Elecraft

Planning the next filters

Classic

List

Threaded

24 messages Options

Chuck Chandler

Planning the next filters

Right now I have the stock SSB and a 400 Hz filter in both main and sub
rcvrs. I would like to add some a few at a time. My first priority is to
add a couple to the main receiver, since it is the biggest chore to get
to.

I plan on a 250 Hz CW filter, but am going back and forth between the 2.1
and 1.8 kHz filters. I know that narrowing the DSP to 1.8 gives a really
difficult to understand voice signal, while the 2.1 allows in more adjacent
signals. However, the roofing filters may not behave in the same manner.
For example, the 1.8 could be set to turn on at 2.1 to take advantage of
skirt shape.

I primarily DX, and operate in some contests as well. I know SSB contests
really need a narrow SSB filter.

Any suggestions on the matter from those who have gone this route before?

73 de Chuck, WS1L

--

===================
Chuck Chandler
[hidden email]
===================
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

David Bunte

Re: Planning the next filters

Chuck -

You will likely get a wide variety of responses, and here is mine. I added
the 400 Hz filter to my main and sub receiver. I work CW almost
exclusively. I rag chew and chase some DX. I barely “dabble” in contests. I
can’t recall a single time when I thought a narrower CW filter would have
made a difference for me. It may help that my antenna is quite modest, and
I don’t have other hames near enough to impede my reception.

Best of luck as you proce d.

Dave - K9FN

On Thu, Mar 8, 2018 at 11:24 AM Chuck Chandler <[hidden email]>
wrote:

> Right now I have the stock SSB and a 400 Hz filter in both main and sub
> rcvrs. I would like to add some a few at a time. My first priority is to
> add a couple to the main receiver, since it is the biggest chore to get
> to.
>
> I plan on a 250 Hz CW filter, but am going back and forth between the 2.1
> and 1.8 kHz filters. I know that narrowing the DSP to 1.8 gives a really
> difficult to understand voice signal, while the 2.1 allows in more adjacent
> signals. However, the roofing filters may not behave in the same manner.
> For example, the 1.8 could be set to turn on at 2.1 to take advantage of
> skirt shape.
>
> I primarily DX, and operate in some contests as well. I know SSB contests
> really need a narrow SSB filter.
>
> Any suggestions on the matter from those who have gone this route before?
>
> 73 de Chuck, WS1L
>
> --
>
>
> ===================
> Chuck Chandler
> [hidden email]
> ===================
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Buck

Re: Planning the next filters

In reply to this post by Chuck Chandler

Get the 1.8 if you work ssb. Experiment with shifting the pass band for
maximum clarity. Youll be amazed just how narrow it can get and still be
understandable.

On Mar 8, 2018 11:24 AM, "Chuck Chandler" <[hidden email]> wrote:

Gary K9GS

Re: Planning the next filters

In reply to this post by David Bunte

Hi Chuck,
I agree with K9FN. I have the 250 Hz filters and rarely use them. The 400 Hz filters are great. So good that I would not miss the 250s at all.
On SSB I also have the stock filters. Here's how I handle crowded phone contests. I set the width to 2.0, sometimes 1.8. Then use the shift knob to clean up the signal. It is amazing how that works and most of the time the desired signal pops right out of the QRM.
I contest and chase DX a lot, 90% CW and favor the low bands too. I don't do digital modes but that would be the only other case where I'd consider a special bandwidth roofing filter.

73,
Gary K9GS
-------- Original message --------From: David Bunte <[hidden email]> Date: 3/8/18 10:45 AM (GMT-06:00) To: Chuck Chandler <[hidden email]> Cc: [hidden email] Subject: Re: [Elecraft] Planning the next filters
Chuck -

You will likely get a wide variety of responses, and here is mine. I added
the 400 Hz filter to my main and sub receiver. I work CW almost
exclusively. I rag chew and chase some DX. I barely “dabble” in contests. I
can’t recall a single time when I thought a narrower CW filter would have
made a difference for me. It may help that my antenna is quite modest, and
I don’t have other hames near enough to impede my reception.

Best of luck as you proce d.

Dave - K9FN

On Thu, Mar 8, 2018 at 11:24 AM Chuck Chandler <[hidden email]>
wrote:

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

donovanf

Re: Planning the next filters

In reply to this post by David Bunte

Hi Chuck,

The purpose of roofing filters in the K3 is commonly misunderstood,
they're primary function is to protect the receiver from overload from
an extremely strong interfering signal within the passband occupied
by the signal of interest. There's an excellent paper on this topic on
the Elecraft web site:

http://www.elecraft.com/K3/Roofing_Filters.htm

You can find the plots for many of the Elecraft filters here
(caution: many of the plots use different frequency spans):

http://www.elecraft.com/K3/K3_filter_plots.htm

There's very little difference in the performance of the Elecraft
400 Hz filters vs. their narrower bandwidth filters. The 400 Hz
filter you already have is all you will ever need in all but the most
extreme (and rare) interference situations.

T he 1 kHz filter is a useful addition to your receiver i f you like
broader bandwidth for casual CW operation, although the 2.1 kHz
filter works nearly as well if you narrow the DSP bandwidth.

There's very little difference between the 2.7 kHz stock filter and the
2.8 kHz 8 pole optional filter except in the most extreme interference
situations such SSB DX contests where its tighter skirts of the
2.8 kHz are sometimes be helpful but the difference is small.

I prefer the 2.1 kHz filter to the 1.8 kHz filter. I have both filters
in my K3 but rarely use the 1.8 kHz filter and only under the most
extreme inteference situations in a SSB DX contest.

73
Frank
W3LPL

On Thu, Mar 8, 2018 at 11:24 AM Chuck Chandler <[hidden email]>
wrote:

Joe Subich, W4TV-4

Re: Planning the next filters

In reply to this post by Chuck Chandler

> I plan on a 250 Hz CW filter,

I may be the odd man out but I can't see much enough difference between
the "400 Hz" (450 Hz @ -6dB) and the "250 Hz" (370 Hz @ -6dB) to justify
the cost. If you feel you *need* a narrower filter in the 8215 KHz IF,
consider the new KFL3C-200 which is truly 200 Hz wide at -6 dB.
See: <http://www.elecraft.com/K3/K3_filter_plots.htm> particularly the
last graph.

> but am going back and forth between the 2.1 and 1.8 kHz filters.

My choice - after trying the 1.5 KHz filters from INRAD and working with
both 2.1 KHz and 1.8 KHz DSP bandwidths (with center frequency offset to
maintain the low frequency cutoff at 100 - 250 Hz) was the 2.1 KHz
filter. I felt the DSP provided sufficient high frequency roll off
if/when it was needed but the 2.1 KHz filter provided more flexibility -
and I ended up using 2.4 KHz - with the 2.1 KHz "roofing" filter set to
kick in at 2.4 KHz - as my "normal" SSB bandwidth.

While amateur activity is not channelized, most day-to day SSB activity
seems to fall on 3 KHz boundaries with occasional 2.5 KHz "steps". Even
in a busy contest, things don't tend to get closer than 2 KHz as long as
the stations can "hear". Most of the QRM on SSB isn't necessarily from
overlapping signals ... it's from transmit phase noise and IMD. That's
not something that *any* receiver filter can deal with as those are
"spurious" signals outside the "necessary bandwidth" generated and
radiated by the transmitting (interfering) station.

73,

... Joe, W4TV

On 3/8/2018 11:22 AM, Chuck Chandler wrote:

Dave Hachadorian-2

Re: Planning the next filters

In reply to this post by Chuck Chandler

I think you are optimally configured right now. I used to have
the 1.8 filters in my K3's, but sold them. I think they hurt
intelligibility more than they helped with QRM rejection.
Remember that these are only roofing filters. The DSP filter has
a higher effective shape factor, especially on SSB. Elecraft
once said in an email that you could estimate the DSP shape
factor by the formula (DSP BW SETTING+ 300 Hz)/(DSP BW SETTING).
Especially for the SSB filters, the DSP shape factor is the
determinant in bandwidth, as long as there is no signal just
outside the DSP width that is causing the AGC to pump, or the A/D
to overload. Those effects, in the very rare event that they
occur, can be minimized by judicious use of the RF Gain control.

As for the 250 Hz CW filter, If you look at the filter curves
http://www.elecraft.com/K3/K3_filter_plots.htm you will see that
this filter should be really labeled KFL3A-370, and is not much
different than your 400.

Dave Hachadorian, K6LL
Yuma, AZ

-----Original Message-----
From: Chuck Chandler
Sent: Thursday, March 8, 2018 9:22 AM
To: [hidden email]
Subject: [Elecraft] Planning the next filters

Right now I have the stock SSB and a 400 Hz filter in both main
and sub
rcvrs. I would like to add some a few at a time. My first
priority is to
add a couple to the main receiver, since it is the biggest chore
to get
to.

I plan on a 250 Hz CW filter, but am going back and forth between
the 2.1
and 1.8 kHz filters. I know that narrowing the DSP to 1.8 gives
a really
difficult to understand voice signal, while the 2.1 allows in
more adjacent
signals. However, the roofing filters may not behave in the same
manner.
For example, the 1.8 could be set to turn on at 2.1 to take
advantage of
skirt shape.

I primarily DX, and operate in some contests as well. I know SSB
contests
really need a narrow SSB filter.

Any suggestions on the matter from those who have gone this route
before?

73 de Chuck, WS1L

--

===================
Chuck Chandler
[hidden email]
===================
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list:
http://www.qsl.net/donate.html
Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Jim Brown-10

Re: Planning the next filters

In reply to this post by Chuck Chandler

It's important to understand that the variable DSP IF is plenty good
enough for most operation. The function of the roofing filter is to
protect the DSP from overload by strong signals outside the passband. I
would consider buying additional filters ONLY if you're a serious
contester. I am, and my favorites are 250 and 400 for CW, 2.1 for SSB,
and I've replaced the 2.7 with 2.8 for the improved flatness it
provides, which reduces incidental AM when transmitting RTTY. I agree
that 1.8 is too narrow for my ears.

73, Jim K9YC

On 3/8/2018 8:22 AM, Chuck Chandler wrote:

Wes Stewart-2

Re: Planning the next filters

In reply to this post by Chuck Chandler

I would replace the 2.7 with the 2.8, keep the 400 and be done with it. I did.

The purpose of the "roofing filters" (I hate that term) is to protect the 2nd
mixer from overload. The wider filter is used on TX as well and the 2.8 has
superior symmetry which helps opposite sideband rejection.

Most of the garbage you will hear, particularly on SSB is from the lousy phase
noise and TX IMD (K3s include here) from adjacent signals falling in your RX
passband. No amount of filtering on your end will fix this.

Wes N7WS

On 3/8/2018 9:22 AM, Chuck Chandler wrote:

Barry Simpson

Re: Planning the next filters

In reply to this post by Buck

Hi Craig

I would go for the 200Hz filters, not the 250Hz. I added the 200Hz to my
main and sub receivers (already have the 400Hz) and they are a really
worthwhile addition.

That's my opinion FWIW.

73

Barry VK2BJ

On 9 March 2018 at 03:57, Craig Buck <[hidden email]> wrote:

> Get the 1.8 if you work ssb. Experiment with shifting the pass band for
> maximum clarity. Youll be amazed just how narrow it can get and still be
> understandable.
>
> On Mar 8, 2018 11:24 AM, "Chuck Chandler" <[hidden email]> wrote:
>
> > Right now I have the stock SSB and a 400 Hz filter in both main and sub
> > rcvrs. I would like to add some a few at a time. My first priority is
> to
> > add a couple to the main receiver, since it is the biggest chore to get
> > to.
> >
> > I plan on a 250 Hz CW filter, but am going back and forth between the 2.1
> > and 1.8 kHz filters. I know that narrowing the DSP to 1.8 gives a really
> > difficult to understand voice signal, while the 2.1 allows in more
> adjacent
> > signals. However, the roofing filters may not behave in the same manner.
> > For example, the 1.8 could be set to turn on at 2.1 to take advantage of
> > skirt shape.
> >
> > I primarily DX, and operate in some contests as well. I know SSB
> contests
> > really need a narrow SSB filter.
> >
> > Any suggestions on the matter from those who have gone this route before?
> >
> > 73 de Chuck, WS1L
> >
> > --
> >
> >
> > ===================
> > Chuck Chandler
> > [hidden email]
> > ===================
> > ______________________________________________________________
> > Elecraft mailing list
> > Home: http://mailman.qth.net/mailman/listinfo/elecraft
> > Help: http://mailman.qth.net/mmfaq.htm
> > Post: mailto:[hidden email]
> >
> > This list hosted by: http://www.qsl.net
> > Please help support this email list: http://www.qsl.net/donate.html
> > Message delivered to [hidden email]
> >
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Bill Frantz

Re: Planning the next filters

In reply to this post by Chuck Chandler

I am a all-band, all-mode operator. I contest, chase DX, and
just make contacts.

I have something close to what you are considering. I have 250,
2.1K, 2.7K, and 13K filters in both my main and sub receivers. I
got the 250 for eliminating loud PSK signals from wiping out
weak ones. It was a bunch of playing with center and width, but
worked quite well.

I also use the 250 on CW, sometimes narrowing the DSP bandwidth
to 50 Hz to minimize noise, or nearby QRM when listening to a DX
station. In a DX chase, hopefully the DX will have a bit of
spectrum all to its own, but with DQRM and people tuning up on
the DX frequency, narrow bandwidth can be very useful. If the DX
frequency is clear, the Audio Peaking Filter can work better
than the 50 Hz DSP bandwidth in reducing noise.

Joe is making me tempted to replace the 250s with 200s -- I've
had problems in contests with nearby strong stations. Jim makes
an good argument for the 2.8K ones for cleaner digital transmission.

As you may have noticed, the filter configuration is a highly
individual choice. With the DSP as good as it is, you should
have fine performance whichever way you choose.

73 Bill AE6JV

On 3/8/18 at 8:22 AM, [hidden email] (Chuck Chandler) wrote:

>I plan on a 250 Hz CW filter, but am going back and forth between the 2.1
>and 1.8 kHz filters. I know that narrowing the DSP to 1.8 gives a really
>difficult to understand voice signal, while the 2.1 allows in more adjacent
>signals. However, the roofing filters may not behave in the same manner.
>For example, the 1.8 could be set to turn on at 2.1 to take advantage of
>skirt shape.
>
>I primarily DX, and operate in some contests as well. I know SSB contests
>really need a narrow SSB filter.
>
>Any suggestions on the matter from those who have gone this route before?
>
>73 de Chuck, WS1L
>

On 3/8/18 at 9:32 AM, [hidden email] (Joe Subich, W4TV) wrote:

>... If you feel you *need* a narrower filter in the 8215 KHz IF,
>consider the new KFL3C-200 which is truly 200 Hz wide at -6 dB.
>See: <http://www.elecraft.com/K3/K3_filter_plots.htm> particularly the
>last graph.

-------------------------------------------------------------------------------------
Bill Frantz | Government is not reason, it is not
eloquence, it is force; like
408-356-8506 | a fire, a troublesome servant and a fearful
master. Never for a
www.pwpconsult.com | moment should it be left to irresponsible
action. Geo Washington

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Jim Brown-10

Re: Planning the next filters

On 3/8/2018 1:08 PM, Bill Frantz wrote:
> I got the 250 for eliminating loud PSK signals from wiping out weak
> ones. It was a bunch of playing with center and width, but worked
> quite well.

Hi Bill,

Recall my post from a few days ago about phase shift being the enemy of
signal decoders, and that filters built with real components (and most
digital domain filters) create phase shift in the passband. The steeper
the skirts, the greater the phase shift, and the more it extends into
the passband.

I don't work PSK31 and don't know about PSK31 decoders, but the authors
of other decoding software (WSJT-X, 2Tone) recommend that radios use
wide bandwidth and let the decoder reject off-frequency QRM. Heavy duty
RTTY guys like W0YK recommend 500 Hz for RTTY, and K1JT recommends the
full bandwidth of SSB channel. I follow their advice, and find it works
well for me.

73, Jim K9YC

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Bill Frantz

Re: Planning the next filters

I understand that advice, and follow it most of the time.
However, I found that when I was trying to receive weak PSK31
signals, strong, nearby signals would make decode impossible.
Using a combination of the 250 Hz filter with the DSP filter,
sometimes as narrow as 50 Hz would allow decode. I think the
problem was that the combination of the strong signal and the
weak signal exceeded the dynamic range of the audio AtoD
converter going into the computer, and the weak signal lost out.
I could be wrong.

This problem occurred with a SignaLink. I have not had as much
of a problem with the KIO3B running FT8, but the circumstances
are different enough between the two modes that I'm not sure
that is an apples to apples comparison. I was able to improve
FT8 decode through the KIO3B when there were strong signals 2-3
KHz away from the FT8 subband by narrowing the DSP bandwidth
from 4K to 2.5K (which automatically switched from the 13 KHz FM
filter to the stock 2.7 KHz filter).

73 Bill AE6JV

On 3/8/18 at 8:38 PM, [hidden email] (Jim Brown) wrote:

>On 3/8/2018 1:08 PM, Bill Frantz wrote:
>>I got the 250 for eliminating loud PSK signals from wiping out
>>weak ones. It was a bunch of playing with center and width,
>>but worked quite well.
>
>Hi Bill,
>
>Recall my post from a few days ago about phase shift being the
>enemy of signal decoders, and that filters built with real
>components (and most digital domain filters) create phase shift
>in the passband. The steeper the skirts, the greater the phase
>shift, and the more it extends into the passband.
>
>I don't work PSK31 and don't know about PSK31 decoders, but the
>authors of other decoding software (WSJT-X, 2Tone) recommend
>that radios use wide bandwidth and let the decoder reject
>off-frequency QRM. Heavy duty RTTY guys like W0YK recommend 500
>Hz for RTTY, and K1JT recommends the full bandwidth of SSB
>channel. I follow their advice, and find it works well for me.
>
>73, Jim K9YC

-----------------------------------------------------------------------
Bill Frantz | Can't fix stupid, but duct | Periwinkle
(408)356-8506 | tape can muffle the sound... | 16345
Englewood Ave
www.pwpconsult.com | - Bill Liebman | Los Gatos,
CA 95032

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Jim Brown-10

Re: Planning the next filters

On 3/8/2018 10:18 PM, Bill Frantz wrote:
> This problem occurred with a SignaLink.

That's not a very good interface. It helps to have an clip light or
other digital level indicator on the interface for the received signal.
My Tascam US100 has a green signal presence light and a red one for
clip, which makes it easy to stay out of trouble.

73, Jim K9YC

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Raymond Sills

Re: Planning the next filters

And.... isn't it the case that the SignaLink is a 16 bit device? That Tascam US 100 is also 16 bits, but it has been replaced by a model that has 24 bit/96 KHz sampling.. indicating a high quality interface.

73 de Ray
K2ULR
KX3 #211

-----Original Message-----
From: Jim Brown <[hidden email]>
To: elecraft <[hidden email]>
Sent: Fri, Mar 9, 2018 1:41 am
Subject: Re: [Elecraft] Planning the next filters

On 3/8/2018 10:18 PM, Bill Frantz wrote:
> This problem occurred with a SignaLink.

That's not a very good interface. It helps to have an clip light or
other digital level indicator on the interface for the received signal.
My Tascam US100 has a green signal presence light and a red one for
clip, which makes it easy to stay out of trouble.

73, Jim K9YC

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Johan Ymerson

Re: Planning the next filters

Yes, the Signalink USB is 16-bit from what I remember. But on the other hand
alost all rigs have a 16-bit audio DAC so that will be the limiting factor
anyway (with K3 being the exception with a 24-bit DAC).

Probably the biggest problem with the Signalink USB is not the interface
itself, but the Windows builtin drivers for the ADC. It thinks the ADC is a
microphone and add +30 dB of gain in software. So those 16 bit's are quickly
reduced to 6 bits if you are not careful. There are some descriptions (it
isn't obvious, it is not visible in the normal mixer controls) on the net how
to set the gain to (almost) 0 dB, but I use Linux so I haven't tried.

/Johan

On Friday, 9 March 2018 15:18:09 CET Raymond Sills wrote:

> And.... isn't it the case that the SignaLink is a 16 bit device? That
> Tascam US 100 is also 16 bits, but it has been replaced by a model that has
> 24 bit/96 KHz sampling.. indicating a high quality interface.
>
> 73 de Ray
> K2ULR
> KX3 #211
>
>
>
> -----Original Message-----
> From: Jim Brown <[hidden email]>
> To: elecraft <[hidden email]>
> Sent: Fri, Mar 9, 2018 1:41 am
> Subject: Re: [Elecraft] Planning the next filters
>
> On 3/8/2018 10:18 PM, Bill Frantz wrote:
> > This problem occurred with a SignaLink.
>
> That's not a very good interface. It helps to have an clip light or
> other digital level indicator on the interface for the received signal.
> My Tascam US100 has a green signal presence light and a red one for
> clip, which makes it easy to stay out of trouble.
>
> 73, Jim K9YC
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]

Joe Subich, W4TV-4

Re: Planning the next filters

> Probably the biggest problem with the Signalink USB is not the
> interface itself, but the Windows builtin drivers for the ADC. It
> thinks the ADC is a microphone and add +30 dB of gain in software.
That issue is not with Windows ... it is with the device maker who
has selected a device that identifies as microphone level. Using
the proper device/device ID will cause Windows to see it as a line
level input and set the drivers correctly.

73,

... Joe, W4TV

On 3/9/2018 10:05 AM, Johan Ymerson wrote:

> Yes, the Signalink USB is 16-bit from what I remember. But on the other hand
> alost all rigs have a 16-bit audio DAC so that will be the limiting factor
> anyway (with K3 being the exception with a 24-bit DAC).
>
> Probably the biggest problem with the Signalink USB is not the interface
> itself, but the Windows builtin drivers for the ADC. It thinks the ADC is a
> microphone and add +30 dB of gain in software. So those 16 bit's are quickly
> reduced to 6 bits if you are not careful. There are some descriptions (it
> isn't obvious, it is not visible in the normal mixer controls) on the net how
> to set the gain to (almost) 0 dB, but I use Linux so I haven't tried.
>
> /Johan
>
> On Friday, 9 March 2018 15:18:09 CET Raymond Sills wrote:
>> And.... isn't it the case that the SignaLink is a 16 bit device? That
>> Tascam US 100 is also 16 bits, but it has been replaced by a model that has
>> 24 bit/96 KHz sampling.. indicating a high quality interface.
>>
>> 73 de Ray
>> K2ULR
>> KX3 #211
>>
>>
>>
>> -----Original Message-----
>> From: Jim Brown <[hidden email]>
>> To: elecraft <[hidden email]>
>> Sent: Fri, Mar 9, 2018 1:41 am
>> Subject: Re: [Elecraft] Planning the next filters
>>
>> On 3/8/2018 10:18 PM, Bill Frantz wrote:
>>> This problem occurred with a SignaLink.
>>
>> That's not a very good interface. It helps to have an clip light or
>> other digital level indicator on the interface for the received signal.
>> My Tascam US100 has a green signal presence light and a red one for
>> clip, which makes it easy to stay out of trouble.
>>
>> 73, Jim K9YC
>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Johan Ymerson

Re: Planning the next filters

On Friday, 9 March 2018 16:29:19 CET Joe Subich, W4TV wrote:
> > Probably the biggest problem with the Signalink USB is not the
> > interface itself, but the Windows builtin drivers for the ADC. It
> > thinks the ADC is a microphone and add +30 dB of gain in software.
>
> That issue is not with Windows ... it is with the device maker who
> has selected a device that identifies as microphone level. Using
> the proper device/device ID will cause Windows to see it as a line
> level input and set the drivers correctly.
>

That can be debated, no other OS (MacOS/Linux/*BSD) identifies it as a
microphone. And why is there no clear option to enable/disable the "software
pre-amp" in Windows?

On the filter discussion; I have (apart from FM filter) 2.8 KHz filters and
400 Hz filters. Very pleased with that. I am not a contester so I want the
full SSB bandwidth, mainly for waterfall view in digital modes. 2.8 KHz
because I couldn't be bothered with filter matching in main and sub
receiver...
400 Hz is kind of the smallest BW usable for digital modes anyway. If you want
to go tighter than that (for CW or possibly PSK31), use the DSP filter is my
advice.

/Johan
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Joe Subich, W4TV-4

Re: USB Audio CODEC (Was: Planning the next filters)

On 3/9/2018 11:00 AM, Johan Ymerson wrote:
> And why is there no clear option to enable/disable the "software
> pre-amp" in Windows?

With the current drivers in Windows 10, disabling the preamp is
automatic. One configures the level control in the Sound Control
Panel (Recording) to display in dB ... any setting above 0 dB
enables the preamp, any setting below 0 dB disables the preamp.

I assume the change had propagated to Windows 8.1 before it ended
mainstream support in January but I would not expect it in Windows
XP, NT, 2000 or any other version of Windows that has been out of
"mainstream" support for several years (obsolete).

73,

... Joe, W4TV

On 3/9/2018 11:00 AM, Johan Ymerson wrote:

> On Friday, 9 March 2018 16:29:19 CET Joe Subich, W4TV wrote:
>>> Probably the biggest problem with the Signalink USB is not the
>>> interface itself, but the Windows builtin drivers for the ADC. It
>>> thinks the ADC is a microphone and add +30 dB of gain in software.
>>
>> That issue is not with Windows ... it is with the device maker who
>> has selected a device that identifies as microphone level. Using
>> the proper device/device ID will cause Windows to see it as a line
>> level input and set the drivers correctly.
>>
>
> That can be debated, no other OS (MacOS/Linux/*BSD) identifies it as a
> microphone. And why is there no clear option to enable/disable the "software
> pre-amp" in Windows?
>
> On the filter discussion; I have (apart from FM filter) 2.8 KHz filters and
> 400 Hz filters. Very pleased with that. I am not a contester so I want the
> full SSB bandwidth, mainly for waterfall view in digital modes. 2.8 KHz
> because I couldn't be bothered with filter matching in main and sub
> receiver...
> 400 Hz is kind of the smallest BW usable for digital modes anyway. If you want
> to go tighter than that (for CW or possibly PSK31), use the DSP filter is my
> advice.
>
> /Johan
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Jim Brown-10

Re: Planning the next filters

In reply to this post by Raymond Sills

Bits is not the issue with interfaces for digital modes -- encode/decode
uses the 48 kHz 16 bit stream, and K1JT says that's plenty good enough.
What matters is the quality of the A/D and the analog electronics, and
that's where the Signalink falls down. Higher bit rates and depths DO
matter when using an interface for a spectrum display. N8LP has long
recommended the ASUS Xonar U5 and U7 for use with his LP-Pan, although
I've heard of hardware failures with the U7.

73, Jim K9YC

On 3/9/2018 6:18 AM, Raymond Sills wrote:
> And.... isn't it the case that the SignaLink is a 16 bit device? That Tascam US 100 is also 16 bits, but it has been replaced by a model that has 24 bit/96 KHz sampling.. indicating a high quality interface.

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]