Wien Bridge osc.

The usual tradeoff is distortion vs. reliable startup over temperature and component variations.

Nice going.

Cheers

Phil Hobbs

Diodes Inc. makes some matched parts and some dual 2n3904-type parts that are pretty reasonable. I got 'em from DigiKey, but can't find them on there at the moment.

Here's another:

Pretty. Nice work.

-- Cheers, James Arthur

Those packaged "matched" pairs really rot my socks. If the silly manufacturer would just dice them in pairs instead of singly, they'd have 100 times better thermal tracking and might actually be useful for something at dissipations over 100 uW.

Razza frazza &%$#@!!

Cheers

Phil Hobbs

ND/2676828

I know. Dice. Mix well. Figure out how to mount two up in one package...

The Diodes parts are off the same wafer (adjacent die, I believe) but without the tight thermal coupling they aren't worth much. I've used them successfully but it wasn't a very tight requirement. Zetex used to have some monolithic pairs but they got rid of them when Diodes took over.

Diode clipping is a bit crude. If you want good Wien bridges, look at Jim Williams AN-43

I'm fooling around with something based on an AD734, but the Analog Devices Spice model seems to blow up LTSpice rather too frequently - the circuit can be sitting there oscillating at a stable level when all of a sudden the simulation will start generating megavolts ..

-- Bill Sloman, Nijmegen

re

Yep. Taking beautifully matched transistors and sawing them apart is criminal.

A cheap source of matched diodes for George though, cheaper than self- testing.

-- Cheers, James Arthur

On 2/15/2012 8:11 PM, George Herold wrote:

Since you're using active devices anyway, I wonder how the following circuit would perform with "matched" complimentary MOSFETs compared to matched bipolars - complimentary switching MOSFETs with similar characteristics are available in a single package for about 16 cents in quantity:

Here's an image of the circuit:

LTSPice netlist:

Version 4 SHEET 1 1192 956 WIRE 96 -416 64 -416 WIRE 192 -416 176 -416 WIRE 192 -368 192 -416 WIRE 288 -320 272 -320 WIRE -176 -224 -176 -240 WIRE -64 -224 -64 -240 WIRE 64 -176 64 -416 WIRE 80 -176 64 -176 WIRE 192 -176 160 -176 WIRE 192 -144 192 -176 WIRE 176 -96 176 -320 WIRE 288 -96 288 -320 WIRE 288 -96 272 -96 WIRE 64 32 64 -176 WIRE 80 32 64 32 WIRE 176 32 176 -96 WIRE 176 32 160 32 WIRE 192 32 176 32 WIRE 288 32 288 -96 WIRE 288 32 272 32 WIRE 144 128 144 112 WIRE -80 144 -112 144 WIRE 64 144 64 32 WIRE 64 144 0 144 WIRE 112 144 64 144 WIRE 288 160 288 32 WIRE 288 160 176 160 WIRE 416 160 288 160 WIRE 112 176 0 176 WIRE -112 192 -112 144 WIRE 144 208 144 192 WIRE 416 208 416 160 WIRE 0 304 0 176 WIRE 0 304 -64 304 WIRE 96 304 0 304 WIRE 192 304 160 304 WIRE 288 304 288 160 WIRE 288 304 272 304 WIRE 416 320 416 288 WIRE -64 384 -64 368 WIRE -32 384 -64 384 WIRE 0 384 -32 384 WIRE -32 416 -32 384 FLAG -176 -144 0 FLAG -176 -240 Vcc FLAG -64 -144 0 FLAG -64 -240 Vee FLAG 144 112 Vcc FLAG 144 208 Vee FLAG -112 192 0 FLAG -32 416 0 FLAG 416 320 0 SYMBOL Opamps\\LT1014 144 96 R0 WINDOW 3 20 -1 Left 2 SYMATTR InstName U1 SYMBOL voltage -176 -240 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 12 SYMBOL voltage -64 -240 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value -12 SYMBOL res 16 128 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 10k SYMBOL res 176 16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value 20k SYMBOL res 288 16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 220 SYMBOL res 288 288 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R4 SYMATTR Value 10k SYMBOL cap 160 288 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C1 SYMATTR Value 16n SYMBOL cap -80 304 R0 SYMATTR InstName C2 SYMATTR Value 16n SYMBOL res -16 288 R0 SYMATTR InstName R5 SYMATTR Value 10k SYMBOL res 400 192 R0 SYMATTR InstName R10 SYMATTR Value 100k SYMBOL nmos 272 -368 R90 SYMATTR InstName M1 SYMATTR Value 2N7002 SYMBOL pmos 272 -144 R90 SYMATTR InstName M2 SYMATTR Value BSS84 SYMBOL res 192 -432 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R6 SYMATTR Value 10k SYMBOL res 176 -192 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R7 SYMATTR Value 10k TEXT -196 -80 Left 2 !.tran 5 uic

Wow, you've just figured out how to generate energy from nothing. Sell it!

NT :)

re

The universe where the simulated circuits develop megavolts doesn't seem to have grid connections to any place that will offer me money for the power ostensibly being generated.

-- Bill Sloman, Nijmegen

That should work, pity that they are cut up.

%20Trick.pdf

The Fourier Methods project has been a bit of a rush job. The pcb's came in yesterday. I've been stuffing and testing all day. (A bit of a snafu delayed them a week.. with no weeks to spare!)

So far so good, though I've gotta add some 'tweaking' caps here and there.

George H.

I had this 'late night beer' idea. That I could run the diodes hot and use the Vf temp. dependence to help roll off the gain... (the opposite of the light bulb.) But a few millivolts is not that much... not a big dR/dV.

Does it make sense any sense to run the diodes warm?

George H.

t -

AN-43 was the first thing I read by Jim! After reading I decided I *had* to use a light bulb stablized Wien bridge, in our lockin. The distortion is right at the 90dB noise floor of the srs7700. (except at the lower frequencies.)

The diodes are not too bad. 60dB is below what I can see on my 'scope. I 're-invented'* the trick of just clipping part of the gain. You get large amplitudes this way too. ~5-6Vp-p

George H.

That's a cool idea, but how do you get the FET's to turn on at the 'same' voltage.

George H.

I'm much more fond of linear gain adjustments, e.g. building an oscillator with a diff pair in it, and using a gentle amplitude detector to set DeltaV_BE. You could probably make a well-behaved Wien bridge based on a MC1496.

Cheers

Phil Hobbs

quantity:

In a complimentary pair IC like the FG694301, the P and N channel threshold voltages are matched. Even if they're not exactly the same there shouldn't be any crossover distortion, because the resistor bridging the MOSFETs handles the signal during the dead time when it's near 0 volts.

quantity:

Ahem, look at the datasheet VGSth figures then look at the ID(Vgs) curves for both fets.

Where's the bug?

quantity:

I dunno. The Vds of the MOSFETs in the circuit is never more than a few millivolts and the current is never more than a few hundred microamps, and I certainly can't read the figures down there! They're basically being used as voltage controlled resistors. I really don't know how well the triode mode resistance two complimentary MOSFETs will match down there, but just using the random junk small-signal MOSFETs available in the LTSpice library looked to give about the same 2nd harmonic distortion figure as George was getting with two unmatched diodes, so I thought the comparison could be interesting.

A variation of this...

http:/

is what I prefer instead of a Wien Bridge. ...Jim Thompson