Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Linearly movable contact point

Salvador12

Full Member level 4
Full Member level 4
Joined
Jul 16, 2014
Messages
224
Helped
3
Reputation
6
Reaction score
7
Trophy points
1,298
Activity points
4,161
Does there exist any device or technology where something like a FET channel exists that is an E field controllable switch but unlike a FET, the device can be made long (like a small width long length rectangle , ideally flexible to some degree) and the conducting part can be swept linearly along the way.
Like switching the E field in sequence modulates where along the length of the device the channel exists, so that a current path can be controlled.

You see unlike in a regular switch I do not intend to switch it off completely, it can stay ON all the time , just the current path needs to be moved around spatially in a linear way.
I cannot do this with regular discrete MOSFET's because they are discrete and even though making a long line of them that connect two conducting surfaces still as I switch them in sequence the current path moves in steps not in a sweeping action.
Ideally if there was one long FET channel then such a thing could be done.

Is there anything close to what i ask, and out of curiosity can semiconductors like FET's be made long and large , instead of having small dies within epoxy molded packages?
 
A sufficiently "granular" lineup of discrete FETs might
be made "close enough" for experimental purposes.

How long and how large is a key question. You can
go far "oversquare", but how far is far enough?
100:1? 1000:1? At some point size and parasitic
effects may start to obscure whatever it is you're
trying to display.

There are conductive inks that you could put on
paper and move a contact point, and some
higher-resistivity sheet-form materials might be
interesting to play with (stainless steel pretty thin,
should have high ohms/sq (for a metal) though
contact resistance might need some care; some
"somewhat conductive" polymers perhaps?).

This is not E-field controllable, but you could get
a handle on the geometric effects that seem to be
your interest.

Problem with E-field control is that you want its
intensity of effect, but you do not want its field
induced reliability / drift effects. Stronger conduction
will be had by thinner dielectrics which then limit
applied voltage, and so on.

What about optical stimuli, photoconduction is
a viable modulation (look at CdS photocells and
you will see fine serpentine detail aimed at getting
your high L in small area - but those are for small
signal, more current means more size of course.
 
A sufficiently "granular" lineup of discrete FETs might
be made "close enough" for experimental purposes.
You know I'm thinking along the lines of this, but maybe there are some other ways to help achieve a smooth current sweep along a conductive surface by using FET's because in order to make this just by the FET's themselves one would need a very very close spacing of these FET's due to their abrupt ON/OFF switching which causes a stepwise motion rather than a sweep. I wonder are there any ways by which the switching ON of each next FET could be made such that the current gradually moves towards the newly switched FET instead of it making a direct new current path right away, making a stepwise current "walking"
 
This sounds like it could be done by depositing layers of doped substances on silicon wafers, then leaving it accessible for touching wires to it. Making your own transistor/fet is within reach, according to Jeri Ellsworth. She stands out as an electronics professional who experiments with the technology. Her techniques are discussed in many places:

hackaday.com/2010/05/13/transistor-fabrication-so-simple-a-child-can-do-it/

www.electronicproducts.com/diy-transistor-fabrication-are-you-up-to-the-challenge/
 
This sounds like it could be done by depositing layers of doped substances on silicon wafers, then leaving it accessible for touching wires to it. Making your own transistor/fet is within reach, according to Jeri Ellsworth. She stands out as an electronics professional who experiments with the technology. Her techniques are discussed in many places:

hackaday.com/2010/05/13/transistor-fabrication-so-simple-a-child-can-do-it/

www.electronicproducts.com/diy-transistor-fabrication-are-you-up-to-the-challenge/
I will read those papers, but ideally what I would want is a device that is like a rectangle (the shape is not most important just for clarity) and that one would be able to control the current path across it. For this to be done perfectly one would need like a linear strip along the length of the rectangle and this strip ideally should have the ability to , for example, conduct only in places where an E field is applied from above. Same idea as a Mosfet, but instead of a discrete device just a macroscopic strip. Now I am not a semiconductors expert, but I doubt something like this can be practically made. I also guess silicon is not that durable to be made into long macroscopic structures.

The closest analogue to this would be just using a lot of small low voltage high current FET's of which there now are many kinds, and just put them close to one another (one after the other) and then switch them in sequence. But I am not sure how smooth would the transition be.

Maybe there are methods that can be used at or close to the FET's connection points that would smooth out the moving ON/OFF switching FET addition to the total current? To make the current move as one current not many stepwise switched currents. Maybe a surface charge on the plate where the FET's are connected would smooth out the current spatial position change? Another , albeit , less favorable option would be to use a series inductor for each FET which would make the current rise slower, similarly to a BJT transistor
 
Distributed E fields are easy, but what is your load?
I don't have a load, this is more like a theoretical idea, I am just thinking how easy it is to control a current path within a flat thin and wide conductor.
Normally within a regular shaped conductor DC flows within the whole of it and AC flows within the outside shell (skin depth determined by frequency), here there is a flat long conductor and it is connected to a circuit so is part of it , but although being part of a circuit, current tends to flow in the shortest path with least resistance, so by sweeping the current path , theoretically one should be able to control the path where it flows.
 
Some of the amorphous or organic thin film
technologies could be low cost ways to get
samples (classical wafer processing costs, per
step, and you need (IMO) only two steps -
semiconductor film and passivation - and
two masks (doping and passivation-cut).
Silkscreen (coarser) technologies are even
cheaper.

With passivation in place, that's your gate
dielectric and you could overlay (stamped,
silkscreened, whatever) a conductive electrode
to make yourself any shaped MOSFET you like.
There's questions of uniformity, work function
and doping that lets you observe an effect at
voltages which don't do damage, but that's
all part of the adventure, right? And if your
access cuts are fixed, you'd have some non-
modulated "access resistance" to de-embed
(think about Kelvin wiring, 4 contact cuts
rather than 2, etc.). And think about how to
get as many different configs on one "die"
as you can, which is the core semiconductor
leverage.
 
ok I thought you meant static or different DC e-fields with a motion control. With AC, the goal is usually to preserve the signal or match the current phase to the voltage or maximize power transfer with a conjugate load. So why not control the impedance by controlling the L/C ratio by log{length/width) ratios and width/gap ratios per unit length. This is one way to control E-field and H-field with choice of dielectric and conductor shape.

Is your goal to widen bandwidth of understanding how to control Zo and eddy current losses or how to make a very wide bandwidth megawatt busbar for DC with AC step pulse loads. If it is the latter then we are getting closer to your requirements.
 

LaTeX Commands Quick-Menu:

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top