Help understanding a simple static sensor

Thread Starter

AndyD-UK

Joined Nov 1, 2019
24
Hi all,

I hope that this is in the right place (sorry if it isn't).

I am a bit of an avid ghost hunter and saw a simple sensor that picks up static using a PP3 battery, LED and MPF102 transistor.

While this works, sort of, when I power it on, the LED tends to stay lit, goes off when I put my hand near it, then either stays off or slowly comes back on. Sometimes it seems to work in reverse and I am struggling to understand what it is doing.

It might help me design this properly if I understood how the transistor is working - it is hot-glued to a PPC enclosure for ease but could this be impacting how it works at all?

Thanks,

Andy
 

DickCappels

Joined Aug 21, 2008
10,152
Yep, that's about as simple as an electronic electrostatic filed sensor can be.

The transistor is a Junction Field Effect Transistor, or "JFET". It has three connections. When the gate voltage is near the source voltage, current can flow through the transistor. Conventional current would then flow from the drain to the source, and that current would illuminate the LED.

When the gate is negative with respect to the source the gate of the deice forms a capacitor (which is really a reverse-biased P-N junction) with the source, so when it gets charged up to the source voltage the reverse bias on the junction goes away MOSFET will tend to stay on until something makes the the charge across that capacitance go the other way (gate goes negative). When it gets charged negative with respect to the source the MOSFET will be "very" off and though there will be a little bit of leakage current across the reverse biased junction, it will be very low and the gate voltage can be influenced by charged items, such as your hand moving near the gate lead.

Be careful with static electricity because if the reverse gate-to-source voltage gets too high it will short the gate to the source and the device will not work again.

(Edited)
 

Thread Starter

AndyD-UK

Joined Nov 1, 2019
24
Thank you both very much - so it sounds like this is working as intended and is just incredibly sensitive - perhaps too sensitive if the above article is anything to go by?
 

DickCappels

Joined Aug 21, 2008
10,152
I had misidentified the type of transistor involved and have corrected the description. It is basically the same.

If it is too sensitive, make the pickup probe/antenna/wire shorter.
 

Thread Starter

AndyD-UK

Joined Nov 1, 2019
24
I had misidentified the type of transistor involved and have corrected the description. It is basically the same.

If it is too sensitive, make the pickup probe/antenna/wire shorter.
I will try that thank you.

Still trying to figure this out and why when there are 2 next to each other, can one be on and the other off.

The joys of learning :)
 

AlbertHall

Joined Jun 4, 2014
12,345
You could connect a very high value resistor between the gate and source, for instance a short length of string. This would leave the LED lit all the time until some negative charge comes along.
 

Thread Starter

AndyD-UK

Joined Nov 1, 2019
24
You could connect a very high value resistor between the gate and source, for instance a short length of string. This would leave the LED lit all the time until some negative charge comes along.
I didn't know this - I will give it a go thank you :D
 

Thread Starter

AndyD-UK

Joined Nov 1, 2019
24
I have another question if you don't mind?

I just did a quick test put the sensor on my mouse mat and the LED stays on, but when I touch the casing that it is hot-glued to, the LED dims, but not if I touch it with 1 finger, only if I grab it with 2?
 

nsaspook

Joined Aug 27, 2009
13,086
I have another question if you don't mind?

I just did a quick test put the sensor on my mouse mat and the LED stays on, but when I touch the casing that it is hot-glued to, the LED dims, but not if I touch it with 1 finger, only if I grab it with 2?
Read in detail the article I linked. http://amasci.com/emotor/chargdet.html#13
10. DIPOLE ANTENNA
After you use this FET device for awhile, you'll get the idea that it has just a single antenna terminal. However, like all voltmeters, it actually has two. The rest of the circuit acts as the other terminal. To demonstrate this, build a miniature version of the detector circuit onto the top of a 9V battery. If you hold the battery as usual, the Gate does act as the antenna, and negative objects make the LED go dark. Now carefully grasp the Gate wire between fingers and lift the whole device into the air. Avoid touching the battery. If you now hold a negatively electrified object near the battery, the LED will get brighter instead of dimmer. Polarity of operation has been reversed. If you lay the whole unit down upon an insulating surface and approach it with electrified objects, you'll find that the FET gate wire responds with one polarity, while the battery and the rest of the circuit responds with the other. Try connecting the gate wire to earth ground, then suspend the rest of circuit with an insulating handle. If you hold up objects having various polarities, you'll find that polarity of operation is opposite that of the gate wire.
...
The FET is not really turned off by negative net-charge. That is an overly simplified description. It is really turned off by a DIFFERENCE in the net-charge between the silicon and of the metal plate. You can either electrify the metal plate negatively, or electrify the silicon positively (which also electrifies the battery, LED, and circuit wires.) Both will turn the FET off by pushing (or pulling) the electrons out of the silicon. Think of the rubber hose again: either you can squeeze it shut with fingers, or you can lower the pressure of the whole water circuit, and the hose will be collapsed by "suction" (by air pressure, actually.)
 

Ya’akov

Joined Jan 27, 2019
9,071
When Bill and I worked at the Boston Museum of Science, we built a panel of these FET detectors to show the fields as people would wave their hands over it.

I remember when he discovered how simple the detectors were, he played with them quite a bit.

I always like to see his (awesomely classic) website cited.
 

cmartinez

Joined Jan 17, 2007
8,220
Thank you both very much - so it sounds like this is working as intended and is just incredibly sensitive - perhaps too sensitive if the above article is anything to go by?
From said article:

Obtain a small capacitor with a value below 100 picofarads. Any value will do. Connect it between the FET's gate lead and one other FET lead (doesn't matter which one.) This greatly reduces the sensitivity of the device. In situations where the sensor is TOO sensitive, this can make a big difference. Capacitors larger than 100pF can be used, but they REALLY wipe out the sensitivity.
 

djsfantasi

Joined Apr 11, 2010
9,156
When Bill and I worked at the Boston Museum of Science, we built a panel of these FET detectors to show the fields as people would wave their hands over it.

I remember when he discovered how simple the detectors were, he played with them quite a bit.

I always like to see his (awesomely classic) website cited.
i remember that exhibit! My kids loved playing with it.
 

MisterBill2

Joined Jan 23, 2018
18,176
I am guessing that the reason for wanting to detect electrostatic fields somehow relates to things presumed to be ghosts. That is a part that I don't understand very well, which is how such can vary an electric field. Do you have an explanation for that?
 

cmartinez

Joined Jan 17, 2007
8,220
Bill, I'm sure everyone here is having the same thoughts. But Andy made a valid question about electronics, and this is an excellent opportunity for him to start learning about the field... I suggest we hold our judgement on this one ...
 

MisterBill2

Joined Jan 23, 2018
18,176
Really, I was not even considering any judgement other than what sort of source impedance the circuit would be dealing with, and what sort of voltages. Those two variables have an effect on the circuit being used to make any measurements. My one judgement is that while an LED is a simple and rugged indicator it is a challenge to quantify the magnitude of signal being sensed using only an LED. brightness as the readout.
Many years ago I did create an E-Field detection circuit that was used as a passive circuit presence detector. That device used a type 2050 thyratron tube with the grids biased through a high value resistor to a point just beyond cutoff. The sense antenna was connected directly to the grids, and when a person approached it the ambient AC field would be increased enough for the tube to go into conduction, operating a relay and switching on a light. 20 years later Electronic Design magazine published a similar circuit using the new Field effect transistors.
 
Top