CW Echo

K1ESE

Both in this weekends CW contest and in the 160 meter contest, there were times when my CW transmissions were being echoed on my 6500.  Using QSK it sounds like someone else is sending at the same time, but at the end of a word what you realize is that the sound is actually your own transmission.

Perhaps the sidetone is being echoed.  It is rare, mostly on a noisy band with a narrow filter, 400 hz or less, and usually with some APF on.

Has anyone else noticed this?

73 de K1ESE
John 

KC0EM

John,

I think you are experiencing what they call long-delayed echoes, LDE. I have seen them written up several times over the years. The first time it happened to me I had my novice license and was working a JA station. It took a while for me to recognize I was hearing myself. It has happened to me a couple times more over the years. I don't know if there's ever been a good theory about why this happens. Maybe someone else knows.

VC
KC0EM

Giulio

I have the same problem here, you can't go at 30 wpm and full break in. 73 de IZ3EAW Giulio

Tim - W4TME

I have entered this problem report into our bug tracker as issue #4506 for additional investigation. Thank you for the defect report.

K1ESE

Tim -

While not an echo, this is another symptom.  It only happens on 80m or 160m in the evening.  Might be related to band noise.

In any event, I fed DAX into Audacity and sent  "TEST".  Here's the results with P/CW Delay set at 0 ms.

Most of it is gone if I set Delay to 70 ms.


73 de K1ESE
John

Eric-KE5DTO

Was Full Duplex enabled when this was going on?  Which antennas were you using for RX and TX?

Tom K3TF

Eric,

A little more information here. I have been experiencing the same problem that John noted. I did not take any scope pictures but this is by ear. 

Flex 6500 with the newest software 1.10.16 

Delay set at 0. 
Not in FDX mode

It appears to be noise level related. CW filter is set at 800 hz. and filter latency set at the auto setting.

With a dummy load no "echo" on all bands.

With a dipole on Antenna 1. No other antenna hooked up. QSK icon is on. Headphones connected. AF Gain half scale. AGC-T about half scale. CW pitch set at 700hz. 

15 meters noise level at S0 no echo.

40 meters noise level at S4 slight echo. I can live with it.

80 meters noise level at S6 echo is bad. It is on my frequency 700hz tone. The echo follows my RIT setting. If I RIT plus or minus I will hear the echo change pitch. I can correct the problem if I dial in about 35 ms delay. If I connect up a dummy load so that the noise level on the S meter goes to 0 there is no echo.

It sounds like if when I release the key if there is noise coming from the antenna at that time, possibly raising the software AGC or S Meter reading I can hear the very end of my last CW dot or dash.

I had a Flex 6500 about 2 years ago and do not remember this problem. Back then I was involved with the xmit clipping of the RF CW envelope when using an external keyer and setting the Flex speed control to a slow setting. I worked with Steve Hicks on that issue. The CW envelope is perfect now. 

 I would be glad to test an older version of SDR Software to verify the problem did not exist before. 

I am an analog hardware EE familiar with Superhet designs. Not much experience with SDR receivers. 

Hope this information helps. 

Regards,
Tom 
K3TF.....

K1ESE

Not full duplex and just one antenna - ANT1 for RX and TX.

Eric-KE5DTO

Thanks for the info.  Does the AGC mode (Slow, Fast, etc) seem to make it better or worse?

Eric-KE5DTO

Thanks Tom.  We appreciate your input.  Same question I asked John above -- Does the AGC mode (Slow, Fast, etc) seem to affect the echo?

Tom K3TF

Eric,

I will check later. All my tests were done with AGC in Fast mode.

Tom.....

Tom K3TF

Eric,

Tried to reproduce the "echo" when listening to a noisy band S6 or greater and now can not get the problem to reproduce.

Today, just for fun, I reloaded SSDR VS 1.10.16 and this time after the software load I decided to turn off the Flex 6500 and disconnect the 13.8V power supply for 5 minutes. I was reading about possible sub-processes coming up in "wrong state"??. Now no "echo" on noisy bands. This is the first time I did the power disconnect for 5 minutes. 

I will keep an eye on the problem showing up again. 

I did not want you going crazy trying to fix a problem that may not be there.

The QSK is working FB, the CW waveform looks great.

Life is good.

73,
Tom
K3TF........