Wrong Filter Question - 8 pole vs. 5 Pole Filters

I asked the wrong filter question...or...what was on my mind and how it came out my fingers did not jive.

Which is better a 5 pole filter or an 8 pole filter.  Supposedly the 8 pole would have sharper skirts....and for only 20 more bucks...well....I would like to know the difference between the two filters.  Is it worth the 20 bucks?  I am sure it is, but as you all know there is some economic rumbling and need to watch the pennies.

This is not about cost....it is about what is the difference between the two types of filters.

Lee - k0WA


In our day and age it seems that Common Sense is in short supply.  If you don't have any Common Sense - get some Common Sense and use it.  If you can't find any Common Sense, ask for help from somebody who has some Common Sense.  Is Common Sense divine?
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
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

[Bill W4ZV]

Lee Buller wrote

Which is better a 5 pole filter or an 8 pole filter.  Supposedly the 8 pole would have sharper skirts....and for only 20 more bucks...well....I would like to know the difference between the two filters.  Is it worth the 20 bucks?  I am sure it is, but as you all know there is some economic rumbling and need to watch the pennies.

This question seems to be asked at least once per week...maybe we should learn how to use search engines!

http://groups.yahoo.com/group/Elecraft_K3/message/901 (earlier today)

73,  Bill

Re: K3 CW Filters Performance

--- In [e-mail], "John Rogers" <johnr1g3pqa@...> wrote:

> Has anyone compared the similar bandwidth 5 pole and 8 pole Ist IF
filters for weak CW readability under normal and noisy conditions?
> ie can anyone comment on practical difference between the 500 and
400, and 200 and 250Hz filters for this usage?

Hi John,

Remember that in a rig like the K3 or Orion, roofing filters do NOT
determine ultimate selectivity. Their purpose is mainly to prevent
strong signals from entering the IF chain and causing IMD or BDR
problems. W2VJN of Inrad says the main area of interest for roofing
filters is only about -30 dB down on the skirts from the center
frequency, so there is really not a significant difference between
5-pole and 8-pole (remember that all of Orion's filters are 4-pole).
In other words shape factor and ultimate selectivity are not critical
for roofing filters.

What you are really asking is what the DSP filters look like, which
actually determine ultimate selectivity. I have not seen any
measurements of these yet but I'm sure someone will eventually post
measurements of them.

Here's a simple block diagram to remember:

Antenna > 1st IF roofing filters > 2nd IF DSP filters
Roofing filters prevent IMD & BDR from entering IF chain
DSP filters provide ultimate selectivity

More info on roofing filters here:
http://www.zerobeat.net/mediawiki/index.php/K3_Roofing_Filters

5-pole plots by OH9NB here:
http://groups.yahoo.com/group/Elecraft_K3/files/

8-pole plots by Inrad here:
http://www.elecraft.com/K3/K3_8_pole_plots.htm

73, Bill W4ZV

[drewko]

I'm guessing that I wouldn't be able to tell the difference, like
monster cable for the stereo. Will there ever be an instance when I
will say, "Boy, if only i had an 8 pole filter; I might have snagged
that rare DX station!" I doubt it.

Actually, if it had been a choice between 8 and 5 in a 500 Hz filter,
I might have chosen the 8. But I don't want a 400 Hz filter however
many poles it has; I want 500 Hz. This has to do with the fact that in
a hypershphere of very high dimensionality nearly all of the volume
lies very close to the surface. Just kidding. The filter graphs mean
next to nothing to me.

I guess when it comes time to sell the rig the 8-pole guys will have
an advantage over us filter ignoramuses and cheapskates. Anyhow, the
price difference between the is now $40 for me according to the order
summary.

73,
Drew
AF2Z



On Mon, 04 Feb 2008 13:29:25 -0800 (PST), you wrote:

>
>
>I asked the wrong filter question...or...what was on my mind and how it came out my fingers did not jive.
>
>Which is better a 5 pole filter or an 8 pole filter.  Supposedly the 8 pole would have sharper skirts....and for only 20 more bucks...well....I would like to know the difference between the two filters.  Is it worth the 20 bucks?  I am sure it is, but as you all know there is some economic rumbling and need to watch the pennies.
>
>This is not about cost....it is about what is the difference between the two types of filters.
>
>Lee - k0WA
>
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
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

[David Woolley (E.L)]

Bill W4ZV wrote:
>

> What you are really asking is what the DSP filters look like, which
> actually determine ultimate selectivity. I have not seen any
> measurements of these yet but I'm sure someone will eventually post
> measurements of them.

DSP filter responses can be calculated exactly, they don't actually have
to be measured.

>
=
--
David Woolley
"The Elecraft list is a forum for the discussion of topics related to
Elecraft products and more general topics related ham radio"
List Guidelines <http://www.elecraft.com/elecraft_list_guidelines.htm>
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
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

[dj7mgq]

> DSP filter responses can be calculated exactly, they don't actually
> have to be measured.

As long as we don't have exact and detailed information about the  
implementation of the filters in the K3, the responses curves can not  
be calculated by third parties.

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
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

[Bill W4ZV]

Toby Deinhardt wrote

> DSP filter responses can be calculated exactly, they don't actually
> have to be measured.

As long as we don't have exact and detailed information about the  
implementation of the filters in the K3, the responses curves cannot  
be calculated by third parties.

Correct.  Even within the K3 there are two choices for calculation of the DSP filters...Finite Impulse Response (FIR) or Infinite Impulse Response (IIR).  These can be toggled by pressing "7" when you are in any CONFIG filter menu.  I forget the exact range of applicable bandwidths but believe both are available at the narrowest WIDTH settings (~150 Hz and below).

73,  Bill

[Bill W4ZV]

Bill W4ZV wrote

Toby Deinhardt wrote

> DSP filter responses can be calculated exactly, they don't actually
> have to be measured.

As long as we don't have exact and detailed information about the  
implementation of the filters in the K3, the responses curves cannot  
be calculated by third parties.

Correct.  Even within the K3 there are two choices for calculation of the DSP filters...Finite Impulse Response (FIR) or Infinite Impulse Response (IIR).  These can be toggled by pressing "7" when you are in any CONFIG filter menu.  I forget the exact range of applicable bandwidths but believe both are available at the narrowest WIDTH settings (~150 Hz and below).

73,  Bill

Actually only for the 100 and 50 Hz filters...from the Elecraft Software Page:
http://www.elecraft.com/K3/k3_software.htm

In Our Previous Release  -   MCU 1.58  and  DSP 1.48

    *

      NEW 100- AND 50-Hz DSP FILTERS: We've completed the addition of high-performance 100- and 50-Hz "FIR" (finite impulse response) DSP filters. These are flatter than the  original "IIR" (infinite impulse response) filters, and exhibit less ringing, which is especially noticeable in the presence of high noise. Some operators may prefer the original filters,
      which have a slightly different sound. To switch between IIR (old) and FIR (new), tap '7'  on the keypad when you're in any filter setup menu entry (e.g., CONFIG:FLx BW).
      To use the FIR (new) filters, tap '7' until you see "IIR OFF". This is now the default for newly-shipped K3s.

[Julian, G4ILO]

Bill W4ZV wrote

Lee Buller wrote

Which is better a 5 pole filter or an 8 pole filter.  Supposedly the 8 pole would have sharper skirts....and for only 20 more bucks...well....I would like to know the difference between the two filters.  Is it worth the 20 bucks?  I am sure it is, but as you all know there is some economic rumbling and need to watch the pennies.

This question seems to be asked at least once per week...maybe we should learn how to use search engines!

http://groups.yahoo.com/group/Elecraft_K3/message/901 (earlier today)

That message was not posted to this list, and it is unavailable to non members of Yahoo.

[Bill W4ZV]

G4ILO wrote

That message was not posted to this list, and it is unavailable to non members of Yahoo.


http://www.nabble.com/forum/Search.jtp?forum=28632&local=y&query=8-pole+vs+5-pole

"Found 18 matching posts for 8-pole vs 5-pole in Elecraft in about 11 threads."  

That's on this list only...going all the way back to May 3, 2007.  Many questions could be answered if we first tried search.  

73,  Bill