(thanks to Cap W0CCA for sharing this)
This mod. deserves attention because it is so easy. The up/down conversion scheme of the Corsairs, going from 9 mhz to 6.3 and then back again and the passband tuning feature that is part of that 6.3 mhz IF has always made me curious how just a straight 9 mhz IF radio might be in terms of spurious response and overall noise. The Omni D/C is such a quiet radio, which does not have the 6.3 IF, I thought maybe this scheme could reproduce same. Actually, I think I made a better Omni D/C with this mod. Much better.
Since the PBT board reconverts to a 9 mhz IF prior to IF amplification and product detection, it is not hard to simply take the output of the Noise Blanker board which contains a sort of 2.4 Mhz “roofing filter” (4 or 8 pole depending on your radio) at 9 mhz, and inject that directly to the IF-AF Board. Familiarity with the plugs and understanding the circuit diagrams on the various boards will take an astute observer about 10 seconds for this trial, totally harmless and reversible. Just a plug in plug out, no soldering. Make sure you get the center of the mini coax on the right terminal when you do this. Wont hurt anything if wrong, just wont hear anything.
This simple experiment resulted in what I felt was a rx that could hear all the more complicated version could, but of course had a pretty wide passband, and seemed to me a LOT quieter, hmm, a lot less random noise without an antenna, hmm, and didn’t have a LOT of the birdies, hmm, led me to look at the PBT board a little. I decided that cutting the circuit at C31 downstream and connecting the filter side of that to the input terminal at terminal 33 and cutting just upstream of the 22pf capacitor near L1 (filter side) and hooking that to terminal 75 would still give me diode switched filters without any of the down conversion / up conversion circuitry of the PBT board. This of course means no PBT, but I am a CW only op and never saw how that control helped at narrow filter passband. (At least I never used it) I also prefer simplicity. I had previously totally bypassed the notch filter as well. This clipping of components can be done totally reversibly if one is careful.
Because the diode switching scheme has various attenuations for the various filters (2.4, 1.8, .5 and 0.25) due to losses varying from a few dB to 10 dB or more, you can pick which filter socket to use to try to balance gain, also a cool experiment. And you can totally jumper one ( did the 1.8 as in stock radio without) to have only that 2.4 roofing filter in on that filter position. BTW this totally screws up SSB receive filter sequence. But if you do it right you can still have SSB operation because it’s the “roofing filter” your SSB mixer needs.
The next bit of magic is I had a 500 hz 9 mhz Ten Tec filter on hand, 8 poles, and a 2.4 khz filter, somewhat butchered but also Ten Tec at 9 mhz. This is important because you need their center frequency with the filters, not one 1.5 khz off or so You can find them on Ebay if need be. All I did was plug those into the PBT board and voila, I had a two stage filter, 2.4 roofing filter (to use a term) and 500 or 2.4 following filters. All at 9 mhz, two mixers gone. The filter switching diodes still did the job. I jumpered the 1.8 position so I had a “wide open” filter position to listen to 2.4 khz of band with very flat skirts when I wanted to. My untrained and totally ham radio experimenters mind (I am a civil engineer, rocks and dirt and concrete and pipe) told me this was a quiet radio. In fact I thought it had no gain at all at first and the experiment was a failure until I hooked it to an antenna !! Voila, everybody was home and the spaces in between VERY quiet.
Not one to ever quit when a victory has been won, I went after the 2.4 khz 9 mhz 8 pole filter. My experience with the KX1 and Wilderness Sierra and HB1B variable bandwidth filters made me very curious about varactors. I replaced the capacitors in the 2.4 Khz filter with MVAM 109 varactors (Kits and Parts) and gathered 39k resistors from the varactor/crystal junctions to a 10k pot and thence to the 8 volt reg. of the Corsair and bingo, I had an adjustable crystal filter. I still have what seems like the 2.4 khz width or greater even, and can crank down to easily 200-300 hz using the pot at which point the inefficiency of my Rube Goldberg design undoubtedly kicks in. But no matter, works fine. With some initiative I can tweak that variable filter a little vis a vis the resistors used which determine voltage across the varactors and thus the capacitance, and heck I could even do a crude sweep for response using a signal generator and scope, but I am pretty happy simply to have it working right now. Next guy can do that and report back. Smile This second step is pretty cool and if you DO have to obtain another 9 mhz crystal filter the 2.4 khz ones are cheap or free compared to the 1.8, 500 and 250 hz variety.
The radio is now without question quieter than the Corsair II. I am still playing and the jury isnt all in yet, but it is a very fun, very reversible and good learning experience. There is to me nothing nasty, “boy is that a compromise”, about this mod. By the way you will notice the top picture of the Corsair doesn’t look like yours. 1) the original massive digital readout unit is replaced with an AADE unit, very simple mod. 2) I removed all that vox/processor/ SSB stuff on that switch board too as I am CW ONLY and I MEAN IT. I do not own a microphone. So who cares. Final touch a new panel and new smoked glass over the readout/S meter and a very different and very simple radio. I love it.
Don’t even start on this kind of effort unless you know what is going on and research each step and wire connection. These are old radios and one mis step can consign it to parts you will sell me cheap.