Raspy receive on SSB

  • 1
  • Question
  • Updated 3 years ago
I have noticed lately when operating SSB that when I'm finished transmitting, sometimes the audio is very raspy for a few seconds on the receiver during a QSO. It's almost as if the AGC isn't working right or perhaps WNB or NB are messing with it, though when I switch those off, it still does it sometimes. Anyone have this experience or is my radio developing a hardware issue. I normally run the rig with AGC set to medium and the gain set to 35.
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes

Posted 3 years ago

  • 1
Photo of Andrew O'Brien

Andrew O'Brien

  • 384 Posts
  • 44 Reply Likes
A latency/buffer issue of some sort perhaps ?
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes
If there is, it must be going on inside the radio. If you were referring to network latency, that is always <1ms. No problem there. No big deal if I can't find another user with the same issue. It's under warranty and they can calibrate the power output while it's there.
Photo of Rick  WN2C

Rick WN2C

  • 306 Posts
  • 47 Reply Likes
I have had the same issue when the WNB is on and set to a mid way setting or even higher. The NB the same when set to mid way or higher and definitely when both are on. But usually only on a strong (S9 or better) station.
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes
Yes, that seems to be the case. Well hopefully that can be improved in a future software update. When the static crashes are strong, it's tough to not be able to use those. The NR by itself just doesn't do the trick on a noisy night. I don't remember the radio doing that in the past. I don't have that problem with NR, NB, and such turned on with my other radios. But glad to know it's not just mine doing it.
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9197 Posts
  • 3557 Reply Likes
I suspect you are using the wrong noise mitigation technique for the type of noise you have and that is what is affecting your receive audio. Band noise (random) is not periodic (pulse type), so the NB and WNB will not bee that effective.  WNB is only effective on highly periodic noise.

I recommend that for band noise, you turn off all noise mitigation properly set the AGC-T first.  If you need additional noise mitigation, use the NR.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3972 Posts
  • 967 Reply Likes
If it is raspy something is not set correctly. It should not be raspy. If you are using WNB, the agustments should be made very slowly to allow the filter to program. Also the AGC-T is sensitive and needs to be set so that the noise begins to go down as you back it off.
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes
I've tried it that way too, Bill. No difference. It's still a random thing. And I generally don't change the agc-t setting at all, so that shouldn't even be an issue. And you say slowly? I've tried moving it up and down very slowly in small increments? How much slower do I have to go? With the hardware being in the radio and supposedly being quite capable of doing it's job and then some, how long should it take to respond from a capable computer with a network latency of <1ms ?
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3972 Posts
  • 967 Reply Likes
As you move the slider the radio is calculating, it takes a couple moments to finish. Not moving the AGC-T much is a problem. It should be moved often as conditions change, Never set it and forget it, both work in tandem. If you are getting distorted audio than something is set wrong.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3972 Posts
  • 967 Reply Likes
I also recommend going back and reading some of the threads on this subject. All this has been fully explained.
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9197 Posts
  • 3557 Reply Likes
"And I generally don't change the agc-t setting at all..."

This is part of your problem.  I can't agree strongly enough with Bill's comments to this statement.  The AGC-T is the most powerful control the radio has for improving the SNR and should be adjusted when you change bands or when band conditions change.  If your SNR is not optimized before you start using NR, NB or WNB, then you are fighting a losing battle.
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes
Tim, some of the noises I have to deal with living in the city and likely bad power line noises in my area are helped by the NB and WNB. It's simply is not just a simple AGC-T and NR adjustment. If necessary, I can do a video one night when I'm dealing with those noises and send it to you. I mainly run 75 meter band later at night and the AGC around 30 to 35 seems to work best as a starting point where I've knocked the noise down some without killing off people I'm trying to hear. Some nights I don't need to turn on any noise mitigation. Many nights, I have to use a little NR and a moderate amount of NB and WNB. I can assure you that those don't stay on if they don't make a difference. And again, I don't always have the raspy receive audio. But when I do, it seems the AGC-T and NR adjustments don't do anything to stop it, while disabling or adjusting the others to such a low level that they don't kill noise anymore the raspy audio is gone.
(Edited)
Photo of Justice Wakefield

Justice Wakefield

  • 1 Post
  • 0 Reply Likes
I am experiencing the same issue. Its seems to be very random. However I noticed it seems to be worse when the noise floor is high.
Photo of Ken - NM9P

Ken - NM9P

  • 4239 Posts
  • 1351 Reply Likes
As an Elmer, I have experienced people's complaints of "raspy" sounding audio caused by too high a preamp-setting, and distortion generated by too-high settings of WNB. NB and NR, or by using ANF instead of working with TNF....

But more often than not I have seen it when people are using their AGC-t backwards - i.e. they are using it as a volume control, starting at the left and turning it up until they get a little noise, and then using the volume control to bring things up the way they want.  Then they use the AGC-T as a volume control as signals vary from weak to strong.  When they take this approach, many signals are not being fully controlled by the AGC.  The weaker signals are boosted by the AF Volume, but not limited by the AGC.  Stronger signals will drive the audio chain into distortion before the AGC ever gets a chance to limit them.  Thus making for a "raspy" sound.

The correct way to adjust AGC-T is to start with the control full to the right and adjust your AF volume so the noise is at a comfortable level.  The move the AGC-T slider to the left until the noise just begins to drop.  This is the "Sweet Spot" or "Knee" point.  At this point, all signals are being controlled by the AGC system.  Make volume adjustments with the slice or the master volume controls.   If you want to reduce noise levels a little, for example on a noisy static-filled band, you can slide the AGC-T control a little more to the left.  But understand that this will reduce the strength of some of the weaker signals.  If you want to hear the weaker signals again, you will need to bring the AGC-T control back to the right a little.  Simply boosting the volume control will eventually put things back into the bad scenario above.

I find that when I have the AGC-T set correctly, then when I engage WNB, NB, NR, (or APF on CW) that they are much more effective.  NOTE:  after engaging these controls, it is sometimes helpful to nudge the AGC-T control SLIGHTLY one way or another to maximize its effectiveness.  Especially under these circumstances:

WNB, when engaged on the type of noise it was designed to battle, will often reduce your noise floor by anywhere from 3-20 dB.  I have seen everything in between, usually between 5 & 10 dB, depending upon the band and the type of noise.  After waiting for the WNB to "settle" it can be helpful to move the AGC-T to the right a LITTLE to compensate for a lower noise floor and bring the weak ones newly revealed out of the noise.

NB - similarly, a reduced noise floor within the slice can sometimes be tweaked with slight adjustments of AGC-T.

NR:  Sometimes a slight tweak of AGC-T can help find the best balance point between readability and noise reduction..... Adjusting RX EQ can also compensate for some of the high frequencies lost to the NR circuitry.  But TOO MUCH will bring back all the noise that you just eliminated with NR!

Changing Receive filter widths - a more narrow filter pass less noise.  A slight AGC-T adjustment will fine-tune noise response.

APF - likewise - when activating APF to enhance CW signals, the noise profile will shift slightly, and a tweak in AGC-T can help fine-tune things.

The bottom line is - The AGC-T control is one of the most powerful, most misunderstood, and most misused  controls on the 6000 series.  It sets the parameters within which all the other DSP functions do their job.   Learning to drive these various controls is a bit like learning to drive a high performance sports car with manual shift.  All of the controls interact.  Automating them would make it easier to drive, but ultimate performance might suffer unless the automation routine is VERY, VERY good!  At present, I prefer to drive them myself.  (I have mine ported to the center button of my FlexControl Knob) 

I'm not saying that some of the DSP functions like WNB, NB, APF, and especially ANF cannot be improved.  They can all be improved. (ANF especially)  But NR is greatly improved, and is currently as good as I have seen it as it has improved over the past three years.  But the key, I think is to learn, by practice, how to master the AGC-T and then learn how to drive the other controls along with it.   

Ken - NM9P
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1853 Posts
  • 672 Reply Likes
Hi Ken,

RE: The bottom line is - The AGC-T control is one of the most powerful, most misunderstood, and most misused  controls on the 6000 series.  It sets the parameters within which all the other DSP functions do their job.   Learning to drive these various controls is a bit like learning to drive a high performance sports car with manual shift.  All of the controls interact.  Automating them would make it easier to drive, but ultimate performance might suffer unless the automation routine is VERY, VERY good!  At present, I prefer to drive them myself.  (I have mine ported to the center button of my FlexControl Knob) 


Since adjusting the AGC-T is needed so often (for band changes, antenna rotation, conditions, etc) I suspect for many users having the automated feature that tracks the noise floor would be a big improvement.  IMHO, Even if not perfect it would likely be much better than neglecting to constantly monitor and adjust it.  And for anyone that wants to continue to manually adjust it you could but you may find that the automation does a fine job.  

It's the number 2 idea on the list with 51 votes so hopefully it will be offered in the near future.  

https://community.flexradio.com/flexradio/topics/option-to-automate-the-agc-t-setting

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
6700 - HW.................. V 1.9.7.85
SSDR / DAX / CAT...... V 1.9.7.169
Win10
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3969 Posts
  • 967 Reply Likes
One of the big beefs some have leveled against Flex is that there are too many automatic features, not anoth things to tinker with. They want more open settings such as what you can do with PSDR. They want even more control. I know there are some that just refuse to learn the correct way to operate such a high tech radio, but if people would just take the time to learn, they would find that things are not broken, just not setup correctly.
Photo of Ken - NM9P

Ken - NM9P

  • 4239 Posts
  • 1351 Reply Likes
I agree, Al, that some form of automatic AGC-T would help the average user, and I support it if it is selectable.  I would also suggest that there be some adjustment of the automation, for example, allowing the user to set some degree plus or minus referenced to the "perfect knee point" that the automation routine selects....

A bit like the +/- setting on automatic exposure cameras that lets the user lighten or darken the picture according to taste.  Once set, the "knee point" will be the same everywhere (Or better yet, saved by band - I want it a little more quiet on 80 in the summertime than on 6 in the winter!)

An adjustment like this would allow fine tuning the position on the knee curve.
That would be sweet, especially if it readjusted itself after other DSP functions are engaged or adjusted....
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes
Bill Buchanan I take offense to that. First, let me say that if you had read my posts from the beginning, you would know that my issue appears to be somewhat random given the same bands conditions lately. Second, you make clear that you think anyone who comes into the community asking for help or advice is nothing but a big dummy. I'm sure I still have very much to learn, but that doesn't preclude the slim possibility of a software or hardware issue. None the less, it's my understanding that's what the flex radio community is here for, so we may share our experiences and help each other with any problems that may arise, as well as share ideas for future improvement of the software. Aside from a very few issues I've had, I'm very happy with my 6300. I cut my SDR teeth on a 1500 for a couple years before purchasing this one, and I also own what some have described as a science fair product. A 200D and i didn't own it a year or more ago when people were commenting about buggy software and having to calibrate power output, etc. I had it up and running as quick as I did my 6300 and upon first operation that other than the frequency calibration, it was ready to go. I enjoy both very much.

As for the other gentlemen who gave me advice and their info to try, I appreciate it very much and will gladly try them and am looking forward to trying out the newly released software ASAP. Also, I did read other threads on similar issues. Some of those threads did not help and I took a chance that maybe someone had some different ideas on the matter.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3969 Posts
  • 967 Reply Likes
Scott, really you should not get so bent out of shape, There is no indication that my last comments were aimed at you. It is a general comment. It is true, time and time again people think something is broken when really they just didn't understand how things work. After they understand things were good. If people would just take the time to understand things, they could get more from their radio. In your case, you may have something wrong with hardware, It is rare but who knows.
(Edited)
Photo of Phil m0vse

Phil m0vse

  • 208 Posts
  • 38 Reply Likes
I have been a Flex owner for just over a year now (first 6300 and then 6500) and for the first 8 months or so, I barely touched the AGC-T and had many of the problems described. I now twiddle the AGC-T knob on my Maestro with nearly every contact and quite often somebody that is barely above the noise will suddenly 'pop' out and be fully readable. I can't overstate how much of a difference learning to use the AGC-T properly has made to my enjoyment of the 6500. 
Photo of Ria - N2RJ

Ria - N2RJ, Elmer

  • 2314 Posts
  • 956 Reply Likes
The "raspy" noise is common with noise blankers. With my older Icom 756ProIII it would sound raspy if I engaged the noise blanker on some signals, particularly those that are wide. With SmartSDR being more "intelligent" it would make sense that it would sound raspy at the start then change to normal.
Photo of Scott N8UMW

Scott N8UMW

  • 155 Posts
  • 47 Reply Likes
Even with the new software, it appears to be only raspy when I have to use NR and WNB together at moderate to high levels. I'm guessing it may be something I'll just have to live with, given my location and noise levels. I will keep trying different things and see if i can find a happy medium. The new software installed without issue, as usual. And sorry Bill. That's just the way out came across to me. Not really bent. And maybe I'm crazy, but the sharp filter for voice seemed to help. Like that new feature.