Solution needed for long time spectrum logging 1KHz - 1 MHz

[Thomas]

Hi all,

As one of our products seems to be disturbt by a local non permanent
EMC problem
we need to analyse what is happening on site.
We are looking for a solution to make long time ( say 4 weeks) data
logging of a spectrum.

Following specs are required:

Frequency span: 1KHz and 1 MHz (min 10KHz to 500 KHz)
Bandwidth: <= 10KHz
Dynamic range: > 25 dB

I was thinking abt a PC interface with "logging on trigger"
We need abt 5 "spectrum logger" so it is getting a bit price
sensitive...
Has anybody experiance with this cheap PC interfaces using FFT?

Any comment or proposal is highly welcome

Best Regards


Thomas

 

[Aidan Grey]

Hi all,

As one of our products seems to be disturbt by a local non permanent
EMC problem
we need to analyse what is happening on site.
We are looking for a solution to make long time ( say 4 weeks) data
logging of a spectrum.

Following specs are required:

Frequency span: 1KHz and 1 MHz (min 10KHz to 500 KHz)
Bandwidth: <= 10KHz
Dynamic range: > 25 dB

I was thinking abt a PC interface with "logging on trigger"
We need abt 5 "spectrum logger" so it is getting a bit price
sensitive...
Has anybody experiance with this cheap PC interfaces using FFT?

Any comment or proposal is highly welcome

Best Regards


Thomas

I don't know specifically what you will require for this job.

Places you might check are as follows:

http://www.agilent.com
http://www.ni.com
http://www.gage-applied.com

I don't know if any of the above can solve your problem. However, they
may give you an idea of what is possible. They may also be able to
advice you on whether a PC interface or standalone test equipment is
best.

I don't know if you can do FFT from 1 KHz to 1 MHz. A big frequency
span like that takes a lot of computing power. Most FFT spectrum analyzers
are from DC to something like 100 KHz.

You may want to also look into getting some kind of power line monitoring.
How do you know your problem is EMC?

You will also need to define a trigger condition. If something happens,
then
you record what is happening for 5 minutes for later analysis.

Aidan Grey

 

[Robert Lacoste]

Hi Thomas,

Why not just using a low-cost 12-bit / 2Msps A/D board in a vanilla PC, then
write a small software repetitively getting a buffer of say 2000 samples at
2Msps, then calculating the FFT, then the log spectrum power of the FFT ?
You will get a 1KHz resolution from 1Khz up to 1MHz, with around 30dB of
dynamic range... Then you will have to define your trigger settings and to
log the FFT when the trigger is exceeded. Labview will seems natural choice
to develop such algorithms quite quickly.

Let me know if I can help you more...

Yours,

 

[Thomas]

Thank you all for your helpful comments

After further investigation we will have a try with a PC based DSO
from picotech which can be used as a FFT spectrum analyser.

Their pre-sales support claims that they can do it. Have a look to
their helpul forum for the detailed response

http://www.picotech.com/support/viewtopic.php?t=942

Have a look to

http://www.picotech.com/digital_oscilloscope.html

for more details on the product

Best Regards


Thomas

 

[mike]

Hi all,

As one of our products seems to be disturbt by a local non permanent
EMC problem
we need to analyse what is happening on site.
We are looking for a solution to make long time ( say 4 weeks) data
logging of a spectrum.

Following specs are required:

Frequency span: 1KHz and 1 MHz (min 10KHz to 500 KHz)
Bandwidth: <= 10KHz
Dynamic range: > 25 dB

I was thinking abt a PC interface with "logging on trigger"
We need abt 5 "spectrum logger" so it is getting a bit price
sensitive...
Has anybody experiance with this cheap PC interfaces using FFT?

Any comment or proposal is highly welcome

Best Regards


Thomas

I have a friend who claims to be able to stuff an antenna into his VCR
video input. He can then go back and listen to the radio by listening
to the video out with a receiver. There are obvious problems
with the sync interval, but if you can manage to change tapes every six
hours, you might be able to get some useful information from it.
Might get you thinking about alternative approaches.
Depending on the circumstances, some of the FCC (or equivalent in your
country) field guys can be VERY helpful.
mike

--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Yaesu FTV901R Transverter, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/