FT-8 Frequency Jump on Flex-6500

N1UW

Thanks to Ria's most excellent set-up instructions, I nailed FT-8 setup on the first try.  But, I notice with slice set to 7.074, it jumps to 7,0735 when I transmit. I setup FT-8 exactly according to Ria's instructions and screen grabs, including "Fake It" for Split Operation.  Out of 50 or so QSO attempts, I have had 1.5 Q's.  I don't know if this frequency changing behavior has anything to do with my low Q rate.

Any thougts?

K3DCW

That is exactly what "fake it" is supposed to do. Rather than putting your rig into split mode, it will shift the VFO slightly to keep your transmitted energy in the center of your filter passband to maximize signal purity.  When done transmitting, it goes back to the normal VFO frequency.

Ria

K3DCW is 100% correct on this one. "Fake it" is your culprit, and it's a feature, not a bug. 

N1UW

So, the frequency shift is a good thing.  "Culprit" is  not meant to imply it is a "bad" thing?

Thanks for helping straighten me out.

Frank
- We have no problems, only insurmountable opportunities.

Chris DL5NAM

With all this digital modes you spread out in "your bandwidth"  off your TRX (what filter you have select ex. 2.7 or 3.0 or ...) You using always USB mode and the real bandwidth of this digital modes are much less then 100 Hz.
That's allow us to have many QSO's at same time on "one" QRG.

You can seen on waterfall display where all the stations are and a marker show's your TX  will work

So dont worry

73 Chris

Bill Kendrick

Typically the issue is timing. If your clock is off then you miss QSOs. VK9MA suffered from this problem. Though several people were calling his clock was off by seconds and he did not decode anyone. If an operator was clever he could have shifted his clock by the same offset and VK9MA would have decoded him.

You should download and install D4. It is an application that syncs your clock with local network servers keeping it within 200 ms. This runs in the background. Once you have this going you may see a change.

Bill
N6RV

p.s. The frequency shift also occurs on the Flex 5000a. Apparently the tone out of WJSTx is at a single frequency and to shift it to the desired transmit tone the DDS in the radio is shifted.

Ria

It’s like the word “awesome” which can describe things that evoke fear, yet most people know it as something that is very good. It’s all good.

Ria

I like meinberg as it is a more intelligent way to steer the clock, rather than D4. But if you think time is an issue, look at the “dt” column and see if it is more than 1 for most stations. Ideally it should be less than 0.5.

K3DCW

Exactly.  Meinberg constantly adjusts the clock in small sub-millisecond amounts to make the large jumps non-existent (other than at first start it up). It smoothes out the system clock and the way it functions in Windows.  Once setup, it is flawless and just works in the background completely out of sight and mind. It eliminates any possibility of your time being off. 

D4 simply adjusts the time all at once which can result in large jumps if you have a very poor system clock or only sync every few hours/days.  

N1UW

Thanks all the great suggestions (timing, etc.) but I think I have sleuthed the real problem.  Due to a killer noise level in the neighborhood I use a DXE magnetic loop receive antenna.  Looking at my short log, I can see my incoming signal reports are in the toilet bowl compared to the signal reports that I am sending out.  This is telling me my Tarheel screwdriver as transmitting antenna is not doing the job. I have some HOA restrictions, thus the teeny antennas.

Rich McCabe

Can someone elaborate on this fake it a little more please

I have not been running it and just started. I have always assumed it would be hard to transmit an audio frequency down very low as it would almost be zero beat.  I realize you can create the same frequency by shifting the transmitter as well.

Of course the station you are working is hearing the same frequency regardless so not sure what that matters.

Is there some benefit from a hardware standpoint keeping the audio frequency in the center of the passband? 

I am sure that question was clear as mud.

Rich

Ria

For conventional radios they are optimized for human speech, so they transmit the most amplitude and best quality within a certain frequency range. For Flex radios I don't think it matters much to be honest because in DIGU everything is "flat." I haven't really noticed a difference in who is hearing me either way. 

Rich McCabe

Thanks Ria.  You just answered what was going to be my next response

Ria

In SSB there is also the issue of audio harmonics passing through the SSB filter so they try to keep it within 1.5 to 2kHz. 

Oh, see in the manual, what do I know:

https://www.physics.princeton.edu/pulsar/K1JT/wsjtx-doc/wsjtx-main-1.7.0.html

• Split Operation: Significant advantages result from using Split mode (separate VFOs for Rx and Tx) if your radio supports it. If it does not, WSJT-X can emulate such behavior. Either method will result in a cleaner transmitted signal, by keeping the Tx audio always in the range 1500 to 2000 Hz so that audio harmonics cannot pass through the Tx sideband filter. Select Rig to use the radio’s Split mode, or Fake It to have WSJT-X adjust the VFO frequency as needed, when T/R switching occurs. Choose None if you do not wish to use split operation.