A little research shows found crystal filters with impedances ranging from 50 ohms to 500 ohms. I speculate that the 10 MHz - 50 ohm filters were internally matched to 50 ohms.
I was going to say from the start of the video that 500 Ohms would be about the right value. I've seen other filters like this one designed for 500 Ohm source and load impedance.
A simple L C network on each end will reduce insertion loss and smooth the pass band ripple. W7ZOI LADPAC software will design that for you painlessly. I have used his software to design crystal filters and LC filters often for some impedance higher than 50 ohms but then matched to 50 ohms. And to match INRAD filters to 50 ohms.
As I am playing around with this kind of things with the SSA 3021X these days I recommend to set the sweep to "manual" and go down to 0.5s as long as you are still far from the correct resistance values. When getting closer more precision is helpful again. I also like the multiple traces stored in "view" mode for comparison, but I still have to get used to the normalization.
Good video! FYI, you could hit the BW key to speed up the sweep rate for that test. Also, nice to see your SA goes to 3.2GHz :) (Mine does too! ) hihi 73
Yes a 1:4 unun would match well it , the peaks and valley on the top probably are caused by the resistors loss,. Using a VNA could also been useful because many experimenters have one.
Nice one there, fortunately most of my filters are 50ohm so I dont bother but when I need to I do exactly this ! Have you considered building a pair of TIA (Termination insensitive Amplifiers ) that the guys at "Solder Smoke" like so much and trying them between the inputs and outputs of the filter section much as they do in the IFs of their radio builds
I hope you did not rip that out of an IC211E and throw the rest away... My first base station rig! I've still got a soft spot for it especially the electromechanical tuning control. As suggested below a VNA could be used to measure the impedance, but have you ever tried to measure it using the very old school method of half voltage? I tried this years ago but my scope was not up to the task, now I've got a decent scope the filter box is empty! Feed in a signal via something like a 10K resistor, measure the voltage across the filter input. Connect the pot and adjust for half voltage. The pot should then be the same value of the input impedance. Andy
A Frequency Modulated channel is about 150 kHz wide. What good is a filter whose bandwidth is so narrow? Also, 600 Ohm is a fairly standard signal generator impedance.
it passes (no attenuation) a band of frequencies and rejects (attenuates) all other frequencies. ua-cam.com/video/hf4grkMuql4/v-deo.htmlsi=aZ27MbnRSdHKYekz
I use matching transformers to get impedance's a little closer and measure with my VNA ... seems to work pretty well ...
That was extremely informative. Thank you
A little research shows found crystal filters with impedances ranging from 50 ohms to 500 ohms. I speculate that the 10 MHz - 50 ohm filters were internally matched to 50 ohms.
I was going to say from the start of the video that 500 Ohms would be about the right value. I've seen other filters like this one designed for 500 Ohm source and load impedance.
Agreed. I think every filter of this style I have ever seen in amateur radios and CB radios had 500 Ohm impedance filters.
A simple L C network on each end will reduce insertion loss and smooth the pass band ripple. W7ZOI LADPAC software will design that for you painlessly. I have used his software to design crystal filters and LC filters often for some impedance higher than 50 ohms but then matched to 50 ohms. And to match INRAD filters to 50 ohms.
As I am playing around with this kind of things with the SSA 3021X these days I recommend to set the sweep to "manual" and go down to 0.5s as long as you are still far from the correct resistance values. When getting closer more precision is helpful again.
I also like the multiple traces stored in "view" mode for comparison, but I still have to get used to the normalization.
Good video! FYI, you could hit the BW key to speed up the sweep rate for that test. Also, nice to see your SA goes to 3.2GHz :) (Mine does too! ) hihi 73
Yes a 1:4 unun would match well it , the peaks and valley on the top probably are caused by the resistors loss,. Using a VNA could also been useful because many experimenters have one.
it might be good old fashioned ripple. I've done it with pots and proper matching -- little difference measured in many cases.
All the best stuff is made in Japan...
Not necessarily. The best tequila still comes from Mexico. :)
That's heavy, doc!
"When this radio gets up to 88, you're gonna see some serious sh1t!"
Nice one there, fortunately most of my filters are 50ohm so I dont bother but when I need to I do exactly this ! Have you considered building a pair of TIA (Termination insensitive Amplifiers ) that the guys at "Solder Smoke" like so much and trying them between the inputs and outputs of the filter section much as they do in the IFs of their radio builds
It won't help in testing much, you still need to know the design impedance of the filter. It is a good idea for the radio design.
@@IMSAIGuy Agreed may well be a interesting device to talk about regardless.....Keep it up !!
I hope you did not rip that out of an IC211E and throw the rest away... My first base station rig! I've still got a soft spot for it especially the electromechanical tuning control.
As suggested below a VNA could be used to measure the impedance, but have you ever tried to measure it using the very old school method of half voltage?
I tried this years ago but my scope was not up to the task, now I've got a decent scope the filter box is empty!
Feed in a signal via something like a 10K resistor, measure the voltage across the filter input. Connect the pot and adjust for half voltage. The pot should then be the same value of the input impedance.
Andy
I did not
Great video but why not measure input and output impeadance with vna before setting resistance?
A Frequency Modulated channel is about 150 kHz wide. What good is a filter whose bandwidth is so narrow? Also, 600 Ohm is a fairly standard signal generator impedance.
narrow FM modulation used in Ham Radio is only 11kHz wide. This filter is most likely a SSB filter
you dismantled and junked one of the most beautiful transceivers ever made, the Icom IC-211. I just restored one.
I did not
A spectrum analyzer input is 50 Ohms unless it is 75 Ohms lol !
What kind of ham stuff are you into?
old extinct hobby. WAS, WAC, DXCC. I randomly turn on the radio and run FT8 these days
@@IMSAIGuy Yeah I don't like the contesting stuff; I am interested in FT8 and similar.
What is a band pass and what is this filter doing to the band pass?
it passes (no attenuation) a band of frequencies and rejects (attenuates) all other frequencies. ua-cam.com/video/hf4grkMuql4/v-deo.htmlsi=aZ27MbnRSdHKYekz
Well nice and good but you didn't terminate both coaxes properly with 50 ohms. Likely doesn't matter much at only 10MHz...
2.64 kHz is lousy for a FM filter.
2.65 kHz is an appropriate value for single sideband voice communication.
You should have used those blue Bournes (10) turn pots. You must watch Dave Jones - "Trap for Young Players" LOL