Flex 6500 echo on transmit audio

  • 1
  • Problem
  • Updated 3 years ago
Over the past three weeks, I have had reports on 6 metres (only band used) that my signal has a low level echo on it, which is distorted and the audio is lacking in high frequency response.  Flex 6500, SDR 1.6.21. Audio is fine through multi switcher on all other rigs.
Photo of Brian Morgan VK7RR

Brian Morgan VK7RR

  • 58 Posts
  • 11 Reply Likes

Posted 3 years ago

  • 1
Photo of James Del Principe

James Del Principe

  • 285 Posts
  • 29 Reply Likes
Brian, are you using the 'monitor' function and if so, are you using headphones or a speaker?   Have you tried tailoring the audio with the equalizer?   Have you tried this into a dummy load?   How about other bands?    You can record your own audio and then play it back to yourself. This is a better representation of your transmit audio than listening with the monitor.   73, Jim
Photo of Al K0VM

Al K0VM, Elmer

  • 583 Posts
  • 95 Reply Likes
Brian,
  A low level echo could be caused by RF feedback into the mic input, or it could also be caused by a ground loop if your external audio system shares the 6500 power supply..  In both cases, I would try running low power (1W)  with some one nearby.

AL, K0VM
Photo of Barry N1EU

Barry N1EU

  • 495 Posts
  • 121 Reply Likes
You should really spell out your audio path from microphone to 6500 in detail.  Is there an echo if you plug microphone right into 6500 and have DAX/MON/EQ/PROC disabled?  If not, slowly work back to your current setup and find out where the echo is introduced.
(Edited)
Photo of Brian Morgan VK7RR

Brian Morgan VK7RR

  • 58 Posts
  • 11 Reply Likes
Hi guys and thanks for your thoughts. I am not using the Monitor function. I am using a speaker. This set up has been in place for several years and nothing has been changed. I have gone down to low power with no change in the problem. The NCS multi switcher controls three other radios and the transmitted audio is perfect on them. There is serious earthing of all radios to a busbar and from there to a ground spike which is over 6 feet into the ground, which is less than 6 feet from the entry point to the shack. Dax, Mon, EQ, Proc. disabling makes no difference. As well, the graphic eq. has little effect on audio quality, whereas when I set it up, this had a very noticeable difference. I can record and play back my audio, plus listen to it in real time on a steam powered radio. Al, I do use a common power supply but again, this has been in use since the release of the 6500. I am very conscious of RF feedback!
Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 4025 Posts
  • 1248 Reply Likes
A couple other simple things to check....
* has something deteriorated in the cable between the multi-switcher and the 6500? Cold solder joint? cable pinched & broken lead?
* have you tried moving the 6500 to a different output of the multi-switcher?  You could have developed a bad channel/contact?
* Have you done a persistence reset?  (Export your profiles first!  then reset and import the profiles)
*Have you tried uninstalling SmartSDR, rebooting both rig and computer, and re-installing SmartSDR?  You might have gotten a corruption in your software.  Or something could have "glitched" in the rig and needs refreshed.

Just a start ...

Ken - NM9P
Photo of Barry N1EU

Barry N1EU

  • 495 Posts
  • 121 Reply Likes
The first two can be knocked off by just plugging the mic into the 6500, which I previously suggested.
Photo of Brian Morgan VK7RR

Brian Morgan VK7RR

  • 58 Posts
  • 11 Reply Likes
I have tried all suggestions except for Ken's (sorry Ken), but have gone close to resolving the problem. This forum is a good place to share ideas and often we know the solution, but need to discuss it with others to realise this.
The issue relates to the Processor and the mike level being seen by the Flex 6500. I use it without the + 20 db bias as the Heil mike does not need it. The NCS multi switcher has a level control for each input and each microphone. It did not matter whether I used the desk mike or the headset, the problem was always there. By elimination, I found that whenever the Processor is turned on, whether in Normal, DX or DX- position, I could hear this problem. Without the processor, mic audio as read on the Flex Level control never goes above - 10 dbm, so never peaks anywhere near 0dbm so no clipping. Looking at the output on my spectrum analyser, enabling the processor one sees (and hears) this audio. It is a measured 55 dbm below "normal audio".
My solution for now is to not use the Processor and ensure that DEXP is set correctly as it eliminates the fan & shack noise when I am not speaking.
I suspect something may have happened with the last but one upgrade that has made this so temperamental, but I don't mind not using the Processor and will check the condition again, with subsequent software releases. Then again Ken, on a rainy afternoon, I might try your idea as it remains a logical step which I have not yet taken.
Barry I had tried the direct connection of the mike as one of the first steps.
Thanks again for all your thoughtful ideas,
Brian
Photo of Barry N1EU

Barry N1EU

  • 495 Posts
  • 121 Reply Likes
So the processor seems to be introducing a -55dB echo to the audio?
Photo of Brian Morgan VK7RR

Brian Morgan VK7RR

  • 58 Posts
  • 11 Reply Likes
Yes Barry. So when I get time I will do a re program of the radio to see if that solves the problem.

73 Brian.
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9152 Posts
  • 3478 Reply Likes
I doubt this is the actual problem.  I could be wrong, but I suspect a ground loop or RFI via common mode currents on the coax.   I recommend you "divide and conquer" the issue.

I would eliminate the NCS by disconnecting it and the remove the antenna systems from the equation by connecting a dummy load directly to the radio.  Test at full power and listen to your transmitted signal on a different radio for the presence of echo.  I suspect you will not have any.  If you do, check the mic cable to make sure the cable shield is grounded and not connected to MIC- since the mic input 6000s are pseudo balanced.

Then start adding one component back at a time and test until you find the component responsible for the issue.
Photo of Brian Morgan VK7RR

Brian Morgan VK7RR

  • 58 Posts
  • 11 Reply Likes
Hi Tim and thanks for the suggestions. I have already removed the NCS and done this exercise, with no change. However I will use my dummy load and try again as this makes sense to me. This is particularly so because if I reduce power to a few watts I have no problem but at  50 watts and above it is there. It is no worse when I run the linear so I had always been thinking along the lines of RF feedback but could not explain why, with the legal limit, it was no worse than with 50 watts. The strange thing  is that the problem just occurred out of nowhere, with nothing being changed.
73 Brian.
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9152 Posts
  • 3478 Reply Likes
This all sounds so very much like RF common mode currents on the coax.  Any time you have coax out in the elements, there is a distinct possibility of a WX related change to the antenna system.