RE: Tuned Stub Filter - alternate approach
I use a 1,2 or 3 cavity filter which has high Q and the loss is acceptable. A cavity filter in the 28/32 cm band is easy to assemble from a 2lb fruit can and if "tweaked" on a network analyzer can provide >30db rejection with a bandwidth of a couple of Mhz's. There are many texts describing their construction. Contact me off line if you have questions. Procedure I use is, set the filter to the desired center frequency of interest with signal gen, network analyzer, or tracking Gen / analyzer. Determine the insertion loss & record at the center freq. and make the measurement. Then correct for the filter loss. Take this number and factor against the fundamental and your home. I usually re-do at the -3db points of the filter as a sanity check. If the numbers look funny they probably are and you need to check the filter center freq. Repeat for each harmonic of interest. You will probably need to adjust your signal attenuation to keep the measurement in scale. A bit cheesy but worked well for RF guy's the last 60 years. Regards, Michael Taylor Hach Group Colorado >-Original Message- >From: richwo...@tycoint.com [mailto:richwo...@tycoint.com] >Sent: Wednesday, March 20, 2002 11:14 AM >To: emc-p...@majordomo.ieee.org >Subject: Tuned Stub Filter > > > >We are considering using a tuned stub filter to suppress the >carrier so that >we do not overload our spectrum analyzer while measuring >spurious emissions. >One question we have concerns the Q of the filter - i.e., will it >sufficiently suppress the modulation sidebands. So, let me ask >the following >questions. In the frequency ranges of 860-1000 MHz and 2.45 >GHz, how much >suppression should I expect from a stub filter and what design >steps can be >taken to have the necessary bandwidth? Or, am I going down the >wrong path >for a filter? > >Richard Woods >Sensormatic Electronics >Tyco International > This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ To cancel your subscription, send mail to: majord...@ieee.org with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Ron Pickard: emc-p...@hypercom.com Dave Heald: davehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: http://ieeepstc.mindcruiser.com/ Click on "browse" and then "emc-pstc mailing list"
RE: Tuned Stub Filter
>-Original Message- >From: richwo...@tycoint.com [mailto:richwo...@tycoint.com] >Sent: Wednesday, March 20, 2002 11:14 AM >To: emc-p...@majordomo.ieee.org >Subject: Tuned Stub Filter > > > >We are considering using a tuned stub filter to suppress the >carrier so that >we do not overload our spectrum analyzer while measuring >spurious emissions. >One question we have concerns the Q of the filter - i.e., will it >sufficiently suppress the modulation sidebands. So, let me ask >the following >questions. In the frequency ranges of 860-1000 MHz and 2.45 >GHz, how much >suppression should I expect from a stub filter and what design >steps can be >taken to have the necessary bandwidth? Or, am I going down the >wrong path >for a filter? > >Richard Woods >Sensormatic Electronics >Tyco International > Rich: I have tried using double-stub tuners to do this job, and it's usually not very good. I seem to recall getting about 35 dB of rejection from a double-stub, and the Q is not great. I now use a combination of K&L tunable band-rejection filters (which have a 5% wide >55 dB rejection band) and high-pass / low-pass filters. One problem with resonant band rejection filters is that if you tune them to notch out the fundamental, you also get rejection at harmonic frequencies too. So this makes measuring harmonic content a problem. I typically use the band rejection filter to notch out the fundamental while looking near the fundamental for spurious emissions. Looking below the fundamental for spurious, I use a low-pass filter. Above the fundamental, I measure harmonics with a high-pass filter that has a cutoff just below the second harmonic. Once above the second harmonic, you can usually use the high-pass filter for measuring spurious emissions too. Needless to say, after a while, you build up a decent collection of filters. (I especially like some high-pass filters I have made from waveguide; they give >80 dB well below cutoff, the cutoff is sharp, and the passband is just about zero dB loss. Actually, waveguides are band-pass filters, and you need to switch to a waveguide of smaller size as you progress up in frequency.) And remember, you have to characterize all of your filters. Ed Price ed.pr...@cubic.com Electromagnetic Compatibility Lab Cubic Defense Systems San Diego, CA USA 858-505-2780 (Voice) 858-505-1583 (Fax) Military & Avionics EMC Services Is Our Specialty Shake-Bake-Shock - Metrology - Reliability Analysis --- This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ To cancel your subscription, send mail to: majord...@ieee.org with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Ron Pickard: emc-p...@hypercom.com Dave Heald: davehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: http://ieeepstc.mindcruiser.com/ Click on "browse" and then "emc-pstc mailing list"
Re: Tuned Stub Filter
When using a stub, its reflected signal combines with the incident signal at the junction. If this reflected signal were equal in amplitude to the incident signal, it would offer (theoretically) infinite rejection. Because a stub has a non-zero loss, there is never complete cancellation. If the reflection is just 1 dB down, (about 89 percent amplitude) it will only give about 20 dB of attenuation. It is pretty easy to do better than this. But I'll stick with it for this example. If the signal is also not shifted exactly 180 degrees, as when slightly off frequency, full cancellation is not possible. If it is 10 degrees off, (900 Mhz instead of, say 810) then it is reduced to about 82 percent due to phase shift. If you had a 89 percent cancellation before, now you effectively have only about about 73 percent, and can expect a null of about 11 dB. This assumes nearly lossless coax. If you include the RC or RL phase shift due to loss, you get the total effect of imperfect Q. These numbers are close if I did the arithmetic right (grin). All this disregards the other effects on DESIRED signals. Placing a stub in parallel with the feedline parallels a reactive impedance which is numerically equal to the product of the tangent of its length in degrees and its characteristic impedance. A quarter wave 810 mhz open stub would at 405 Mhz be 45 degrees long and shunt 50 ohms of capacitive reactance with your coax. I have had some luck, in non-critical applications, with placing two tee-fittings or two splitters in a line, then adding a length of coax in parallel with the main feedline but a half wavelength (in coax) longer than the distance between the fittings. The signal is delayed 180 degrees and then summed with the original signal. If the two junctions are colocated (no distance at all, just four receptacles at one spot) this reduces to a parallel quarter-wave open stub. The farther apart the junctions are, the less upset (except for SWR in the paralleled section) you get. There are multiple notches due to this setup. When using tee's there are SWR reflections from the tee's, which might be reduced by using 93 ohm coax (100 ohms is perfect if it's available) for the notching section parts. If you use hybrids instead of tee's you avoid the matching problems. Let me know what you end up with. It's a fun experiment. Cortland --- This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ To cancel your subscription, send mail to: majord...@ieee.org with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Ron Pickard: emc-p...@hypercom.com Dave Heald: davehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: http://ieeepstc.mindcruiser.com/ Click on "browse" and then "emc-pstc mailing list"
