Electrical friction?

On 7/13/2013 2:11 PM, Phil Hobbs wrote:

And here's one with a BF862 as the current source, with and without a floating bias voltage.

Cheers

Phil Hobbs

Version 4 SHEET 1 880 680 WIRE -224 32 -432 32 WIRE -80 32 -224 32 WIRE 208 32 -80 32 WIRE 80 96 32 96 WIRE 208 96 208 32 WIRE 208 96 144 96 WIRE 256 96 208 96 WIRE 384 96 320 96 WIRE 176 128 96 128 WIRE 176 144 176 128 WIRE -432 160 -432 32 WIRE -224 160 -224 32 WIRE -80 160 -80 32 WIRE 32 192 32 96 WIRE 96 192 96 128 WIRE 96 192 32 192 WIRE 144 192 96 192 WIRE 256 192 240 192 WIRE 384 192 384 96 WIRE 384 192 336 192 WIRE 32 320 32 192 WIRE 80 320 32 320 WIRE 208 320 144 320 WIRE 256 320 208 320 WIRE 384 320 384 192 WIRE 384 320 320 320 WIRE -432 368 -432 240 WIRE -224 368 -224 224 WIRE -224 368 -432 368 WIRE -80 368 -80 240 WIRE -80 368 -224 368 WIRE 208 368 208 320 WIRE 208 368 -80 368 WIRE -80 400 -80 368 FLAG -80 400 0 SYMBOL diode 256 112 R270 WINDOW 0 44 17 VTop 2 WINDOW 3 71 74 VBottom 2 SYMATTR InstName D2 SYMATTR Value {diode} SYMBOL ind -64 256 R180 WINDOW 0 49 82 Left 2 WINDOW 3 47 41 Left 2 SYMATTR InstName L1 SYMATTR Value 1m SYMBOL cap -240 160 R0 WINDOW 0 -54 14 Left 2 WINDOW 3 -57 54 Left 2 SYMATTR InstName C1 SYMATTR Value 1µ SYMBOL current -432 240 R180 WINDOW 0 21 -17 Left 2 WINDOW 3 -174 -162 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName I2 SYMATTR Value PULSE(0 1 0 0 0 10u) SYMBOL diode 256 336 R270 WINDOW 0 42 10 VTop 2 WINDOW 3 70 72 VBottom 2 SYMATTR InstName D1 SYMATTR Value {diode} SYMBOL diode 80 336 R270 WINDOW 0 44 17 VTop 2 WINDOW 3 74 79 VBottom 2 SYMATTR InstName D3 SYMATTR Value {diode} SYMBOL diode 80 112 R270 WINDOW 0 44 17 VTop 2 WINDOW 3 70 83 VBottom 2 SYMATTR InstName D4 SYMATTR Value {diode} SYMBOL voltage 240 192 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 -32 56 VBottom 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value {Vbias} SYMBOL njf 240 144 R90 WINDOW 0 20 92 VRight 2 WINDOW 3 74 70 VRight 2 SYMATTR InstName J1 SYMATTR Value BF862J TEXT 88 400 Left 2 !.model Di D(Vfwd=0) TEXT -216 -32 Left 2 !.tran 10m TEXT 112 -32 Left 2 ;Frictional equivalent resistor, hacked a bit more TEXT 240 0 Left 2 ;JL/PH July 12, 2013 TEXT 432 112 Left 2 !.model 1 AKO: BAT54\n.model 2 AKO: Di TEXT 408 192 Left 2 !.step param diode list 1 2\n.step param Vbias list

That's totally great! I might have to bog one together. It'd make a cool 'scope shot if nothing else.

George H.

=.8

Hi, Phil,

For some reason I can't see your post. That happens some time.

Seems to me that, even in my ideal version, there is some weird distortion. Makes sense, the sine wave is being bent all the time.

You are making word salad again.

In sliding friction, force is independent of velocity.

In a viscous coupling (sub-turbulent) force is proportional to velocity.

They are different. The equations are different.

I'm not qualified to say if viscosity "comes about_because_ of friction." That's all words again.

Resistance steals power from a tank as V^2. The fristor steals power as V. Some other gadget could steal constant power, and make a different envelope.

There is a generalization of the fluctuation-dissipation theorem due to Stratonovich. It's probably applicable.

An interesting concept, this 'fristor'.

Regards, Mikko

Ever hear of hot electrons?

In some meanings, I have. Like hot-electron bolometers, or hot-electron effects in SQUID shunt resistors. Any electron which is far from thermal equilibrium qualifies, I suppose.

But I'm puzzled about why do you ask?

Regards, Mikko

The topic was "electrical friction"...

What's the equation for this fristor? And what kind of friction is it analogous to?

Best regards, Spehro Pefhany

Forget it.. this conversation was from July.. it's dead, Jim.

--sp

I = sgn(V) / F

where F is the fristance.

It's analogous to sliding friction, constant force at any nonzero velocity. Sort of like a felt-pad damper.

Yup, It's the linear decay of the envelope that says 'friction' to me.

As far as the noise, I'd guess that the diodes add full shot noise of whatever current they are biased to. Or will it be more than that?

George H.

Well, it obviously relies on some relative of Maxwell's Demon. Maxwell's famous thought experiment was that if you had two sealed chambers connected by a small orifice with a very small valve, and you had a demon controlling the valve who let all the gas molecules through one way but not the other, eventually all the gas would wind up in one of the two chambers.

Various people over the years have shown that even if you had a control system that took zero power, e.g. an atomic-scale check valve, the thermal excitation of the check valve itself would be enough to prevent this from working.

The same argument would seem to apply to the fristor, if it were possible to have it work in thermal equilibrium (which it probably isn't, since it needs infinitely variable conductance).

Cheers

Phil Hobbs