My Dreams Of The Future - Filtering

  • 9
  • Idea
  • Updated 5 years ago
  • Under Consideration
  • (Edited)
Hey beta testers, what a privilege to be a part of the evolving 6000 series radio development.. I have now used my Flex 6700 long enough to realize what tremendous (yes, tremendous) potential this radio has. When I bought this radio, I bought it with all the available "bells & whistles" knowing that someday this would be an awesome thing!. Having started as a ham back in 1952 with my S-38 radio from Sears, look where we have come from since then. Can you see where we are going? Well, Gerald and the boys at Flex certainly have that future vision in their mind's eye. Lets help em out with that vision.

Now, what should the future 6000 look like? I have a short list my own dreams here:
1. Smooth (yes I mean smooth) QSK at 30 plus wpm.
2. For the weak signal cw DX guy like me, an audio peaking filter with 5 hz or less pass band, and extremely steep skirts (ringing be damned). This peaking filter would be selectable by quick menu click or Flex Control pushbutton, with its pass band frequency (and perhaps, width as well) adjustable by the Flex Control knob. My K3 has this, and I use it a lot.
3.CW filter skirts with much steeper skirts than those which presently exist under PowerSDR, particularly for the more narrow filters (which, I am sure is much more difficult to develope).. the 50 hz filter skirts on PowerSDR was never that good.

My list will grow as time passes, and I'm sure you all have your own wants for this future radio. Lets hear them.

Dave, W5UN
Photo of W5UN_Dave

W5UN_Dave

  • 315 Posts
  • 30 Reply Likes

Posted 6 years ago

  • 9
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9153 Posts
  • 3503 Reply Likes
Dan,

#1 and #3 are in progress.

I have added #2 to the feature request database. This is an interesting idea. Thanks for sharing it.
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1798 Posts
  • 627 Reply Likes
Dave, from another weak signal CW DXer, I also like the audio peaking filter (#2 from above). The FTDX9000 has an APF feature that allows the user to select the amount of peaking (a menu setting) and a knob to adjust the frequency to peak. It works well but the 6xoo could do it much better and give the user more control (with the ability to control the width)l.

This is one of the best tools for making a weak CW signal pop.

Regards, Al / NN4ZZ
Photo of W5UN_Dave

W5UN_Dave

  • 315 Posts
  • 30 Reply Likes
Al, without the AFP I would have missed half of the low power DXpeditions on 160 meters. It is an absolute necessity here. My K3 has a good one, so did my previous Yaesu FT1000D. My Flex 5000 has no such capability, despite my repeated pestering Gerald to provide this capability. If it is not implemented (and soon) on the 6700, this radio will not be usable by me on 160 meters this coming season for serious DXing.
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1798 Posts
  • 627 Reply Likes
Dave, agree AFP is a great feature. On the Yaesu FTDX9000 it was added as one of the PEP enhancements in 2009. . I was one of the field testers and wrote it up on my web page.

http://www.nn4zz.com/PEP.htm#Contour_...

I'm sure the 6700 can do it even better, and give us more control of the AFP parameters.

Regards, Al / NN4ZZ
Photo of K1UO - Larry

K1UO - Larry

  • 823 Posts
  • 132 Reply Likes
Tim,
don't let the APF idea die internally! this is an idea whose time has come... especially one with the ability for user adjustments. As another weak signal Lowband DXer you can count me in as another vote for this feature.

Regards,
Larry K1UO
Photo of Steve - N5AC

Steve - N5AC, VP Engineering / CTO

  • 1046 Posts
  • 1025 Reply Likes
Dave / Larry,

Do you perceive any difference in adding an APF versus an RF filter that can be very narrow? I ask both from a user interface perspective and from a technology perspective. The APF can be implemented as an RF filter (baseband), but presented with a different user interface or it could be a different filter (not sure why you would do this).

Steve
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1798 Posts
  • 627 Reply Likes
Steve,
If I can also add some feedback, here are some differences I've seen in the audio peaking filters compared to an RF filter for CW use.

- high frequency audio (hiss) may be attenuated more effectively. And it can make listening less tiring over time.

- low frequency audio (rumble) may be attenuated more effectively

- fi there are two close CW signals the audio peaking filter make the signal you want to copy pop compared to the one that you don't want to copy. The Yaesu FTDX9000 also allows you to set the filter as an audio notch instead of a peaking filter and this can also work very well to attenuate a close by CW signal.

In some cases using both the appropriate RF filter and the Audio peaking/notch filter together works better than either one alone. My experience isn't on an SDR but the results may be similar.

Regards, Al / NN4ZZ
Photo of W5UN_Dave

W5UN_Dave

  • 315 Posts
  • 30 Reply Likes
First, let me say that the purpose of an APF is not to eliminate nearby interference.

Unless one has done serious, "digging out of the noise" DXing used a GOOD audio peaking filter (APF), one will not know of the great advantages it gives over trying to narrow the RF filter. Existing RF filters trying to do this are a joke . I make my case by referring to the PowerSDR 50 hz filter on my Flex 5000. It is not 50 hz, and the skirts are much too wide! A peaking filter is a different animal. Filter skirts must be extremely sharp, and bandpass very narrow (even as narrow as one hz). If Flex can make such a filter in the RF domain, show me, do it for PowerSDR! Talk to any of the top 160 meter DXers and they will tell you an APF is essential. I first used a homemade APF in my 2 meter EME cw days. Many times there was no way to get the cw info without the APF since signals were at, and even below the noise floor. My APF was adjustable in those days, and could be made as narrow as 1 hz. Ring, you bet, but this ringing works to advantage by causing the signal to "jump" out of the noise on the rings. Tiring, yes, but it serves its purpose.

My thoughts and experiences.
Dave, W5UN
Photo of W5UN_Dave

W5UN_Dave

  • 315 Posts
  • 30 Reply Likes
OH, and by the way the 6000 radio planned bells and whistles are great, but an APF is not a bell or a whistle. IT IS A TOOL!
Photo of Tim K8XS

Tim K8XS

  • 48 Posts
  • 8 Reply Likes
Dave and Al,
What cw tone do you typically use (ie 400Hz, 600Hz,...). Do you need to adjust the tone in real-time? If the answer is yes, what is the range you would need (ie 400-700)? When you are doing your
"digging out of the noise" DXing, do you still need a full break-in capability? Would a latency of 1 sec for a 1 Hz BW be unacceptable? What latency could you live with? Just curious .
Thanks,
Tim K8XS
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1798 Posts
  • 627 Reply Likes
Tim,
I didn't follow the latency question related to QSK. I would expect the feature to work whether using QSK or not (although I don't typically use QSK for weak signal DX).

I guess there are several ways to do the audio peaking/notching (i.e. audio dsp). I made this diagram to show the peaking or notching that I'm familiar with. Two of the three variables are menu setting, one is a real time adjustment.

The top diagram is for peaking and only the desired signal is shown. The bottom diagram is for notching and two signals are shown close together. The one on the left is the desired signal. The one on the right is the offending signal which is shown as a stronger signal in this example. Note the diagram is not to scale.

Click on the diagram for a full size view. Does this help or am I missing something?

Regards, Al / NN4ZZ

Photo of Tim K8XS

Tim K8XS

  • 48 Posts
  • 8 Reply Likes
Al,
I was thinking if the latency through the filter was 1 sec and Dave had a QSK requirement of 30 wpm, then those two items seem to be mutually exclusive. Any action that you would hear between your sending would have happened 1 second in the past. To me that isn't QSK.
-Tim
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1798 Posts
  • 627 Reply Likes
Tim, Ok I see what you are referring to now. I guess it depends on how it's implemented. If the audio DSP processing is very fast then QSK should be ok. I can't detect any noticeable delay in the FTDX9000 or other radios with that feature but I'm sure there is some. That would be for Steve and developers to comment on. I don't use QSK, for weak signal DX anyway but others may so we can see what Dave thinks.

Regards, Al / NN4ZZ
Photo of W5UN_Dave

W5UN_Dave

  • 315 Posts
  • 30 Reply Likes
Let me say, I have no interest in full, between the dits and dahs QSK when using the peaking filter, as signals are too weak for that anyway, and cw speeds are normally slow (t be used. If you have ever used a K3, notice the latency on cw at any speed. Turn the K3 volume up, and then plug in the headphones. It is there always. I think the 6700 can do much better with this. In fact, I expect it (and I will give the Flex boys time. It may take a few years)
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1798 Posts
  • 627 Reply Likes
Steve / N5AC -- any thoughts on the audio DSP peaking and notching feature? Tim indicated he added it to the features database. It really is a great capability for the "CW weak signal DXers." Even for good signals, the audio DSP can be adjusted to make CW sound smooth and clean for easy tireless copy.

This is a feature that can help the 6000 series become more attractive to the CW OPs interested in SDRs.

I'm sure you are busy with other pressing work but your feedback would be appreciated if you have time.

Regards, Al / NN4ZZ
Photo of Charles - K5UA

Charles - K5UA

  • 318 Posts
  • 88 Reply Likes
I concur that an audio peak filter (and audio notch filter) would be highly desirable for any serious DX or contest work.