• 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...

Shockwave

Sr. Member
Sep 19, 2009
3,792
3,354
273
Anyone notice today's export radios with FM are more susceptible to adjacent channel interference on AM than an older radio like a Cobra 148GTL? If you don't care about FM or wire in some sort of switching, there is a way to improve that. Most radios with FM, only have one 455Khz. ceramic filter in their IF strip. In order to work with both AM and FM, a lousy compromise had to be made. FM on this band is plus or minus 5Khz. maximum deviation and the right filter for that would be 10Khz. wide. AM is about 6Khz. wide and these radios use a filter that is 8Khz. wide. Too wide for AM and too narrow for FM.

Get a Murata 455 filter out of an old 8719 Uniden board to use it for AM in these radios and notice the improvement. It's not just less bleedover and heterodynes. It also improves the signal to noise ratio by reducing the high frequency content in any white noise and interference.
 
Last edited:

In 148 AM is filtered by SSB crystal filter, that is why it is better.
Mike
The 148GTL has a 3Khz crystal filter @ 7.8Mhz. for SSB use only. The 148 did not compromise the performance on one mode for the other. It also has a 455Khz. IF strip that uses a 6Khz. wide filter on AM only. One bad example of a radio that used the SSB filter on AM is the 142GTL. Very narrow sounding with no treble on AM as a result and they had to shift the IF offset out twice as far to get the filter to work on SSB.

Update: In an effort to be completely accurate here, it just occurred to me that there is still one crystal inline with the 7.8MHz. IF strip on AM. This crystal is right in front of the dual gate 455 converter. While the crystal is inline on AM its purpose is to remove byproducts from the 7.8 mixer prior to the 455 conversion so they don't cause interference problems later in the circuit. It is the 455Khz. ceramic filter that determines bandwidth on AM.
 
Last edited:
very cool tip!

i hadn't noticed that they used an 8kc filter.

have they always done this with radios that had FM?
LC
 
very cool tip!

i hadn't noticed that they used an 8kc filter.

have they always done this with radios that had FM?
LC
I believe so. If you've ever listened to someone with heavy deviation on FM with an export receiver, you can hear much more distortion around the word peaks than you would using an HF rig on FM. That's because the export radios 8Khz. filter is chopping off both sides of the FM signals bandwidth.
 
Last edited:
  • Like
Reactions: NZ8N
I wonder if the RCI-2950 uses this?? Maybe clean up that white noise. I know I don't run FM at all.. Just AM and SSB..
Just about sure it does too. That's why the white noise on these export radios tends to be more high pitched and ear piercing. Some of that is cleaned up in the AF section but it doesn't work nearly as well as it does in the IF section with respect to the signal to noise ratio.

If you're curious as to the bandwidth of the filter in your radio. Open it up and look at the little black 455 rectangular shaped ceramic filter. After the 455 part number you'll see a letter like "H" or "K". That is the bandwidth indicator on the filter and a Google search for the datasheet will let you know what that filters specific characteristics are. If I recall correctly "H" was best for AM and "K" was just wide enough to be used on FM.
 
Last edited:
Just about sure it does too. That's why the white noise on these export radios tends to be more high pitched and ear piercing. Some of that is cleaned up in the AF section but it doesn't work nearly as well as it does in the IF section with respect to the signal to noise ratio.

If you're curious as to the bandwidth of the filter in your radio. Open it up and look at the little black 455 rectangular shaped ceramic filter. After the 455 part number you'll see a letter like "H" or "K". That is the bandwidth indicator on the filter and a Google search for the datasheet will let you know what that filters specific characteristics are. If I recall correctly "H" was best for AM and "K" was just wide enough to be used on FM.

Screenshot_20190328-004304~2.png

Look like 455H in the 2950DX model.. just found a pic.. looks like 3KHz to me. From Barkett page for 2950DX..

CFW455H

Screenshot_20190328-003345~2.png
 
Last edited:
The CFW455HT is a 6Khz. wide filter and ideal for AM so the problem may not be as widespread as I first thought. The 2950 must use some simple LC filtering in the FM IF strip because I don't think there is a wider FM filter in the radio and that one won't pass FM.
 
shockwave, im looking at the murata catalog in the CFW section and i can't find any with a 'K' suffix.

looking at a schematic of an old galaxy 99v, and comparing it to the schematic of a DX959 (no FM), it looks like they are calling out the same part: 455KCHT

is is possible that they mean 455KC (kilocycles) and the HT is the 6khz slope?

if that were the case, then as you said, they would be using LC filtering in the FM section instead.
LC
 
I did a little more research. It turns out most of these radios are using the correct "H" version with a 6Khz. bandwidth spec but these filters are typically testing at over 8Khz. wide! When I swept the 455Khz. IF strip, I found the same thing discussed in this INRAD article: http://mail.inrad.net/pipermail/list_inrad.net/2006-September/000739.html All tested at over 8Khz with an advertised spec of 6Khz. That is putting the adjacent channel rejection at around 50db. Not so good.
 
I did a little more research. It turns out most of these radios are using the correct "H" version with a 6Khz. bandwidth spec but these filters are typically testing at over 8Khz. wide! When I swept the 455Khz. IF strip, I found the same thing discussed in this INRAD article: http://mail.inrad.net/pipermail/list_inrad.net/2006-September/000739.html All tested at over 8Khz with an advertised spec of 6Khz. That is putting the adjacent channel rejection at around 50db. Not so good.

Any better swap in options for say the 2950??
 
Any better swap in options for say the 2950??
Yes, the "I" version of this part shaves 1Khz. off each side so it probably has a real spec of + or - 3Khz. Murata also makes metal cased versions of these parts that are larger and have more poles of filtering inside. They are much better than the smaller plastic ones but harder to find and more expensive. You could also cascade a pair of the "H" parts together in series for improved response.
 
When swapping 455KHz filter more important is BW at -60dB than on -6dB.
In my radio I have filter with BW 6KHz @ -6dB and 11KHz @ -60dB. That makes a difference.
I also have 4/8 filter, but it kills FM.
Mike
I think it's safer to say both numbers are equally important. That's because the bandwidth figures let you know how the filter will interact with the desired signal while the bandstop specs let you know how the filter will interact with undesired signals or interference. Another point is most bandstop figures are given at 60db down. These filters are lowering the bar a little by giving it at only 50db down.
 

dxChat
Help Users
  • No one is chatting at the moment.
  • dxBot:
    Tucker442 has left the room.
  • @ BJ radionut:
    LIVE 10:00 AM EST :cool:
  • @ Charles Edwards:
    I'm looking for factory settings 1 through 59 for a AT 5555 n2 or AT500 M2 I only wrote down half the values feel like a idiot I need help will be appreciated