reverse-biased diode noise source?
Walter Harley
functions as a wideband noise source. However, I can't find much
information about how to put this into practice. How much noise does it
generate, for how much current? Does it have to be in avalanche, or
does it generate noise even at leakage-current levels? Is there any
difference between different diodes, e.g., different voltages? What is
the frequency spectrum like?
Thanks,
-walter
Leon Heller
"Walter Harley" <wal...@cafewalter.com> wrote in message
news:am46nc$ojh$1...@216.39.172.65...
I recently built a little noise source using this property - a zener diode
followed by a broadband amplifier. Output is reasonably flat up to 30 MHz,
according to the radio I checked it on.
Since this isn't really a desirable property of the devices, you basically
have to 'suck it and see' if you want to use it. A co. called NOISE/COM
supplies noise diodes rated from 10 Hz to 3 GHz if you want one with a
proper spec. They will calibrate units built with their diodes (for a
price).
Leon
--
Leon Heller, G1HSM
leon_...@hotmail.com
http://www.geocities.com/leon_heller
Tom Faloon
Any silicon or germanium rectifier diode, transistor base emitter junction,
or zener diode will act as a noise generator when reverse biased to
breakdown (avalanche) region.
However, this is generally an unwanted effect, and manufacturers do not
specify characteristics.
It is possible to buy diodes with published characteristics, but these are
specialist devices,
and cost and minimum order quantity put them in reach of business users
only.
You can obtain useful output from everyday devices, but you will not be able
to predict results accurately in advance, or assume repeatability from
component to component.
In the simplest application, simply reverse connect the diode to a supply,
via a current limiting resistor, increase the supply voltage until reverse
current flows,
and a noise voltage will then be produced across the diode.
You can expect a fairly 'white but not perfect' noise spectrum over a wide
frequency range.
Non perfection is most likely to be due to secondary noise functions in the
diode, the
most pronounced being flicker noise. (1/f) which will be present below about
100 kHz.
The bandwidth will depend on the components and layout used. (Junction
capacitance, lead inductance, PCB track capacitance & inductance.)
Bandwidths into the GHz region are possible.
The noise level generally increases with reverse current. Level will depend
on the device used
As a rough figure think fractions of a nanoVolts / root Hz. (Perhaps X 10 to
100)
Useful devices.
A standard rectifier diode. (Not always practical, because most diodes have
a fairly high breakdown voltages.)
Base emitter junction of a bipolar transistor is more practical. (Reverse
breakdown is often around 5V)
(Leave the collector unconnected.
Zener diode, but use devices above about 6 or 7 Volts.
(We call all these devices Zener diodes, but in fact they operate in two
different modes. Devices operating below about 5 Volts rely on the Zener
effect. Higher
voltage devices use the avalanche effect.)
Tom Faloon
remove 'z' from address to email direct.
http://www.faloon.co.uk
**********************************************
Walter Harley <wal...@cafewalter.com> wrote in message
news:am46nc$ojh$1...@216.39.172.65...
Phil Hobbs
low frequency noise source by shining the light from a LED on a
photodiode--_NOT_ a phototransistor or photodarlington. Drive the LED
from a quiet current source (e.g. a battery and metal film resistor or a
well-designed current source with a big voltage drop across its (metal)
sense resistor). Provided the incident light is really unmodulated, the
photocurrent exhibits exactly full shot noise. Shot noise is white and
Gaussian to very high accuracy, so you can calibrate the noise power
spectral density by measuring the DC and applying the shot noise
formula,
i sub N = sqrt(2*e*I sub dc) (1-Hz BW)
where e is the charge on the electron (1.602e-19 coulombs).
This method is much better than the Zener approach, because it's based
on fundamental physics rather than device parameters that have to be
calibrated from unit to unit. Since it's easy to generate a
photocurrent that has exactly full shot noise--shine a flashlight on
your photodiode--it's easy to test by comparison whether your LED drive
current source is quiet enough.
Cheers,
Phil Hobbs
John Larkin
Tom,
a 1N758 (10 volt) zener makes pretty nice gaussian noise at 1 mA; at
much lower currents the waveform starts getting asymmetric and then
oscillatory. At 1 mA, noise density is about 340 nV per root Hz, or
about 60 uV RMS in a 30 KHz bandwidth.
John
Tom Faloon
I like it!
Tom Faloon
Phil Hobbs <pc...@SpamMeSenseless.us.ibm.com> wrote in message
news:3D860C94...@SpamMeSenseless.us.ibm.com...
Tom Faloon
That gives Walter a good idea about level. He also asked about bandwidth.
Do you have any bandwidth figures for this device? I looked at the data
sheet, but no info - as expected.
Nor does it give anything which might help calculate it (e.g. diode
capacitance)
Many thanks,
Tom Faloon
remove 'z' from address to email direct.
http://www.faloon.co.uk
John Larkin <Jo...@0.com> wrote in message
news:tf3cou8hlojnh7e5f...@4ax.com...
> On Mon, 16 Sep 2002 15:17:59 +0100, "Tom Faloon"
> a 1N758 (10 volt) zener makes pretty nice gaussian noise at 1 mA; at
John Larkin
<pc...@SpamMeSenseless.us.ibm.com> wrote:
>Now that amplifiers are so quiet, it's easy to make a well-calibrated
>low frequency noise source by shining the light from a LED on a
>photodiode--_NOT_ a phototransistor or photodarlington. Drive the LED
>from a quiet current source (e.g. a battery and metal film resistor or a
>well-designed current source with a big voltage drop across its (metal)
>sense resistor). Provided the incident light is really unmodulated, the
>photocurrent exhibits exactly full shot noise. Shot noise is white and
>Gaussian to very high accuracy, so you can calibrate the noise power
>spectral density by measuring the DC and applying the shot noise
>formula,
>
>i sub N = sqrt(2*e*I sub dc) (1-Hz BW)
>
>where e is the charge on the electron (1.602e-19 coulombs).
>
>This method is much better than the Zener approach, because it's based
>on fundamental physics rather than device parameters that have to be
>calibrated from unit to unit. Since it's easy to generate a
>photocurrent that has exactly full shot noise--shine a flashlight on
>your photodiode--it's easy to test by comparison whether your LED drive
>current source is quiet enough.
>
>
>Cheers,
>
>Phil Hobbs
>
Phil,
that's neat, except that it's not a lot of signal... something like 18
pa / rootHz for a 1 mA photodiode current. A simple 50 ohm resistor
coupled into a TIA would make about that. If you run the pd's 1 mA
into a good TIA with, say, 1K transimpedance, you're still only up to
18 nV/rtHz.
I do know that 1950-vintage radar jammers used 931 photomultiplier
tubes as noise sources.
John
Phil Hobbs
John Larkin wrote:
> > Phil,
>
> that's neat, except that it's not a lot of signal... something like 18
> pa / rootHz for a 1 mA photodiode current. A simple 50 ohm resistor
> coupled into a TIA would make about that. If you run the pd's 1 mA
> into a good TIA with, say, 1K transimpedance, you're still only up to
> 18 nV/rtHz.
>
> I do know that 1950-vintage radar jammers used 931 photomultiplier
> tubes as noise sources.
>
> John
(Longish post warning)
John,
You're right that the noise current isn't too big, but the key is the
impedance level. Photodiodes are very nearly ideal current sources
(with a shunt capacitance). Thus the noise power level can be made as
large as you like by increasing the load resistance, at least until the
RC time constant becomes a problem. It's almost always possible to make
it much higher than the kT per Hz you get from a resistor.
Making an amplifier that's much quieter than the shot noise is as hard
as wiring an LF411 to a 10k resistor--you'll be shot noise limited any
time the photocurrent drops more than 2kT/e volts across the load
resistor (~50 mV at room temperature), provided the resistor noise is
greater than the amplifier noise. (This was the origin of my somewhat
cryptic comment about amplifiers being quiet nowadays.) TIAs are
typically unhappy with 50 ohm source impedances, because the noise gain
becomes so high that the amplifier voltage noise tends to dominate.
Even when it doesn't, the amplifier noise is a far higher proportion of
the total output than with a decent-sized photocurrent, so the
uncertainties and drifts of the amplifier noise will require
calibration.
Furthermore, the noise power from a resistor is proportional to the
absolute temperature, whereas photocurrent shot noise involves only the
dc current and fundamental constants, so it's easy to make it very
stable with time and conditions, unlike noise diodes and resistors.
Also unlike noise diodes, it's Gaussian to very high accuracy(*) and
shows no measureable 1/f noise.
Photomultipliers are far worse than either, because their noise PSD
depends on their gain, which changes _dramatically_ with stray magnetic
fields, history, age, ambient helium concentration, and bias voltages.
(There are PMTs whose gain changes by a factor of 2 when you turn them
from East-West to West-East in the Earth's magnetic field.) PMT noise
is also very far from Gaussian--it consists of similar-sized pulses
arriving in a Poisson process, with occasional very large blips from
ions hitting the photocathode. Low-pass filtering helps somewhat, but
that won't fix the calibration problem.
Cheers,
Phil Hobbs
(*) I've measured the Gaussian statistics of photocurrent shot noise,
and found that it's Gaussian to within 0.1 dB out to at least 7.1 sigma,
which corresponds to a threshold crossing rate of about 10**-11 times
the bandwidth. I couldn't go any further, because my last data point
gave 2 counts in 69 hours, and the threshold crossing rate goes down as
exp(-0.5(Vthresh/Vrms)**2)--going 1 dB further out would have reduce the
predicted crossing rate to 1 per year in my 1 MHz BW.
For anyone who's interested in this or other photodetection issues and
other low-noise optical measurements, I have some possibly interesting
papers at
http://www.pergamos.net.
(The exact URL for the paper containing the Gaussian shot noise plot is
http://users.bestweb.net/~hobbs/canceller/noisecan.pdf.)
Phil Hobbs
John Larkin wrote:
> > Phil,
>
> that's neat, except that it's not a lot of signal... something like 18
> pa / rootHz for a 1 mA photodiode current. A simple 50 ohm resistor
> coupled into a TIA would make about that. If you run the pd's 1 mA
> into a good TIA with, say, 1K transimpedance, you're still only up to
> 18 nV/rtHz.
>
> I do know that 1950-vintage radar jammers used 931 photomultiplier
> tubes as noise sources.
>
> John
(Longish post warning)
John,
Cheers,
Phil Hobbs
Jonathan Kirwan
<pc...@SpamMeSenseless.us.ibm.com> wrote:
>For anyone who's interested in this or other photodetection issues and
>other low-noise optical measurements, I have some possibly interesting
>papers at
>http://www.pergamos.net.
>(The exact URL for the paper containing the Gaussian shot noise plot is
>http://users.bestweb.net/~hobbs/isicl/isiclao3.pdf.)
Thanks much. It'll be some of my reading for the next few days.
Jon
John Devereux
The book is excellent too - wish I had it a few years ago,
when I really needed it!
--
John Devereux
Andre
> On Tue, 17 Sep 2002 18:18:23 -0400, Phil Hobbs
> <pc...@SpamMeSenseless.us.ibm.com> wrote:
>
> >For anyone who's interested in this or other photodetection issues and
> >other low-noise optical measurements, I have some possibly interesting
> >papers at
> >http://www.pergamos.net.
> >(The exact URL for the paper containing the Gaussian shot noise plot is
> >http://users.bestweb.net/~hobbs/isicl/isiclao3.pdf.)
FOr all of 54 pence you can make an audio noise source using a PIC
12C508 .
Bipolar output with my modifications too .
Email me for the code or see www.gtonline.net/private/mandoline . I
can also supply pre-programmed chips if you are really stuck .
-Andre
John Larkin
<pc...@SpamMeSenseless.us.ibm.com> wrote:
Phil,
after due consideration, I have decided to
a) stick with zeners as noise sources and
b) buy your book, because it sounds cool.
John
Phil Hobbs
I hope you like it. The second printing has a number of corrections, so
if you have the earlier one, do get the errata from
ftp://ftp.wiley.com/public/sci_tech_med/electro-optical.
Cheers,
Phil Hobbs
Jonathan Kirwan
>FOr all of 54 pence you can make an audio noise source using a PIC
>12C508 .
>
>Bipolar output with my modifications too .
>
>Email me for the code or see www.gtonline.net/private/mandoline . I
>can also supply pre-programmed chips if you are really stuck .
I was just looking to see some interesting ideas on the front end of
photodiode systems, not for audio noise generators.
Jon