Roofing filters

[WA1GFZ]

Anybody out there know of any receivers that use a 70.455 MHz roofing filter?

[W1VD]

FT-1000?

[KL7OF]

can you explain what a roofing filter is? tnx...Steve

[Steve - WB3HUZ]

A roofing filter refers to the filter used in the first IF stage. It has a major impact on the close-spaced dynamic range of the receiver. Most tranceivers made in the last 20+ years usually have fairly wide filters in the first IF. This is to accomodate wider modes like FM (and to probably save some money). Those who want better close-in dynamic range will insert more narrow filters here. Many newer radios come with the narrow filters installed or they can be inserted via front panel settings.

[N3DRB The Derb]

I've heard slopbukits talk about them, but they always pronounced it "rooofin" so I never had any idea what they were talking about.

[VE7 Kilohertz]

can you explain what a roofing filter is? tnx...Seve

I think that if you live in the country, a roofing filter is that screen that goes over your woodstove chimney flue to keep the ashes from landing on your roof and setting your receivers on fire.   

Merry Christmas

Paul

[WA1GFZ]

TNX Jay the mp has one

[Steve - WB3HUZ]

Don't worry about it, neither did they.

I've heard slopbukits talk about them, but they always pronounced it "rooofin" so I never had any idea what they were talking about.

[WA1GFZ]

Paul,
Most Racals uses two 6 pole roofin filters in the first IF, 20 and 25 KHz wide.
These are the 40.455 KHz first IF filters. A few lucky people find the 8 KHz units and I hear the 1792 used a single 16 KHz. I'm looking for a couple 16 KHz units for my wide band hot rod unit. My NB RA6830 Hot Rod has a pair of 8s.
I'll be running quarter mile tests against it when HPSDR Mercury shows up.

[VE7 Kilohertz]

Hi Frank,

Keep me posted on your progress. I have the 1792 here and it is a bit better at pulling out weak signals but doesn't have the same audio quality as the 6790GM that I have made some mods to. I changed the AGC MED characterists and a few other things I can't remeber.

Cheers

Paul

[Steve - WB3HUZ]

Those are rather wide roofing filters. Most are going for 3 kHz these days.

[K1JJ]

I actually tried a roofing filter from INRAD for my FT-1000D last year. (For ssb use only)  The FT-1000 has an aftermarket 6khz Collins mech filter for TX and RX AM work.

It was a 2.4 khz? -  if I remember correctly.

Anyway, my after-market INRAD IF filter is 3.0 khz.   By adding in the narrower roofing filter, it limited the fidelity of the overall system, so I sent it back. It made a noticable difference in audio Yellowyness.

Supposedly they are a good addition to the overall system, but Inrad didn't have the wider ones available.

For the corntester and DXer looking for better selectivity numbers, it's probably OK.  If they made a 3kc model, I'd probably still be using it.

BTW, Chuck/K1KW tells me he tried one also and decided to send it back too. I shud have taken his advice, but had to find out for myself.


T

[WA1GFZ]

Paul,
The 1792 filter should drop right into the 6790. BTW I just found out the SD5000 first mixer board in the 6830 came right out of the 6793. So far the 6830 synthesizer is only in the 6830. It stuffed all the discrete logic in an LSI 40 pin chip.
Steve, a 3 KHz. roofer would sound crappy on AM. I added horse power in the first IF and second mixer of the NB hot rod so dynamic range limit is phase noise now.
The wide band hot rod I found a special Racal module for the second mixer from a 6840 that runs plus 23 dBM into the second mixer RAY6 and there is only about 10 dB of gain ahead of the second mixer. The  front end/first mixer module is passive with 10 dB of loss.
Tom, a 3 KHz roofer with the right gain distribution would work well in a first IF
for SSB. I just read the FT1000 has MDS /noise problems ahead of the second mixer and there is an inrad fix. I'd add a couple small relays so I could switch in the Inrad myself.

[Steve - WB3HUZ]

For sure, 3 kHz isn't for AM (although it could be with a sync detector). Then again, I don't think too many AMers are worried about working weak signals and have good close-in DR. But, many of the newer rigs allow you to select the roofing filter. Several radios allow you to select 15, 6 or 3 kHz. Most automatically switch in the correct roofing filter based on the bandwidth you select in the later/final IF.
 
For the weak signal CW types, a 3 kHz roofing filter is required, especially during corntests.

[WA1GFZ]

I don't have the room inside the radio to have a dual run of filters so just built two different configurations. I could build 2 custom modules if I didn't have SDR to play with.

[k4kyv]

For sure, 3 kHz isn't for AM (although it could be with a sync detector).

Then it would no longer be AM.  Just SSB with carrier - a different animal altogether.

A true sync detector receives both sidebands simultaneously and reinserts the carrier exactly on frequency and in phase (or 180° out of phase) with the original carrier so that both sidebands are demodulated and the resultant audio voltages add vectorially.

Some of the so-called "synchronous detectors" used on riceboxes are nothing more than solid state versions of the old "sideband slicers" that were popular in the 50's - aka phasing type SSB reception.  They do lock onto the AM carrier, but are capable of copying only one sideband or the other, but are not capable of coherent detection of both at the same time.

[Steve - WB3HUZ]

The person can transmit whatever they want. If I choose to only use the carrier and one sideband in detection, it's still AM.

Also, if my detector oscillator is synchronized to the incoming carrier, it's a synchronous detector. How many sidebands are involved  is irrelevant.

For sure, 3 kHz isn't for AM (although it could be with a sync detector).

Then it would no longer be AM.  Just SSB with carrier - a different animal altogether.

A true sync detector receives both sidebands simultaneously and reinserts the carrier exactly on frequency and in phase (or 180° out of phase) with the original carrier so that both sidebands are demodulated and the resultant audio voltages add vectorially.

Some of the so-called "synchronous detectors" used on riceboxes are nothing more than solid state versions of the old "sideband slicers" that were popular in the 50's - aka phasing type SSB reception.  They do lock onto the AM carrier, but are capable of copying only one sideband or the other, but are not capable of coherent detection of both at the same time.