q: Watts per root-Hertz?
D. John Lee
A friend of mine was talking about a spectrometer in some project
they were managing at work, and were puzzled by the fact that
the scientist owning the instrument mentioned that the sensitivity(?)
was given in units, if I recall correctly, of Watts per square-root
Hertz (W*Hz^(-0.5)). I could not come up with any meaningful
interpretation for such a collection of units. It is the
square root that really confuses me.
I stress that I, or my friend, may be misquoting the units, but I
was just curious if anyone knew of any sort of meaningful
related quantity with Hertz^(0.5) in the denominator. Any
relevant tutorial info on spectrometer specs like this would be
appreciated.
Thanks
Dave
Doug Goncz
ratio.
>units, if I recall correctly, of Watts per square-root
>Hertz (W*Hz^(-0.5)).
This is inconsistent. Is it W/(Hz^0.5) or W*(Hz^0.5)?
Yours,
Doug Goncz,
Experimental Machinist ( DOT 600.260-022 ) ( A.A.S.M.E.T. )
Replikon Research ( USA 22044-0094 )
http://users.aol.com/DGoncz or /ReplikonVA
Mark W. Lund, PhD
Noise power is proportional to the bandwidth, the more hertz
of bandwidth the more noise. But most detectors have a signal
that comes out in volts or amps, not watts, so the noise
would be in volts/root hz or amps/root hz, since the power is
proportional to the square of the volts or the amps.
So I agree with you that watt/root hz is probably wrong.
best regards
mark
"D. John Lee" wrote:
--
----------------------------------------------------------------------
Mark W. Lund, PhD "He represented, indeed to a superlative
VP, Engineering degree, the great moral fallacy of our
MOXTEK, Inc. time, that collective virtue may be pursued
452 West 1260 North without reference to personal behavior."
Orem Utah 84057 --Malcolm Muggeridge
SOFT X-RAY WEB SITE http://www.moxtek.com
Brigham Young University e-mail: lu...@xray.byu.edu
tar...@imap2.asu.edu
: Hi:
: A friend of mine was talking about a spectrometer in some project
: they were managing at work, and were puzzled by the fact that
: the scientist owning the instrument mentioned that the sensitivity(?)
: was given in units, if I recall correctly, of Watts per square-root
: Hertz (W*Hz^(-0.5)). I could not come up with any meaningful
: interpretation for such a collection of units. It is the
: square root that really confuses me.
He probably means noise equivalent power (NEP), which usually has those
units. Essentially, it means that an input signal with that power at 1 Hz
bandwidth has an SNR of 1. It attempts to represent the minimum
detectable signal, which is why he called it sensitivity.
Ben B.
Kent Nickerson
a sensitivity expressed in Watts/Hz, or "Volts per root Hertz" since the
background thermal noise power *and* electrical shot noise both have
power which is proportional to bandwidth. (In optics, photon
statistical noise and electronic shot noise are indistinguishable except
in things like photomultipliers, which I assume isn't being used).
However, incoherent detectors used in the vast majority of instruments
(and certainly wide band one like spectrometers) generate an output
voltage or current which is proportional to input power. The
sensitivity of the detector is set by thermal or shot noise (units =
V/root Hz) AFTER the detector. When this noise floor is referred to the
signal hitting the detector, it translates to signal Watts per root
Hertz. Again, an artifact of incoherent detection.
Kent
D. John Lee wrote:
>
> Hi:
>
> A friend of mine was talking about a spectrometer in some project
> they were managing at work, and were puzzled by the fact that
> the scientist owning the instrument mentioned that the sensitivity(?)
> was given in units, if I recall correctly, of Watts per square-root
> Hertz (W*Hz^(-0.5)). I could not come up with any meaningful
> interpretation for such a collection of units. It is the
> square root that really confuses me.
>
Scharnberg F_alk
On 11 Sep 1999, Doug Goncz wrote:
> >units, if I recall correctly, of Watts per square-root
> >Hertz (W*Hz^(-0.5)).
>
Sorry, it is not.
W * Hz^(-0.5) = W / (Hz^0.5)
Falk
PS: If you don't believe me:
y * x = 1 => y = 1/x
y * x^1 = x^0 => y = x^(0-1)
Nelson Wallace
My guess is your friend was talking about "D-star", which has units of
( cm * sqrt(Hz) ) / watt
It is a common measure of sensitivity for infrared instruments.
It is the inverse of the noise equivalent power, with the detector area
and the electrical frequency factored out, so various systems can be
compared.
For more on this, take a look in R. D. Hudson's classic book, "Infrared
System Engineering", Chapter 7, equation 7-7.
Then again, your friend could have been talking about something else.
Regards, Nelson Wallace
http://home.earthlink.net/~nwwallace/index.htm
___________________________________________________________
David Moodie
>
> Hi John,
>
> Noise power is proportional to the bandwidth, the more hertz
> of bandwidth the more noise. But most detectors have a signal
> that comes out in volts or amps, not watts, so the noise
> would be in volts/root hz or amps/root hz, since the power is
> proportional to the square of the volts or the amps.
>
> So I agree with you that watt/root hz is probably wrong.
>
> best regards
> mark
>
> "D. John Lee" wrote:
>
W/root Hz is probably correct in this situation, I belive you didn't
take account of the conversion factor for the photoreceiver and as such
amps (volts after the amp) are proportional to optical power.
The noise equivalent power (NEP) of a photoreceiver is often refered to
the input, hence W/root Hz is used where W is the optical power
(propartional to voltage at the output of the receiver).
In the case of photodiodes the NEP is given by
noise current (A/root Hz)
------------------------
photo sensitivity (A/W)
which is thus expressed in W/root Hz
the NEP is effectively the light level required to obtain a signal level
of unity.
regards
David
Dr. Mark W. Lund
conversion. I haven't worked in infrared systems design for 10
years, and am working with a used brain, so I am likely to make
a few gaffers! :)
best regards
mark
David Moodie wrote:
--
EandorY
<snip>
>I ...am working with a used brain, so I am likely to make
>a few gaffers! :)
Heh-heh-heh.
What's the secondary market value on used brains? Can you get them
rebuilt/refurbished/modified? I'm in the market for an upgrade - how do I
find a good one? Should I take it to a specialist and get it checked out
before buying? Is there an industry spec for MTBF?
Also, is it better to be the owner of a used brain, or one that's still in
mint condition? Are the unused ones more valuable as collector's items?
Are they in heavier demand around the Washington, D.C., or the Los Angeles
area?
Eric
(Sorry, I guess that's a little off-topic)