PSN-L Email List Message

Subject: Re: noise
From: ChrisAtUpw@.......
Date: Sun, 3 Apr 2005 20:14:06 EDT


In a message dated 03/04/2005, 1goss@........... writes:

> I am having problems with noise during the night. Oddly it starts about 
> 9:00pm and ends about 11:00 am.
> 
> 9:00pm WinSDR screenshot  https://home.comcast.net/~bryangoss/noise9pm.jpg

Hi Bryan,

       This is NOT ocean background! The background is about 10 cycles per 
minute, so 10 mins should give ~100 peaks.  

> 11:00am WinSDR screenshot  
> https://home.comcast.net/~bryangoss/noise11am.jpg

       This could well be the ocean background.

> I thought it could be because the shop has a concrete floor and the air in 
> the shop heats up to around 70F during the day and cools to around 36F at 
> night. I believe the slower cooling of the floor during the night could be 
> causing convection in the box. I tried to seal the box off and covered it with a 
> blanket but that did not work. Could it be caused by the temperature change in 
> the mineral oil and the vertical dampener?

    Ouch! Gee! You need to keep the temperature a lot more constant. Seal the 
joints of the perspex case with the special 2" clear tape that you use to 
make glass joins in greenhouse windows. Then get a large cardboard box, seal the 
top joins with 2" gaffer tape and place it over the perspex box. Get a 15 W 
light bulb and place it inside the top of the perspex box. This heating should 
keep a stable temperature gradient inside, but you should really think long 
term of finding somewhere which only changes temperature by a few degrees from 
day to night. If the cardboard box and the bulb give a significant improvement, 
think in terms of making a 2" thick cellotex case - see the bottom photo at 
http://pages.prodigy.net/fxc/ 

    It can't be a direct effect of changes in oil viscosity, since the 
damping is greatest when the temperature is least and you see the opposite. I 
suspect that you may be correct in assuming that you have turbulent air cooling 
during the night.

> I ran FFT but to be honest, I'm not sure how to read it.

       The vertical axis is the logarithmic signal amplitude. The horizontal 
axis is the logarithmic frequency, with higher frequencies toward the right. 
It will be 100, 10, 1, 0.1, 0.01, 0.001 Hz. There is a cursor on the screen 
which reads out in frequency and period figures at the top of the screen. Your 
plot shows excessive low frequency signals, which could well be associated with 
cooling.

       Regards,

       Chris Chapman
In a message dated 03/04/2005, 1gos=
s@........... writes:

I am having problems with noise= during the night. Oddly it starts about 9:00pm and ends about 11:00 am.

9:00pm WinSDR screenshot  https://home.comcast.net/~bryangoss/noise9pm.jpg


Hi Bryan,

       This is NOT ocean background! The backg= round is about 10 cycles per minute, so 10 mins should give ~100 peaks. = ;

11:00am WinSDR screenshot = https://home.= comcast.net/~bryangoss/noise11am.jpg


       This could well be the ocean background= ..

I thought it could be because t= he shop has a concrete floor and the air in the shop heats up to around 70F=20= during the day and cools to around 36F at night. I believe the slower coolin= g of the floor during the night could be causing convection in the box. I tr= ied to seal the box off and covered it with a blanket but that did not work.= Could it be caused by the temperature change in the mineral oil and the ver= tical dampener?


    Ouch! Gee! You need to keep the temperature a lot more constant. Seal the joints of the perspex case with the special 2= " clear tape that you use to make glass joins in greenhouse windows. Then ge= t a large cardboard box, seal the top joins with 2" gaffer tape and place it= over the perspex box. Get a 15 W light bulb and place it inside the t= op of the perspex box. This heating should keep a stable temperature= gradient inside, but you should really think long term of finding somewhere= which only changes temperature by a few degrees from day to night. If the c= ardboard box and the bulb give a significant improvement, think in terms of=20= making a 2" thick cellotex case - see the bottom photo at http://pages.prodigy.net/fxc/

    It can't be a direct effect of changes in oil viscosity,=20= since the damping is greatest when the temperature is least and you see the=20= opposite. I suspect that you may be correct in assuming that you have turbul= ent air cooling during the night.

I ran FFT but to be honest, I'm= not sure how to read it.


       The vertical axis is the logarithmic si= gnal amplitude. The horizontal axis is the logarithmic frequency, with highe= r frequencies toward the right. It will be 100, 10, 1, 0.1, 0.01, 0.001 Hz.=20= There is a cursor on the screen which reads out in frequency and period figu= res at the top of the screen. Your plot shows excessive low frequency signal= s, which could well be associated with cooling.

       Regards,

       Chris Chapman

[ Top ] [ Back ] [ Home Page ]