Pages: [1]   Go Down

Author Topic: Testing Unknown Mechanical Filter  (Read 370 times)

N2LXM

  • Member
  • Posts: 137
Testing Unknown Mechanical Filter
« on: December 04, 2021, 01:21:08 PM »

While digging through my storage aera for parts for a winter project. I found what appears to be a small mechanical filter. The only marking on it are 512453. What would be a good way to test it? Not knowing the center frequency or bandwidth. As for equipment, I have a HP 8061A RF / sweep generator, Frequency counter, Oscilloscopes, and a LP technologies, LPT-2250 150KHz - 10000 MHz Spectrum analyzer. Any information would be helpful.
Logged

HAMHOCK75

  • Member
  • Posts: 1297
Re: Testing Unknown Mechanical Filter
« Reply #1 on: December 04, 2021, 03:20:05 PM »

If you can afford a $50 nanoVNA, testing ( even repairing ) mechanical filters can be relatively painless.

https://www.eham.net/community/smf/index.php/topic,127583.msg1155979.html#msg1155979
Logged

VE3NNM

  • Posts: 52
    • Archive (2019) of my now-defunct web site
Re: Testing Unknown Mechanical Filter
« Reply #2 on: December 04, 2021, 06:12:27 PM »

I agree with Hamhock, a nanoVNA would be great for an application like that, VK3YE and VE6WGM are doing crystals not mechanical filters in the following videos but you can get the gist of the process anyway.

https://www.youtube.com/watch?v=0lsWqKtLTfA

https://www.youtube.com/watch?v=rVIHVi-7brs

However since you already have a spectrum analyzer you might want to whip up a quick wideband noise generator rather than wait for your nanoVNA order to be delivered. Since you don't need high-end features such as a consistent output level from DC to daylight or being able to program a specific output level a hobby-grade circuit would be just fine. All you need to do is drive the filter with noise and look on the spectrum analyzer for a glaringly obvious change in level. I wouldn't go to wide though or you might not see the blip. Start low and then you can just tune your Spectrum Analyzer's center frequency upwards until you see something.

http://www.n5ese.com/noise.htm


https://kb8ojh.net/projects/noisegen/
The original article (with circuit corrections in the comments) the above link references is gone from the web, it's available on the Internet Archive:
https://web.archive.org/web/20170919232519/eeweb.com/blog/extreme_circuits/simple-white-noise-generator

« Last Edit: December 04, 2021, 06:27:50 PM by VE3NNM »
Logged

W9AC

  • Member
  • Posts: 373
Re: Testing Unknown Mechanical Filter
« Reply #3 on: December 05, 2021, 07:36:46 AM »

Adding to the list of references, here's a link to an article authored by Don, W5QN and myself that appeared in a 2015 edition of The Colllins Signal magazine. 

https://www.immigration-lawyer-us.com/images/q32015.pdf

Note the use of series caps on the test jig to resonate the Collins FA filters.  Also notice that a spectrum analyzer with tracking generator was used for testing but that's easily replaced with a VNA.  The article predated widespread use of the nanoVNA device.   The end of the article shows an extremely rare photo of myself -- one of the few known to exist. 

Paul, W9AC
Logged

VE3WGO

  • Member
  • Posts: 666
Re: Testing Unknown Mechanical Filter
« Reply #4 on: December 05, 2021, 09:56:37 AM »

or keep it simple.  No nanoVNA, no noise generator.  you already have more than you need.

your 8601A is a superb sweeper, so just feed a low level signal (0 dBm for example) into the filter through perhaps a 5k ohm resistor, and terminate the other end of the filter with a similar 5k resistor, then watch the output on a scope.  You will find the centre frequency in seconds.

Most of these filters are somewhere between 455 kHz (mechanical filters) to 9 MHz (crystal filters).

73, Ed
« Last Edit: December 05, 2021, 10:00:03 AM by VE3WGO »
Logged

N2LXM

  • Member
  • Posts: 137
Re: Testing Unknown Mechanical Filter
« Reply #5 on: December 05, 2021, 11:38:44 AM »

I was able to hook up the filter an found its center frequency. It is a 455 KHz filter with a bandwidth of about 3 KHz. Thank you, all the great information.
Logged

K6BSU

  • Member
  • Posts: 313
Re: Testing Unknown Mechanical Filter
« Reply #6 on: December 16, 2021, 04:52:51 PM »

But now you need to find the terminating impedances.
Logged

N2LXM

  • Member
  • Posts: 137
Re: Testing Unknown Mechanical Filter
« Reply #7 on: December 17, 2021, 12:25:21 PM »

Thank to all with your input on this subject. Once I get the IF strip I am working on going, I will then spend some time researching the termination for the filter. And if all goes well I will be able to use it. Old Man winter is coming, So this will be a nice project of the winter.
Logged
Pages: [1]   Go Up