between itself and other objects equal to the difference in the charges. So an object with zero charge can still be sensed if the sensor platform itself has a charge with respect to other objects. If the charges get large walking across a nylon rug on a dry day, think "fried" mosfet. If the charges get very large think Lightning! Try to observe the output of you detector during a thunder shower and you may get an indication just before a lighting discharge of the very large e-field between earth and clouds. Place the E-field detector inside a metal screened box and note the Faraday cage effect] (all in the safety of your home of course).

Finally, one more thought: if a "delta e-phobic" bot approaches the edge of a table there should be a large change in the e-field which may (perhaps) be used to reverse and avoid "Geronimo"!

Anyway those are some of my observations.

End Quote

In a later post (Message #31041) to the BEAM email list [Wilf]] had this to say about the FET based circuit:

Begin Quote

That E field (electro-static voltage) sensor just amplifies a changing voltage field strength capacitively coupled to the gate of a high impedance FET. The output goes up or down or oscillates.

A much nicer circuit is the proximity detector which measures stray capacitance to nearby objects and provides an output when the capacitance is above some threshold.

I don't think we have yet designed such a circuit although the capaciflector operates on such a principle.

But there are some exixting circuits such as Steven Bolt's Green Thumb soil moisture detector which can be adapted. That circuit uses two RC time constants to compare the value of a known resistor with the unkown soil resistance. But if you can compare resitances you can also compare capacitances. By exchanging the RC components and using an antenna instead of a soil probe, the circuit turns into a proximity detector.

End Quote