2.10B narrow filter anomaly

[Gill W4RYW]

I have been using ver 1.78 that shipped with rig and been quite pleased. Figured I should update.When tuned to a steady tone (CW too!) I now get a beat note effect on the width setting 100hz and 50hz only.  I have a 250hz filter. Turn on IIR fixes problem. Only hear on FIR DSP. Sounds like two identical freq tones beating - tuning the sig to a precise freq causes effect to disappear. This is annoying to listen to.

SN#707

Gill W4RYW

[Bob Cunnings NW8L]

You may be on to something. Using the 100 Hz FIR filter and injecting
a steady carrier, I looked at the receiver output using Spectrum Lab.
I observed multiple spurious tones, 2 of which are only 15 dB down
from the desired signal. The freq spread between the desired signal
and the 1st spur is 2x the difference between the signal and filter
center, so you have zero beat at filter center. And so on.  Take a
look at it with an audio spectrum analyzer, it's interesting. Zero
beat with the other -15 dB spur is at the upper filter skirt.
Aliasing?

The result is the same for the 50 Hz filter, but the 50 and 100 Hz IIR
filters are squeaky clean under the same conditions.

Subjectively speaking, when listening to on air signals, the signal
starts to sound bad to my ears as you tune it close to the upper
filter edge zero beat point, but within 20 Hz of filter center the
effect seems subtle, and easily overlooked. However, viewed on the
spec an, it's clear that something isn't quite right.

I observe no such problem with ver 1.78 FIR filters under the same conditions.

Maybe someone else can take a look at this.

Bob NW8L

On Sat, Jun 28, 2008 at 8:28 AM, Gill W4RYW <[hidden email]> wrote:

>
> I have been using ver 1.78 that shipped with rig and been quite pleased.
> Figured I should update.When tuned to a steady tone (CW too!) I now get a
> beat note effect on the width setting 100hz and 50hz only.  I have a 250hz
> filter. Turn on IIR fixes problem. Only hear on FIR DSP. Sounds like two
> identical freq tones beating - tuning the sig to a precise freq causes
> effect to disappear. This is annoying to listen to.
>
> SN#707
>
> Gill W4RYW
> --
> View this message in context: http://www.nabble.com/2.10B-narrow-filter-anomaly-tp18171265p18171265.html
> Sent from the Elecraft mailing list archive at Nabble.com.
>
> _______________________________________________
> 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
>

_______________________________________________
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

[Gill W4RYW]

Bob,

Thanks for the confirmation. I have spectrum lab - will take a look after FD. I also think the level drops several db (to the ear) as the filter switches from 150hz to 100hz, most noticable on weaker sigs.

73's

Gill

<quote author="Bob Cunnings">
You may be on to something. Using the 100 Hz FIR filter and injecting
a steady carrier, I looked at the receiver output using Spectrum Lab.
I observed multiple spurious tones, 2 of which are only 15 dB down
from the desired signal. The freq spread between the desired signal
and the 1st spur is 2x the difference between the signal and filter
center, so you have zero beat at filter center. And so on.  Take a
look at it with an audio spectrum analyzer, it's interesting. Zero
beat with the other -15 dB spur is at the upper filter skirt.
Aliasing?

The result is the same for the 50 Hz filter, but the 50 and 100 Hz IIR
filters are squeaky clean under the same conditions.

Subjectively speaking, when listening to on air signals, the signal
starts to sound bad to my ears as you tune it close to the upper
filter edge zero beat point, but within 20 Hz of filter center the
effect seems subtle, and easily overlooked. However, viewed on the
spec an, it's clear that something isn't quite right.

I observe no such problem with ver 1.78 FIR filters under the same conditions.

Maybe someone else can take a look at this.

Bob NW8L

<quote>