Variable filter circuit design. - RF Cafe Forums

RF Cafe Forums closed its virtual doors in late 2012 mainly due to other social media platforms dominating public commenting venues. RF Cafe Forums began sometime around August of 2003 and was quite well-attended for many years. By 2012, Facebook and Twitter were overwhelmingly dominating online personal interaction, and RF Cafe Forums activity dropped off precipitously. Regardless, there are still lots of great posts in the archive that ware worth looking at. Below are the old forum threads, including responses to the original posts. Here is the full original RF Cafe Forums on Archive.org

-- Amateur Radio

-- Anecdotes, Gripes, & Humor

-- Antennas

-- CAE, CAD, & Software

-- Circuits & Components

-- Employment & Interviews

-- Miscellany

-- Swap Shop

-- Systems

-- Test & Measurement

-- Webmaster

hawk7820

Post subject: Variable filter circuit design. Posted: Sat Feb 25, 2006 1:40 pm

Lieutenant

Joined: Sat Feb 25, 2006 1:39 pm

Posts: 1

Location: LONDON

hello guys, I am new here. glad to find such a good forum.

I am working on a project "optical listening device" whicn could be able to monitor the status of chmical pipes. we use a photodiode to detect the reflected laser beam by the mirror which is attached to the pipe. Then we gonna to analyze the siganl detected in terms of frequency.(Any fluid has a certain frequncy when it is flowing, any abstacle in the pipe will slow down the fluid and then change the freqency.)

our proposal is that we use a band pass filter to get rid of the noise and only focus on the spectrum that indicates the normal status of the pipes. This filter also acts as a refernce, which could distinguish the normal and abnormal(say the pipe is blocked) status in the pipe.And then we have to biuld a variable filter to detect the rest of the spectrum.

but now i am not clear how the variable filter works and how to compare the signal received and the reference signal, and then trigger an alarm if any abstacle in the pipes slow down the fluid too much or completely block the pipe.

looking forward to your reply.

cheers.

Top

IR

Post subject: Posted: Sat Feb 25, 2006 5:06 pm

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm

Posts: 373

Location: Germany

Hello hawk7820,

Welcome to RF Cafe Circuits Forum!

To answer your question the technology of implementing a filter depends pretty much on the frequency of operation, required bandwidth (BW) and other paramaters. It doesn´t necessarily has to be a classic variable RF Filter - most common one that comes to mind is by using Varactor Diodes. You can implement a filter by DSP (FIR, IIR) or an active filter with switched capacitors... in order to help you more please provide more information about the frequencis you work with the BW of operation etc..

You can compare between the detected and reference signals by different magnitudes: If the detected signal points over a blocked pipe, it will be out of the filter's pass region and therefore attenuated. The amount of attenuation points on how much the pipe is blocked..You have to tune the filter for that.

Good luck!

_________________

Best regards,

- IR

Top

fred47

Post subject: Variable FilterPosted: Wed Mar 01, 2006 1:28 pm

General

Joined: Wed Feb 22, 2006 3:51 pm

Posts: 104

Hi!

Here's another approach, based on defining your requirements:

You mention eliminating undesired noise. Have you categorized and described that noise? Is it constant or does it change? Is the noise correlated to the change you're trying to observe?

Do you know the frequency and bandwidth of the signal you're trying to observe? DSP techniques work well at lower RF frequencies, and adaptive filtering is quite advanced in that area, but it is sample-rate limited,

Over what range must the filter be tuned? A 1% variation is a lot easier than a 200% variation, especially at RF frequencies.

Good Luck!

Is filtering even the correct approach? Maybe spectrum analysis is better, given your needs.

Top

nubbage

Post subject: Variable Filter Circuit DesignPosted: Thu Mar 02, 2006 5:01 am

General

Joined: Fri Feb 17, 2006 12:07 pm

Posts: 218

Location: London UK

I agree with Fred47.

This looks more like a job exactly right for Fast Fourier Transform processing, or DSP, but a final choice would depend on the essential parameters already mentioned.

With such a software suite running on an industrial grade rugged PC, most of the electronic design is done, leaving you to concentrate on a roofing filter and low noise pre-processing front-end.

Posted  11/12/2012