-
You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
455Khz. Ceramic filters & bleedover in Exports...
- Thread starter Shockwave
- Start date
Once you start going below 5 or 6 Khz. the filter cuts off too much treble in the receiver audio and everyone starts sounding muffled on AM. Better results are obtained with higher quality filters, having more poles and sharper bandwidth skirts.
I'm more familiar with the old school Murata filters in a metal case for 455 Khz. and seen good results using them in many receivers. These had 9 poles and many different bandwidths covering everything from SSB to FM. The one Kop has posted is interesting because it looks like it has a pair of tuned matching transformers.
That gives some flexibility in bandwidth adjustment if you can sacrifice a bit of gain. The input could be tuned a little higher in frequency while the output could be tuned a little lower, adjusting the overall bandwidth. I suspect it may work well when just tuned normally. The addition of tuned circuits alone, sharpens the bandwidth.
Last edited:
The same letter in different types or series of filters does not always indicate the same bandwidth. To be sure, we need to check the data sheet for the part in question. We also have to consider that while the CFW455HT is a 6Khz. filter for AM while it's only + or - 3Khz. from the center frequency and that's all the usable frequency response it can provide on AM.
From what folks were saying they were showing more like 8khz when swept. I just see the lescomm is doing narrower 10.7 and I think 455mhz as well to cut back on the hash and trash noise that's my goal to improve the rx on the ranger export I have.
Before you change AM filters to try and reduce noise, is the noise acceptable on SSB? If the signal to noise ratio has not noticeably improved on SSB with half of the receiver bandwidth, than we should not expect narrowing the AM filter to improve the results there either.
On the other hand, if most of the noise is high frequency content, narrowing the receiver filter removes some of that hash and you may find a better trade off between fidelity and signal to noise ratio, using a narrower AM filter.
On the other hand, if most of the noise is high frequency content, narrowing the receiver filter removes some of that hash and you may find a better trade off between fidelity and signal to noise ratio, using a narrower AM filter.
It's that Rci63ffc2 no ssb just am/fm cool loud radio for am which is pretty much where most if the locals stay here all 12 or so of them. Like I say was think the the 455i would be the way to go. What about narrowing the 7.8 how narrow can it go before degradation ?
From your chart looks to me like HUis the winner but is that a murata filter or another brand?
I can't find a schematic for that radio or even an internal picture and I'm a bit surprised that it would be using a 7.8 Mhz. filter without SSB. These filters are the crystal type rather than the 455 Khz. ceramic and have much sharper bandwidth skirts. It would be unusual for a CB or 10 meter rig to be using both IF filters simultaneously on AM. Although, many HF rigs do exactly that to provide the best performance. If this radio did, it should also have very good selectivity as a result.
Last edited:
I just wanted to show there were options. I do not however think that the 10.7~455 is all of the problem.
http://ps-2.kev009.com/jshorney/2510.htm Thanks to Jim Shorney we have this observation of the HR2510 filters compared to the HTX-100.
So far it seems we're all on the same page.
The difference between my communications receivers and a "CB-Export Radio" is what is allowed into the radio in the first place. We'll exclude the high IF modern radios and look at only 34-38MHz 10.7MHz mixes with a nod to the 7.8MHz mixes.
Let's get the brain strain out of the way first.
http://www.rfcafe.com/references/electrical/image-frequency.htm
as well as
http://www.arcticpeak.com/radiopages/imagefrequencies.htm
It does the math for you
and on to mixer spurs for the next part of the problem.
https://leleivre.com/rf_mixerspur.html
Now that we have an idea what can and does leak into our receiver via image and spur we need to examine just how much more of the radio spectrum we are listening to that is absolutely unnecessary.
Generally a CB radio listens to just .45MHz plus the skirts for a little over .6MHz at the 6db down points. Add the six or more "bands" added to an "Export" radio and we begin to have problems.
So it becomes a question of ratios and percentages, Listen through 600KHz of HF noise to get to your desired 450KHz bandwidth cut down to 6KHz at the 10.7 MHz (or 7.8MHz) filter and further at the 455KHz filter...
or
Expect a 6 "band" export radio to sample near 3MHZ of HF noise to get to your desired 450KHz bandwidth cut down to 6KHz at the 10.7 MHz (or 7.8MHz) filter and further at the 455KHz filter.
Do you think modern radios could benefit from a little band pass filtering on the front end ? Simply the bigger sample you take to get what you want the more crap you're going to get!
This isn't a how to or even a complete definition of the problem I have more questions than answers and only a few common sense practices to create a livable signal to noise ratio.
Comments ?
http://ps-2.kev009.com/jshorney/2510.htm Thanks to Jim Shorney we have this observation of the HR2510 filters compared to the HTX-100.
So far it seems we're all on the same page.
Okay, still with us but not the whole story.
The difference between my communications receivers and a "CB-Export Radio" is what is allowed into the radio in the first place. We'll exclude the high IF modern radios and look at only 34-38MHz 10.7MHz mixes with a nod to the 7.8MHz mixes.
Let's get the brain strain out of the way first.
http://www.rfcafe.com/references/electrical/image-frequency.htm
as well as
http://www.arcticpeak.com/radiopages/imagefrequencies.htm
It does the math for you
and on to mixer spurs for the next part of the problem.
https://leleivre.com/rf_mixerspur.html
Now that we have an idea what can and does leak into our receiver via image and spur we need to examine just how much more of the radio spectrum we are listening to that is absolutely unnecessary.
Generally a CB radio listens to just .45MHz plus the skirts for a little over .6MHz at the 6db down points. Add the six or more "bands" added to an "Export" radio and we begin to have problems.
So it becomes a question of ratios and percentages, Listen through 600KHz of HF noise to get to your desired 450KHz bandwidth cut down to 6KHz at the 10.7 MHz (or 7.8MHz) filter and further at the 455KHz filter...
or
Expect a 6 "band" export radio to sample near 3MHZ of HF noise to get to your desired 450KHz bandwidth cut down to 6KHz at the 10.7 MHz (or 7.8MHz) filter and further at the 455KHz filter.
Do you think modern radios could benefit from a little band pass filtering on the front end ? Simply the bigger sample you take to get what you want the more crap you're going to get!
This isn't a how to or even a complete definition of the problem I have more questions than answers and only a few common sense practices to create a livable signal to noise ratio.
Comments ?
Since a bandpass filters only rejects out of band signals, there would have to be an out of band signal strong enough to be causing problems in the receiver like desensing, before adding one would help noticeably. CB radios and exports already have a fixed tuned circuit in the front end, tuned for the center of the band. HF rigs have one for each band.
The best receivers adjust the resonance of the tuned filter circuits before the front end transistor, as the frequency of the receiver is changed. This allows the filter to track with the receiver frequency and the bandwidth of the filter can be much tighter than one passing the entire band. Then the front end can also reject strong, adjacent in band signals.
The best receivers adjust the resonance of the tuned filter circuits before the front end transistor, as the frequency of the receiver is changed. This allows the filter to track with the receiver frequency and the bandwidth of the filter can be much tighter than one passing the entire band. Then the front end can also reject strong, adjacent in band signals.
"If we assume that the USB and LSB carrier frequencies represent the edges of filter, we can see that the 2510 has a 5 KHz wide filter, while the HTX has a 3 KHz filter! This is enough to make a noticeable difference in the performance of the receiver."
That's exactly what they did with the Cobra 142 GTL in order to get away with one 5 KHz filter in the single conversion design. It provides poor sideband rejection and lousy high frequency response on AM. When the sideband offsets are more than 1.5 Khz away from center, you know you have a compromised crystal filter in use.
That's exactly what they did with the Cobra 142 GTL in order to get away with one 5 KHz filter in the single conversion design. It provides poor sideband rejection and lousy high frequency response on AM. When the sideband offsets are more than 1.5 Khz away from center, you know you have a compromised crystal filter in use.
Last edited: