[Drake] Roofing Filters

[Bob Camp]

Hi

Keep in mind that the filter is not the only thing that determines the 
dynamic range of of a radio. The phase noise of the local oscillator 
also has a major impact on close spaced third order dynamic range.  For 
a lot of years the R-4 series has had a ranking near the top of the 
list for phase noise performance. Mixers, filters, and switching also 
get into the act. Sometimes they do it  in a big way, sometimes not.

Another item that you can get hung up on is a specific dynamic range 
measure. You can focus on a third order close spaced measure or on wide 
spaced second order number. They are radically different things but 
booth evaluate the performance of a radio. A radio with a tuned front 
end like the R-4's or an R-390 will do significantly better on second 
order intercept than a broad band front end radio.

Unfortunately when you hook a real radio up to a real antenna a whole 
bunch of things happen all at the same time. Depending on which 
antenna, which radio, and what time of which day you may be hit by one 
thing or by another. In some situations a radio with a negative 
intercept for both second and third order will do a better job than a 
radio with 80 and 40 db intercepts. It all depends on what the limiting 
factor is right now with your antenna listening to one specific signal.

Even if you are talking about a CW contest situation with six foot tall 
helical front end filters on the antenna, third order may not be all 
you have to worry about. I am always amazed at how well some radios 
hear things with no antenna attached.  The IM performance of your radio 
may actually be better than the performance of that ten year old  coax 
connector on the antenna ....

A simple real world check on all of this:

Pop the antenna on the radio and make sure the noise out of the radio 
goes up when you are tuned to a quiet part of the band you want to use.
Stick in a 10 db pad. Does the noise still go up? if so keep stepping 
up the pad.

Once you have a pad that keeps you from hearing antenna noise back off 
by 10 db. Most of the time you will find that you have a 10 or 20 db 
pad on the lower HF bands and a 10 db pad on 20 meters. Of course this 
trick only works if the band is open....

At this point you are still noise limited by the antenna and not by the 
pad and receiver combination.

Next tune over to a busy part of the band and try to tune a weak 
signal. Drop the pad down by 5 or 6 db  and see if that helps pick out  
the weak signal. If the signal is easier to pick out with the pad in 
then you are dynamic range limited. If the pad has no effect then 
dynamic range is not an issue. You have to do this with care because 
you will naturally favor a louder signal. When the guy at the stereo 
store switches to the expensive speakers he always cranks up the volume 
.....

Finally find a signal that sounds like distortion. If the distortion 
signs with KB8TQ it's just my normal signal ... Watch the S meter as 
you cut the pad in and out. A 10 db pad will knock in radio second or 
third order distortion down by significantly more than 10 db. If it 
only has 10 db of effect then the distortion is from that coax 
connector and not from the receiver. Or it may be the splatter from my 
200KHz wide signal ...

No need to fix what isn't broke .......

If it is broke then the question is weather you want to spend major 
bucks on a compromise multi band radio or build a cheap single band 
radio that will blow any multi band away. It's not clear to me that 
doing heavy mods on an R-4C is really any easier than building a simple 
radio from scratch ....

	Enjoy!

		Bob Camp
		KB8TQ




On Aug 15, 2004, at 12:21 PM, dmartin wrote:

> Been doing some reading lately on the concept of roofing filters and 
> their
> influence on near-in blocking dynamic range, etc., performance. The TT 
> Orion
> seems to be the first factory offering to take advantage of a variety 
> of
> various bandwidth front end roofing filters. Of course, Rob has long 
> offered
> his Sherwood 600 Hz roofing filter option for the 4C. Tom Rauch, W8JI, 
> has a
> great article in the current WorldRadio where he gets into receiver
> performance concepts. He pretty much disses the stock R-4C on close-in
> performance and concludes that a Sherwood 600 Hz roofing filter makes 
> the 4C
> acceptable, at best, on CW performance. However, my confusion over
> the "acceptable on CW" is this: if you check the performance tables on 
> Tom's
> website at http://www.w8ji.com/receiver_tests.htm, you'll note that 
> although
> the close-in performance of the stock 4C is as poor as several 
> references
> state, the "R-4C heavy mod" specs at least equal to >exponentially 
> exceed<
> anything else tested, including the Icom 7800 and current Orion. Along 
> with
> the addition of Sherwood's 600 Hz however, the "R-4C heavy mod"
> includes "solid state double balanced high level mixers". I'm quickly 
> aware
> only of Rob's IC third mixer mod. Anyone know any details on the 1st 
> and 2nd
> mixer mods on Tom's test 4C?
>
> Not considering such mods, mind you, just want to better understand 
> how such
> apparently superior performance, equal to and exceeding $3,000-$10,000
> radios, could have been squeezed out of what began life as a 4C? From 
> this I
> would conclude that an otherwise stock 4C but with Rob's 600 Hz roofing
> filter and possibly his 3rd mixer mod might make for a CW receiver very
> competitive with anything currently offered today, no?
>
> Dan
> WB4GRA
>
> --
>
> --
>
> ______________________________________________________________
> Drake mailing list
> Home: http://mailman.qth.net/mailman/listinfo/drake
> Help: http://mailman.qth.net/mmfaq.html
> Post: mailto:Drake at mailman.qth.net
>