QSK?

  • 2
  • Problem
  • Updated 4 years ago
  • Acknowledged
What happened to QSK.  I have delay set to zero but I hear lots of audio popping while sending..not just the rx noise.  Happens with zero power as well.  Only sending at 26 wpm with an external WinkeyUSB connected to the front panel key input.

Also I was told by someone at flex there would be pin diode cw keying available as a user option.  If this was in the design I imagine it is already in there but not  yet available in the software?

73
Greg


Photo of Greg

Greg

  • 656 Posts
  • 63 Reply Likes

Posted 4 years ago

  • 2
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9029 Posts
  • 3431 Reply Likes
CW is always keyed using the PIN diode.

Does the slice control panel indicate that you are in QSK mode?
Photo of Mack McCormick

Mack McCormick, Elmer

  • 310 Posts
  • 114 Reply Likes
Greg,
As a primary CW op there is absolutely no audio popping here. I too use a WinKey USB Keyer. As one of the Alpha testers I particularly pay attention to CW performance. My delay is set to zero. 

The pin diode is in the system and works. It is only engaged above a particular speed and I don't remember that number. 

I have no popping at any speed.

All of the CW Sidetone is generated in the radio so the computer should not be an issue.

Mack
W4AX
Alpha Tester
Photo of Greg

Greg

  • 656 Posts
  • 63 Reply Likes
Tim,
The slice shows QSK...see pic.  But this is even with the delay set to 280.  I understand that the QSK indicator is only supposed to come on if delay = 0.  

Mark  - If the T/R relay for CW is a pin diode then why do I get all the relay chatter when keying with delay set to zero?  I would expect the radio to be just about silent when keying.  

Let me answer that question myself.  If you recall I am on a bunch of meds including pain meds for a severely pinched nerve in my lower back.  Had another cocktail of delaudid and toradol today too.  I am getting the relay chatter....fortunately it is coming from my amp next to the radio....the way things are set up it sounded as if it was coming from the radio.  My bad!!

The popping or artifacts I think are still in there.  The way I feel right now maybe it is either my imagination or just to critical.  For 6 years I have used a different rig whose QSK is the best I have personally used.   This comes close but I don't notice the popping on that rig.  Maybe popping is a poor term.  Maybe its the aggressiveness of the RX coming back after each break.???

Glad I realized my foul on the relay noise.


Photo of Greg

Greg

  • 656 Posts
  • 63 Reply Likes
Here is the pic...
Photo of Barry N1EU

Barry N1EU

  • 495 Posts
  • 121 Reply Likes
Greg, perhaps you can make a short recording of your rx audio with T/R transitions that captures the popping.  Looking at a graphical display of the captured audio waveform will by very helpful in determining whether there are audio artifacts during the changeover.
Photo of Greg

Greg

  • 656 Posts
  • 63 Reply Likes
I will try to do that today.
Photo of Steve N4LQ

Steve N4LQ

  • 568 Posts
  • 92 Reply Likes
I also see the QSK indicator lit no matter where "delay" is set on my 6300. Relay noise is just a tiny ping with one antenna but with a rx antenna connected to the xvtr port relay noise is very loud and there is no QSK even though the QSK indicator is still on.
Photo of EMIL-DL8JJ

EMIL-DL8JJ

  • 48 Posts
  • 9 Reply Likes

Hello guys,
here my comments:

External Key K1EL. 25 to 90 WPM
Output Power slider on 100W

1. RX = ANT1, TX = ANT1

Speed slider (SSDR) set to 100 WPM.
K1EL - 25 up to 90 WPM (test)
RESULT: QSK Ok, but not realý hear between the characters.

Speed slider (SSDR) set to 5 WPM
K1EL - 25 up to 90 WPM (test)

RESULT: BAD, you will heard you own signal pumping

Important question: Way there is dependency between internal speed slider settings and the QSK in ALL caces?
Exactly under the 20WPM of the Speed slider (SSDR) you will heard pumping.
Over 20 all OK.

2. RX = RXA, TX = ANT1

RESULT: BAD, you head relay klicking on every single charachter IF the speed slider (SSDR) is on 100WPM
 
IF the speed slider(SSDR) is on 5WPM you will heard relay klick only with the first one (like semi break-in) .QSK symbol in the slice is aways NOT activ.


3. RX = RXB or XVTR = QSK, TX = ANT1

RESULT: QSK is OK & no relays clicking. QSK symbol in the slice is active. But under 20 WPM on internal speed slider (SSDR) you will hear signal pumping

IMPORTANT: All tree behavior are the same with the internal IAMBIC (SSDR) key ON and only external paddle.

I thing you have some bad bugs with the new version.

Thanks
Emil, de DL8JJ

Photo of Steve - N5AC

Steve - N5AC, VP Engineering

  • 1004 Posts
  • 967 Reply Likes
Emil,

I've entered a task to go look at each of your comments here and get back to you -- thanks for the detailed analysis.  Tracking #DE2031
Photo of Steve N4LQ

Steve N4LQ

  • 568 Posts
  • 92 Reply Likes
Also, if the keyer is set at 19 wpm and below, when you let up on the key it sounds like a duck quack. Nobody could stand that for long! You must set the keyer above 20 wpm then for slower speeds use an external keyer or bug.
Photo of Steve N4LQ

Steve N4LQ

  • 568 Posts
  • 92 Reply Likes
Same results here with the 6300 except no QSK with separate antenna on xvtr input...Loud relay clicking. "Full duplex QSK" does not exist despite the claims by Flex. It does help if you turn the "master AF gain" full up and reduce the slice AF gain and sidetone. This helps with the pumping sound you mentioned. I think they need work on the timing sequence. It seems that the receiver is not muting fast enough and some transmitter noise is present between characters. Steve N4LQ
Photo of Steve - N5AC

Steve - N5AC, VP Engineering

  • 1007 Posts
  • 969 Reply Likes
Steve, I checked this today and there is a relay flip coded for QSK in the 6300 when you are in a XVTR/ANT1 or XVTR/ANT2 setting.  We've fixed this and it will be available in a point release we hope to have out in the next few weeks.
Photo of Steve N4LQ

Steve N4LQ

  • 568 Posts
  • 92 Reply Likes
Thanks Steve. Better tell the others cuz they don't seem to know. Now about this low power 85 watts max issue? Suppose that's a code thing?
Photo of EMIL-DL8JJ

EMIL-DL8JJ

  • 48 Posts
  • 9 Reply Likes

Steve,

plese check also my comments on 6700. That will be great.

Thanks.

Emil, de DL8JJ

Photo of Paul Christensen, W9AC

Paul Christensen, W9AC, Elmer

  • 313 Posts
  • 132 Reply Likes

Steve, N4LQ suggested using a dedicated gigabit LAN card in my PC for the Flex 6700 connection.  I ordered a new board online, price was about USD $10. 

In my case, CW QSK is now equal to the K3, even when using an external keyer.  It was good before, it's superb now.  Gone are the switching artifacts (clicks & pops) in the headphones and audio can be heard between high-speed keyed elements with Delay set to zero.  This still means the SSDR speed slider must be set to 20WPM or above to avoid a well-known switching artifact when the slider is at 19 and below.

Installation took a few minutes and SSDR found the 6700 right away.  If you're a die-hard CW QSK user, consider adding a dedicated gigabit LAN card to your PC.

Paul, W9AC


       

Photo of Steve - N5AC

Steve - N5AC, VP Engineering

  • 1007 Posts
  • 969 Reply Likes
Paul ... this should not be the case.  The QSK sending mechanism is completely self-contained in the radio.  Assuming your observations are valid, my only explanation is that the sending and receiving of the TX status update from the radio to the PC is delaying QSK processing.  Operating on this premise, I've entered a defect (DE2036) to find out why this is the case.
Photo of Paul Christensen, W9AC

Paul Christensen, W9AC, Elmer

  • 313 Posts
  • 132 Reply Likes
Thanks, Steve.  Comparing to a dedicated LAN card may offer some clues.  Observing the interactions on a scope, no change is noted -- only the complete removal of switching artifacts heard in the headphones.  With the change, QSK performance is truly remarkable.   Maybe the 19 WPM issue could be reviewed at the same time if its related?  Tnx! 
Photo of Steve - N5AC

Steve - N5AC, VP Engineering

  • 1007 Posts
  • 969 Reply Likes
Yes, I have a defect for that as well, Paul.  Thanks
Photo of Steve N4LQ

Steve N4LQ

  • 568 Posts
  • 92 Reply Likes
The switching speed doesn't necessarily change but there seems to a delayed mute command coming back through the network which keeps the receiver open during the onset of the sidetone. This results in noise on the leading edge of each character. With a direct PC connection the mute command gets to the rig soon enough to prevent sidetone corruption. Using the dedicated NIC puts QSK in real-time instead of delayed. I no longer need an external audio filter to eliminate the keying artifacts. This isn't true with the speed set under 19 wpm. In that case the noise is on the trailing end of the sidetone. Letting up on the key produces a HONK. I suspect the mute isn't being held quiet long enough. Steve N4LQ