input power of spectrum analyzer

[krishmaniac@hotmail.com]

I have a spectrum analyzer with 50ohm input. It is marked the max.
30dBm/50VDC at the input. The 30dBm roughly calculates to be 136dBuV
at 50ohm impedance. The signal i am observing has frequency components
at f1,f2,...fn with magnitude close to 110dBuV. I did not understand
what the max.30dBm/50VDC at the input of analyzer stands for. Is it
the maximum signal level at any frequency level or is it the total
power of the signal at 50ohm should be less than 30dBm? if its the
power, how can i know how much power in the spectrum as the spectrum
may have infinite frequency components. Someone please clearify me
this doubt.
thanks
kristo

 

[Max65]

On 14 Set, 05:02, "krishman...@hotmail.com" <krishman...@hotmail.com>
wrote:

I have a spectrum analyzer with 50ohm input. It is marked the max.
30dBm/50VDC at the input. The 30dBm roughly calculates to be 136dBuV
at 50ohm impedance. The signal i am observing has frequency components
at f1,f2,...fn with magnitude close to 110dBuV. I did not understand
what the max.30dBm/50VDC at the input of analyzer stands for. Is it
the maximum signal level at any frequency level or is it the total
power of the signal at 50ohm should be less than 30dBm? if its the
power, how can i know how much power in the spectrum as the spectrum
may have infinite frequency components. Someone please clearify me
this doubt.
thanks
kristo

The marked value refers to the maximum total power allowable to avoid
any thermal damage at the input front-end (the DC voltage is just the
input coupling capacitor maximum voltage).
The thermal conversion of EM energies no matter of the frequency
components of the signal, therefore any harmonic components should be
considered and summed to compute the total thermal energy. Usually,
the harmonic components are not considered indeed, because they are
quite low in energy if compared to the fundamental signal energy.
E.g. whether a very "dirty" signal has it's third harmonic component
10db under the fundamental, the fundamental participates to the
thermal effect ten times more than its third harmonic, and we are
talking about a very "dirty" signal (for signals having harmonic
components 20dB under the fundamental, those harmonics participate
just for 1% of the total).
Anyway, if you need to measure that kind of distorted signals, or
other composite signals, take care of the total power to avoid damages
at the input attenuator, or worse, at the 1st mixer.
Have a nice day.
Massimo.

 

[Fred Abse]

On Thu, 13 Sep 2007 20:02:02 -0700, krishmaniac@hotmail.com wrote:

I have a spectrum analyzer with 50ohm input. It is marked the max.
30dBm/50VDC at the input. The 30dBm roughly calculates to be 136dBuV at
50ohm impedance. The signal i am observing has frequency components at
f1,f2,...fn with magnitude close to 110dBuV. I did not understand what the
max.30dBm/50VDC at the input of analyzer stands for. Is it the maximum
signal level at any frequency level or is it the total power of the signal
at 50ohm should be less than 30dBm?

"Maximum input" means just that. The total *power* (30dBm = 1 Watt) which
may be applied to the input without damage to the input attenuator or
mixer, irrespective of its spectral makeup.

More importantly, you need to know the input level for a specified degree
of compression, (usually 1dB), which should be specified in the manual.
That will be some way below maximum permissible, (I'd guess +10dBm).
Exceeding this will result in intermodulation products occurring in the
instrument, and what you see won't be what you really have.

if its the power, how can i know how
much power in the spectrum as the spectrum may have infinite frequency
components. Someone please clearify me this doubt.

You need to know the total (broadband) power of your signal. Use some sort
of broad(ish) band power meter (thermocouple, bolometer, etc.) first, to
get an idea.

Then use additional attenuator(s) at the analyzer input. to get the total
power down to (say) -10dBm, at which the analyzer will be happy. Power
attenuators can be expensive, especially ones with flat characteristics
extending up into the high UHF range.

Be careful, a zapped analyzer frontend can cost several months' salary to
get fixed, not to mention the nose bleed, if it belongs to someone else

There was (probably still is), a good application note on the use of the
spectrum analyzer on the Agilent site. It has, among other things, one of
the best explanations I've seen about why the need to apply corrections to
noise measurements when using video (VBW) filtering.


--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[Fred Abse]

On Fri, 14 Sep 2007 05:37:33 -0700, Fred Abse wrote:

There was (probably still is), a good application note on the use of the
spectrum analyzer on the Agilent site. It has, among other things, one of
the best explanations I've seen about why the need to apply corrections to
noise measurements when using video (VBW) filtering.

http://www.home.agilent.com/agilent/redirector.jspx?action=ref&cname=AGILENT_EDITORIAL&ckey=1000000159%3Aepsg%3Aapn&lc=eng&cc=US&nfr=-536902453.0.00

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[Wayne P. Muckleroy]

The 50V input refers to an impulse or static voltage that can come from ESD
or a (long) charged up cable. Always dicharge yourself and the connecting
cable (center-conductor) before connecting up to the sensitive front-end.

<krishmaniac@hotmail.com> wrote in message
news:1189738922.158443.45840@19g2000hsx.googlegroups.com...

I have a spectrum analyzer with 50ohm input. It is marked the max.
30dBm/50VDC at the input. The 30dBm roughly calculates to be 136dBuV
at 50ohm impedance. The signal i am observing has frequency components
at f1,f2,...fn with magnitude close to 110dBuV. I did not understand
what the max.30dBm/50VDC at the input of analyzer stands for. Is it
the maximum signal level at any frequency level or is it the total
power of the signal at 50ohm should be less than 30dBm? if its the
power, how can i know how much power in the spectrum as the spectrum
may have infinite frequency components. Someone please clearify me
this doubt.
thanks
kristo