John Wilke K9RZZ
11-17-2006, 04:39 PM
I posted this to the Palstar e-list and thought I'd repost it here:
<< This site if you are not aware of it has a lot of info on preselecctors.
http://www.geocities.com/amdxlog/mfj956.html
The MFJ-956 is about $60 and I understand it does a good job.
This guy does a lot of AM listening and has a lot of tricks to take care of
nearby AM stations.
>>
Boy, I can vouch for this guy, he has some good ideas. :cool:
I've been following his suggestions of lowering the impeadance of the LC circuit inside a preselector to raise the Q, and thereby selectivity ... and I am building my own.
http://members.aol.com/j999w/pa1.jpg
http://members.aol.com/j999w/pa2.jpg
My goal is to hear more European stations on the AM broadcast band (NOT easy from the midwest), so I need a _narrow_ preselector to hear those 9 khz splits. As a consistent testing signal, I use St. Kitts on 555khz. With the Palstar R-30 and a 150 ft, bent random wire, I can barely hear the carrier. With the preselector as you see it, the signal is loud enough, and the 550/560khz slop is lowered enough so that I can just make out audio, so it does help. When I get to the point that I can copy St. Kitts, 555khz regularly with the R-30 and a random wire, then I'll hook up my modified Icom R-71a, string some K9AY's or EWE's, and be in DX heaven. Taking the pre-selector on a mini-DXpedition to a local park with the R-30 and beverages ... would be the ultimate! :laughing:
I found a guy on eBay that winds those coils out of Litz wire for crystal sets and I also have a pair of Ramsey SA7 pre-amps that need to go in there. Those big FT-140-43 toroids are definitely overkill, but it seemed that smaller FT-82-75's had too much loss. (It might be a type 75 vs 43 material issue :confused: ). The biggest gain in selectivity is using the toroid transformers to reduce the impedance in the circuit as the guy in the webpage above talks about. Litz wire might be overkill as well, but I really wanted to make sure I got the highest Q circuit for maximum selectivity. Loss comparison is done by switching between the preselector in passive mode and just the R-30 with this 12:1 transformer:
http://members.aol.com/j999w/pa3.jpg
(which by the way makes a 20 db gain in signals compared to just jamming that random wire into the back of the R-30, and it helps throughout the shortwave bands).
I think with some rotary switches, this little box could be made into an active/passive - pre-selector/antenna phaser/notch filter !
I _really_ want to build a regenerative amplifier for this unit, such as the WA1ION MWT-3 ... then I would be bordering on selectivity that is too narrow. I've got the parts, just not the time to do that just yet.
When I run this gizmo down to longwave, those Litz wire coils really play and it takes a lot of re-tuning to cover the band, but I tell you, there's no AM BC slop down there. :p
When I'm happy with it's performance, it will all go into a nice metal box instead of this well used 'tupperware' container.
A project in evolution. :D
John K9RZZ
Milwaukee, WI
<< This site if you are not aware of it has a lot of info on preselecctors.
http://www.geocities.com/amdxlog/mfj956.html
The MFJ-956 is about $60 and I understand it does a good job.
This guy does a lot of AM listening and has a lot of tricks to take care of
nearby AM stations.
>>
Boy, I can vouch for this guy, he has some good ideas. :cool:
I've been following his suggestions of lowering the impeadance of the LC circuit inside a preselector to raise the Q, and thereby selectivity ... and I am building my own.
http://members.aol.com/j999w/pa1.jpg
http://members.aol.com/j999w/pa2.jpg
My goal is to hear more European stations on the AM broadcast band (NOT easy from the midwest), so I need a _narrow_ preselector to hear those 9 khz splits. As a consistent testing signal, I use St. Kitts on 555khz. With the Palstar R-30 and a 150 ft, bent random wire, I can barely hear the carrier. With the preselector as you see it, the signal is loud enough, and the 550/560khz slop is lowered enough so that I can just make out audio, so it does help. When I get to the point that I can copy St. Kitts, 555khz regularly with the R-30 and a random wire, then I'll hook up my modified Icom R-71a, string some K9AY's or EWE's, and be in DX heaven. Taking the pre-selector on a mini-DXpedition to a local park with the R-30 and beverages ... would be the ultimate! :laughing:
I found a guy on eBay that winds those coils out of Litz wire for crystal sets and I also have a pair of Ramsey SA7 pre-amps that need to go in there. Those big FT-140-43 toroids are definitely overkill, but it seemed that smaller FT-82-75's had too much loss. (It might be a type 75 vs 43 material issue :confused: ). The biggest gain in selectivity is using the toroid transformers to reduce the impedance in the circuit as the guy in the webpage above talks about. Litz wire might be overkill as well, but I really wanted to make sure I got the highest Q circuit for maximum selectivity. Loss comparison is done by switching between the preselector in passive mode and just the R-30 with this 12:1 transformer:
http://members.aol.com/j999w/pa3.jpg
(which by the way makes a 20 db gain in signals compared to just jamming that random wire into the back of the R-30, and it helps throughout the shortwave bands).
I think with some rotary switches, this little box could be made into an active/passive - pre-selector/antenna phaser/notch filter !
I _really_ want to build a regenerative amplifier for this unit, such as the WA1ION MWT-3 ... then I would be bordering on selectivity that is too narrow. I've got the parts, just not the time to do that just yet.
When I run this gizmo down to longwave, those Litz wire coils really play and it takes a lot of re-tuning to cover the band, but I tell you, there's no AM BC slop down there. :p
When I'm happy with it's performance, it will all go into a nice metal box instead of this well used 'tupperware' container.
A project in evolution. :D
John K9RZZ
Milwaukee, WI