Re: Measuring the noise floor

From: PaulCsouls (paulcsouls_at_worldnet.att.net)
Date: 03/03/05 

Date: Thu, 03 Mar 2005 04:40:39 GMT

On Wed, 02 Mar 2005 18:36:20 -0800, qrk <SpamTrap@reson.com> wrote:

>On Wed, 02 Mar 2005 20:49:29 GMT, PaulCsouls
><paulcsouls@worldnet.att.net> wrote:
>
>>
>>Thanks. I didn't know about the low noise setting, I'll have to check
>>the manual again. The resolution bandwith with the 3588 was 17k. The
>>4195A only lets me set 10k and 30k in that range. Though I would
>>expect to see a descrepancy between the data from the 3588 and 4195A,
>>the lack of change between the on and off conditions worries me about
>>the condition of the 4195A. The preamp is probably what I need. I'll
>>look into renting one of those when the 4195A come back from cal.
>>
>>Thank you
>>
>>Paul C
>>
>To get to the low noise mode, press the Receive Attenuation button.
>You will have a selection of
>IF RNG Normal
>IF RNG Low Distortion (only in spectrum mode)
>IF RNG Hi Sensitivity
>
>Low distortion and high sensitivity mode will give you a better noise
>floor. I'm a bit baffled by your bandwidth setting limitations. The
>4195 allows you to use bandwidth settings down to 3 Hz. Might take a
>few days to make the measurement at 3 Hz depending on your setup.
>
>What is your basic setup? Freq range, mode, circuit impedance, ...
>If your doing noise floor measurements, you want to use the spectral
>density mode (dBuV/Hz, dBm/Hz, uV/rtHz) which reports noise in a 1 Hz
>bandwidth.
>
>Mark

Yes I am using the 'noise' mode reading in dBm/Hz with a 1Hz bandwidth
and the 4195A does let me set the bandwidth down to 3Hz. I had set the
3588 to 17KHz and tried to get the same settings on the 4195A. The
4195A only allows certain quantum jumps in RBW ; 300Hz 1KHz 3KHZ 10KHz
30KHz. It lets you key in other settings but it jumps to some level. I
did try it down at 300Hz and it took a long time to make it across the
screen but it didn't help the readings. I missed the sensitivity
settings on the Receive Attenuation buttons. I'l have to try that. I
did play with attenuation settings but couldn't get the machine to
recognize the on from the off condition of the circuit. I know it's
probably operator error, that makes it all the more frustrating.

Paul C.

Quantcast