[Elecraft] Hearing the effect of narrower roofing filter

Sun Aug 31 12:25:02 EDT 2008

Guy, 

> I need the roofing filter is to get down 30 db as quickly as 
> possible, without giving up too much of the +/- 250 Hz.

A careful comparison of the 500 Hz 5-pole filter, 400 Hz 8-pole 
filter, "250 Hz" 8 pole filter and 200 Hz 5 pole filter shows 
some interesting things ... 

1) my "500 Hz" filter is really 470 Hz at - 6dB ... this is 
   consistent (460 - 490 Hz) with measurements reported by 
   others.  I intend to check the two other 500 Hz filters 
   received with my second K3/KRX3 when I have some time. 

2) the -30 dB bandwidth of my 500 Hz filter (780 Hz) is not 
   significantly different than the -30 dB bandwidth of the 
   400 Hz filter reported on the Elecraft web site (680 Hz) 
   when cascaded with a DSP filter at 400 or 500 Hz.  

   I set my K3 so the "500 Hz" filter engages at 450 Hz.  It 
   is on-line when normalizing the CW filters but is off-line 
   when selecting a composite bandwidth wider than the filter. 

3) the 200 Hz five pole filter (209 Hz measured) is narrower 
   than the 250 Hz filter down to at least -40 dB without any 
   any benefit from cascading with DSP.  

For CW specifically, it's hard to justify the 400/250 Hz pair 
with their relatively small difference in bandwidth and their 
significantly greater cost compared to the 500/200 Hz pair. 

To be an effective alternative to the 400 Hz (450 Hz from 
Inrad curves, 435 Hz from Elecraft curves) filter, the 250 Hz 
8-pole really needs to be held to the nominal 250 Hz bandwidth 
so that the difference will be useful.  Even in a "worst case" 
(e.g. 40 meters in Europe) scenario, reducing the pre-DSP 
bandwidth from 450 to 370 Hz is not likely to be significant.  

> I have used both of these filters for years in my FT1000MP 
> and am completely familiar with their shape and use in a 
> contest. 

The 250 Hz filters in an FT1000MP is an entirely different 
situation.  In the FT-1000D/FT-1000MP, there are two filters 
- each approximately 350 Hz wide - in cascade yielding a 
composite 250 Hz (at -6dB) bandwidth.    

73, 

   ... Joe, W4TV 

> -----Original Message-----
> From: elecraft-bounces at mailman.qth.net 
> [mailto:elecraft-bounces at mailman.qth.net] On Behalf Of Guy, K2AV
> Sent: Sunday, August 31, 2008 10:46 AM
> To: elecraft at mailman.qth.net
> Subject: Re: [Elecraft] Hearing the effect of narrower roofing filter
> 
> 
> 
> 
> 
>     ..... Maybe
> I should have ordered the 200 Hz 5-pole instead, but I got it 
> into my head that maybe the 8-pole filters were "better" in 
> some way.  I didn't repeat this mistake when I ordered 
> filters for the subreceiver.
> 
> Dick, K6KR
> 
> I know I will incur some religious wrath from some.... 
> however here is the OTHER side of that argument.  It's not 
> that I am denying the issues that are raised to question 
> getting the two 8 pole CW filters (400 and 300 in actuality).
> 
> I am solving a SINGLE happenstance which occurs over, over, 
> over and over again in contests, so much so that until the K3 
> it was in my mind the LIMITING issue in improving 40m CW DX scores. 
> 
> I am at a contest station in eastern NC just off Pamlico 
> Sound. On 40m to Europe we are using a 5 element wire quad 
> suspended across a 220 foot NE/SW catenary between two towers. 
> 
> Broadcast signals above 7.1 routinely peg meters and light 
> all the lights. 
> So do some US stations in the NE, and so do some number of 
> VERY LOUD European stations (VLS).  WHAT they are doing to be 
> so loud is not part of my exposition. They just are very loud.
> 
> At some point in the contest, one has worked all the VLS, 
> loud, medium and well-antenna'd QRP signals. What remains is 
> a bewildering and seemingly bottomless pool of stations that 
> can hear us (QRO on 5 elements) and are trying to work us on 
> antennas with the gain of a basement floor joist wet noodle 
> antenna. There are hundreds and hundreds of these. We have 
> some number of recordings made in Europe of these, and they 
> are quite weak over there. 
> 
> It is guaranteed, only a matter of time, and sometimes 
> immediate, that a VLS will settle in the next slot above or 
> below. They may be as close as 350 Hz. The problem now is 
> working the QRP wet noodle station who is into the noise 
> without hardware AGC pumping or other effects from a VLS 
> INSIDE the roofing filter that was set to hear stations who 
> will call sometimes +/- 200-250 Hz.  
> 
> I need the roofing filter is to get down 30 db as quickly as 
> possible, without giving up too much of the +/- 250 Hz. The 
> 400 Hz is a good width even for VLS +/- 500 Hz.  If a VLS 
> squeezes me on one side, I only want to give up +/- real 
> estate in the roofing filter on that side. The procedure is 
> to drop to 300 hz roofing and DSP and move the RX center 50 
> hz away from TX frequency and the VLS. I give up listening so 
> far on the side of the VLS, but keep the real estate on the 
> other side.  
> 
> It is a matter of the width out to the edge and the largest 
> db drop per 10 hz in the skirts thereafter. The roofing 
> filter keeps the VLS from getting into the hardware AGC or 
> pushing the DSP to the extreme.  
> 
> I have used both of these filters for years in my FT1000MP 
> and am completely familiar with their shape and use in a contest. 
> 
> I have heard it said that the 200 5 pole filter will do 
> better for picking out signals in a very crowded situation, 
> perhaps so in an extremely tight situation. But thus far I 
> have been able to go narrow with the 300 8 pole and it is the 
> DSP handling the work there. 
> 
> At some point I would like to be able to tune the center of 
> the DSP CW selectivity up/down at 25 Hz rate using RIT 
> *WITHOUT* moving the position of the roofing filter relative 
> to the band.  The point of the roofing filter is to reject 
> the VLS in the next slot above and below me. Tune the whole 
> thing up or down to listen to a weak station off-frequency 
> for whatever reason and you let one of the VLS in under the roof.
> 
> Just the view from the other side of the river... :>)
> 
> 
> 73, Guy. 
> 
> -- 
> View this message in context: 
> http://n2.nabble.com/Hearing-the-effect-of-narrower-roofing-fi
lter-tp470635p795536.html
Sent from the Elecraft mailing list archive at Nabble.com.

_______________________________________________
Elecraft mailing list
Post to: Elecraft at mailman.qth.net
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft    

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com