Lambda diode applications #1

B

bitrex

Guest
Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

<https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0>

Version 4
SHEET 1 1524 1128
WIRE 128 -432 -800 -432
WIRE 976 -432 208 -432
WIRE -800 -304 -800 -432
WIRE -432 -256 -608 -256
WIRE -224 -256 -352 -256
WIRE 272 -256 -224 -256
WIRE 576 -256 272 -256
WIRE 688 -256 576 -256
WIRE 272 -160 272 -256
WIRE 272 -160 160 -160
WIRE 400 -160 272 -160
WIRE -800 -144 -800 -224
WIRE 160 -112 160 -160
WIRE 400 -64 400 -160
WIRE 112 -48 64 -48
WIRE -608 -16 -608 -256
WIRE 160 96 160 64
WIRE 480 96 160 96
WIRE -224 128 -224 -256
WIRE 160 144 160 96
WIRE -912 176 -912 128
WIRE -800 176 -800 128
WIRE 400 192 400 16
WIRE 400 192 224 192
WIRE -608 208 -608 64
WIRE 400 272 400 192
WIRE -912 304 -912 256
WIRE -800 304 -800 256
WIRE 976 304 976 -432
WIRE 1296 304 976 304
WIRE 1360 304 1296 304
WIRE 1296 400 1296 304
WIRE 64 416 64 -48
WIRE 160 416 160 240
WIRE 160 416 64 416
WIRE 240 416 160 416
WIRE 400 416 400 352
WIRE 400 416 240 416
WIRE 976 416 976 304
WIRE 976 576 976 480
WIRE 1104 576 976 576
WIRE 1296 592 1296 480
WIRE 1296 592 1168 592
WIRE 1104 608 1056 608
WIRE -224 640 -224 192
WIRE 48 640 -224 640
WIRE 240 656 240 416
WIRE 240 656 112 656
WIRE 48 672 -48 672
WIRE 1056 688 1056 608
WIRE 1296 688 1296 592
WIRE 1296 688 1056 688
WIRE -224 768 -224 640
WIRE -48 784 -48 672
WIRE 480 784 480 96
WIRE 480 784 -48 784
WIRE 976 784 976 576
WIRE 32 864 -48 864
WIRE 176 864 96 864
WIRE 176 976 176 864
WIRE 176 976 96 976
WIRE 256 976 176 976
WIRE 576 976 576 -256
WIRE 576 976 336 976
WIRE -224 992 -224 848
WIRE -48 992 -48 864
WIRE -48 992 -224 992
WIRE 32 992 -48 992
WIRE 176 1008 96 1008
WIRE 976 1056 976 864
WIRE 176 1088 176 1008
FLAG -912 128 Vcc
FLAG -912 304 0
FLAG -800 128 Vee
FLAG -800 304 0
FLAG 976 1056 0
FLAG 1136 624 Vcc
FLAG 1136 560 Vee
FLAG -608 208 0
FLAG 80 688 Vcc
FLAG 80 624 Vee
FLAG 688 -256 HiPass1
IOPIN 688 -256 Out
FLAG 1360 304 HiPass2
IOPIN 1360 304 Out
FLAG 64 1024 Vee
FLAG 64 960 Vcc
FLAG 176 1088 0
FLAG -800 -144 0
SYMBOL njf 112 -112 R0
SYMATTR InstName J1
SYMATTR Value 2N4393
SYMBOL pnp 224 240 R180
SYMATTR InstName Q1
SYMATTR Value 2N3906
SYMBOL voltage -912 160 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value 12
SYMBOL voltage -800 160 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V2
SYMATTR Value -12
SYMBOL cap 960 416 R0
SYMATTR InstName C1
SYMATTR Value 47n
SYMBOL res 1280 384 R0
SYMATTR InstName R3
SYMATTR Value 100
SYMBOL res 960 768 R0
SYMATTR InstName R4
SYMATTR Value 10k
SYMBOL voltage -608 -32 R0
WINDOW 123 24 124 Left 2
WINDOW 39 0 0 Left 2
SYMATTR Value2 AC 1
SYMATTR InstName V3
SYMATTR Value SINE(6 1 1k)
SYMBOL cap -240 128 R0
SYMATTR InstName C2
SYMATTR Value 47n
SYMBOL res -240 752 R0
SYMATTR InstName R7
SYMATTR Value 10k
SYMBOL res -336 -272 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R8
SYMATTR Value 10k
SYMBOL res 224 -448 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R9
SYMATTR Value 10k
SYMBOL res 384 -80 R0
SYMATTR InstName R2
SYMATTR Value 10k
SYMBOL res 384 256 R0
SYMATTR InstName R1
SYMATTR Value 2.2k
SYMBOL Opamps\\LT1014 64 928 M0
SYMATTR InstName U3
SYMBOL res 352 960 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 45 58 VTop 2
SYMATTR InstName R5
SYMATTR Value 1Meg
SYMBOL cap 96 848 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C3
SYMATTR Value 100n
SYMBOL voltage -800 -320 R0
WINDOW 123 24 124 Left 2
WINDOW 39 0 0 Left 2
SYMATTR Value2 AC 1
SYMATTR InstName V4
SYMATTR Value SINE(0 0.5 1k)
SYMBOL Opamps\\LT1055 80 720 M180
SYMATTR InstName U1
SYMBOL Opamps\\LT1055 1136 656 M180
SYMATTR InstName U2
SYMBOL res 144 -32 R0
SYMATTR InstName R6
SYMATTR Value 220
TEXT -944 432 Left 2 !;tran 100m
TEXT -944 392 Left 2 !.ac oct 10 100 1000k
TEXT -1040 16 Left 3 ;SIGNAL + DC BIAS ~6V
TEXT -240 1104 Left 3 ;DC REMOVAL SERVO
 
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
<user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0
Click to expand...

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.
 
On 7/8/19 11:03 AM, John Larkin wrote:
On Mon, 08 Jul 2019 08:44:07 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.

The circuit is bizarre and wrong in more ways than I have time to
enumerate.
Click to expand...

a Convincing analysis...
 
On 7/8/19 10:55 AM, bitrex wrote:
On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q
and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower
R and thus increasing frequency and bandwidth at that.


if it were also behaving as just a strictly a linear resistor equivalent
to R3 on the right, but a smaller value, the phase shift above ~1k would
also be monotonic decreasing
Click to expand...

or rather, the phases are both monotonic decreasing but there couldn't
be an inflection point
 
On Mon, 08 Jul 2019 08:44:07 GMT, Jan Panteltje
<pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.
Click to expand...

The circuit is bizarre and wrong in more ways than I have time to
enumerate.


--

John Larkin Highland Technology, Inc trk

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com
 
On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.
Click to expand...

if it were also behaving as just a strictly a linear resistor equivalent
to R3 on the right, but a smaller value, the phase shift above ~1k would
also be monotonic decreasing, I don't believe it could break down and
then break back up again in this RL-highpass equivalent configuration as
it does.
 
On a sunny day (Mon, 8 Jul 2019 10:55:22 -0400) it happened bitrex
<user@example.net> wrote in <vPIUE.133012$xm4.78259@fx45.iad>:

On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.


if it were also behaving as just a strictly a linear resistor equivalent
to R3 on the right, but a smaller value, the phase shift above ~1k would
also be monotonic decreasing, I don't believe it could break down and
then break back up again in this RL-highpass equivalent configuration as
it does.
Click to expand...

I am not questioning the circuit does what it does (according to spice).
See what I wrote.
But you also use a signal from the PNP emitter, so not really a 2 pole lambda thing.

If what it does is what you wanted it to do, then that is great.
But I think it does not make a good example of using a lambda diode.
 
On 7/8/19 12:59 PM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 10:55:22 -0400) it happened bitrex
user@example.net> wrote in <vPIUE.133012$xm4.78259@fx45.iad>:

On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.


if it were also behaving as just a strictly a linear resistor equivalent
to R3 on the right, but a smaller value, the phase shift above ~1k would
also be monotonic decreasing, I don't believe it could break down and
then break back up again in this RL-highpass equivalent configuration as
it does.

I am not questioning the circuit does what it does (according to spice).
See what I wrote.
But you also use a signal from the PNP emitter, so not really a 2 pole lambda thing.

If what it does is what you wanted it to do, then that is great.
But I think it does not make a good example of using a lambda diode.
Click to expand...

If I remove the signal entirely and just apply a bias voltage for V3.
Ramp the bias voltage between about 2 and 8 volts. The current that V3
sources increases linearly. The current that op amp U1 sinks also
increases, and is approximately the same current that V3 is sourcing,
since the output is servoed to 0 volts by U3 and its input draws no
current from the output.

So the bias current increases from 0.2 mA to almost 1 mA as the DC bias
voltage is increased. That current can only pass thru the lambda diode
structure. meanwhile the voltage drop across the structure stays
approximately constant thru that range.

If the current forced thru it is increasing but the drop across it is
staying approximately constant then its equivalent impedance must be
decreasing with increasing bias current.

basically IMO it's a valid use for the structure to fiddle the bias such
that it's operating not over into the negative resistance region but
around the peak of the curve, so it's acting like a current-controlled
resistance. at high freq the impedance of the equivalent of R3 on the
right gets boosted and at low it gets reduced. but the impedance there
also affects the nominal inductance value of the virtual "inductance"
too, unfortunately, which explains the sloppy phase shift trace, looks
like it's "robbing" phase from the left side to give to the right.
 
On 7/8/19 12:59 PM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 10:55:22 -0400) it happened bitrex
user@example.net> wrote in <vPIUE.133012$xm4.78259@fx45.iad>:

On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.


if it were also behaving as just a strictly a linear resistor equivalent
to R3 on the right, but a smaller value, the phase shift above ~1k would
also be monotonic decreasing, I don't believe it could break down and
then break back up again in this RL-highpass equivalent configuration as
it does.

I am not questioning the circuit does what it does (according to spice).
See what I wrote.
But you also use a signal from the PNP emitter, so not really a 2 pole lambda thing.

If what it does is what you wanted it to do, then that is great.
But I think it does not make a good example of using a lambda diode.
Click to expand...

Basically, try experimenting with R3 in the standard gyrator on the
right, which sort of the defines the "ESR" of your equivalent inductance
in the RL high pass.

Increase the value to 1k and see how the response on the high end
improves but you get fucked on low end, lower it to 50 and vice-versa.
circuit on left is a compromise
 
On 7/8/19 12:59 PM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 10:55:22 -0400) it happened bitrex
user@example.net> wrote in <vPIUE.133012$xm4.78259@fx45.iad>:

On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.


if it were also behaving as just a strictly a linear resistor equivalent
to R3 on the right, but a smaller value, the phase shift above ~1k would
also be monotonic decreasing, I don't believe it could break down and
then break back up again in this RL-highpass equivalent configuration as
it does.

I am not questioning the circuit does what it does (according to spice).
See what I wrote.
But you also use a signal from the PNP emitter, so not really a 2 pole lambda thing.

If what it does is what you wanted it to do, then that is great.
But I think it does not make a good example of using a lambda diode.
Click to expand...

No, you're correct somethings not right, it's not biased correctly to do
what I claim - it's acting too much like a linear resistor as it's
currently biased.
 
On 7/8/19 4:44 AM, Jan Panteltje wrote:
On a sunny day (Mon, 8 Jul 2019 03:38:33 -0400) it happened bitrex
user@example.net> wrote in <_pCUE.141123$qW.57208@fx43.iad>:

Use to boost the bandwidth of a gyrator, used here
in a rob-Peter-to-pay-Paul high-pass filter configuration

https://www.dropbox.com/s/lblj2gdxuu4fo77/LambdaFilter.png?dl=0

I think this is not right,
if you really used he negative resistance aspect it would increase Q and decrease bandwidth.
I think you are merely damping the circuit, causing it so see a lower R and thus increasing frequency and bandwidth at that.
Click to expand...

Scratch my followups, it wasn't operating the way I thought it was. The
low-frequency response should be approximately the same. The idea is to
use the cusp of the lambda diode curve where the equivalent resistance
has a horizontal tangent to boost the impedance looking into the gyrator
at high frequency and extend the HF response, something like this:

<https://imgur.com/a/20UHS2E>

I haven't found the best operating point yet, though.
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top