Thermal and shot noise / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Thermal and shot noise

1 2 next

Move Thread

LAN_403

Ash Small

Sat Feb 03 2018, 11:41PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Can snubber caps help eliminate thermal and shot noise?

Uspring

Tue Feb 06 2018, 10:24AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Yes, they can. Whether that is useful depends on what you are trying to achieve. In signal processing applications there is often a noise component on top of your signal. Filtering that is possible, but that will affect also the signal. Caps will take away high frequency components of the noise but also those of the signal. So if you are also interested in high frequency components of the signal, that doesn't help.

Ash Small

Wed Feb 21 2018, 02:38AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I imagine thermal and shot noise to be a quantum thing, which manifests itself as a disturbance which induces an effect similar to that in a geiger counter (very oversimplified).

It's an experimental audio circuit I'm working on, with the ability to switch various different capacitors in and out, so I'll get it boxed up and see what happens.

I'm using diodes as a voltage reference, I've paralleled some, hoping that will help to reduce thermal and shot noise. Does this make sense, or am I on the wrong track?

Edit: The reference voltage is a part of the signal path.

Dr. Slack

Wed Feb 21 2018, 09:08AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

One two port component is one degree of freedom.

Thermal noise (equipartition noise) has a given power per per degree of freedom, the normal 4ktb thang.

If you change the impedance of the component, you change the ratio of voltage and current noise.

Shot and impulse noise occur in addition to thermal noise. Shot noise is a random thing which follows the sqrt(I) expectation, so higher current means better SNR.

Restricting your bandwidth reduces the total observed noise power, but it leaves the SNR unchanged at any given frequency. Restricting your observation bandwidth to your signal bandwidth is a standard processing thing you ought to be doing anyway.

Caps sold as 'snubber' are just as good as any other capacitor for building filters.

Be aware that voltage references are notoriously noisy, especially 'bandgap' types, which achieve their low tempco by taking a small difference between two larger values and amplifying it by a factor of 10. If you don't need thermal stability, then a silicon diode or two fed by a (quiet) constant current will be much quieter.

FWIW, the quietest noise reference I've seen described is a half-discharged lead acid cell. You can model that as a very large capacitor!

Sulaiman

Wed Feb 21 2018, 09:12AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Could you use lower noise bandgap (typ. 1v23) reference diodes such as LM385 ?

I'm not keen on the term 'snubber capacitor' in this context as I usually associate it with supression of transient switching spikes,
just 'capacitor', or 'bypass capacitor' would be more appropriate,
and yes, a capacitor will shunt current noise.
Can you show the part of the circuitry in question ?

Uspring

Wed Feb 21 2018, 10:00AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Thermal noise arises from fluctuations of thermal energy. More specifically, this energy is shared between motion of molecules and also minute electric fields in your device. There is a constant exchange between these energy forms, which leads to a fluctuation of voltages and currents.

Shot noise comes from the discrete values of e.g. charge. If you have a DC current of say, a 100 electrons per second, the number of electrons will rarely be exactly 100 in a given second, but will fluctuate around it. This creates noise.

Similarly in e.g. cameras, light is received in packets, aka photons, which will create noise in your image.

Ash Small

Mon Mar 12 2018, 11:48PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Thanks for the replies. I was using diodes (LED's) to give me the -ve reference voltage for the gate of a JFET. I've wired it up so I can switch between diodes and resistors, with a parallel cap, and I'm generally preferring the sound of the resistors, and the caps add some real fatness to the sound. Switching the caps in parallel with the diodes doesn't appear to make an audible difference.

There are some new SiC power JFET's out.....Watch this space

Edit: I associate 'snubber' with 'smoothing capacitor'.....It evens out voltage 'irregularities'

Sulaiman

Tue Mar 13 2018, 03:44AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

It seems that your understanding of 'snubber' is different;

International Rectifier

Cornell Dubilier
Snubbers are any of several simple energy absorbing circuits used to eliminate voltage spikes caused by circuit inductance when a switch â€” either mechanical or semi-conductorâ€”opens

Wikipedia
Snubbers are frequently used in electrical systems with an inductive load where the sudden interruption of current flow leads to a sharp rise in voltage across the current switching device, in accordance with Faraday's law. This transient can be a source of electromagnetic interference (EMI) in other circuits. Additionally, if the voltage generated across the device is beyond what the device is intended to tolerate, it may damage or destroy it. The snubber provides a short-term alternative current path around the current switching device so that the inductive element may be discharged more safely and quietly. Inductive elements are often unintentional, but arise from the current loops implied by physical circuitry. While current switching is everywhere, snubbers will generally only be required where a major current path is switched, such as in power supplies. Snubbers are also often used to prevent arcing across the contacts of relays and switches and the electrical interference and welding/sticking of the contacts that can occur (see also arc suppression).

Vishay
In power electronics applications, snubber capacitors are in charge of reducing the electromagnetic interference by clamping voltage and current ringing. Directly mounted on IGBT Modules, for stray inductance reduction, Snubber capacitors are supplied in configurable terminals for versatility purpose.

Ash Small

Tue Apr 24 2018, 06:42PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Sulaiman, I was thinking 'snubber' in the context of eliminating 'spikes' from shot noise. Maybe my interpretation is unconventional, but I see noise as spikes.

Now, a new question about shot noise in particular..... I've a hunch larger wattage resistors will result in less shot noise for a given current.

Does anyone have any 'opinion' either way on this?

Thanks in advance

hen918

Tue Apr 24 2018, 09:55PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

the problem is the 'spikes' are minuscule. When ever something is conducting very slightly, a small number of electrons are moving (in mostly random directions, but one direction more than the others). The thing is, the current required to get that low a number of electrons that you will be affected by shot noise is in the sub picoamp level. And capacitors ain't gonna stop it.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob