Guidelines For Best Transmitted Signal Quality With Digi Modes

[w-u-2-o]

1. Follow the guidelines here to ensure that you are "48KHz clean" throughout all of your audio setttings:

https://apache-labs.com/community/viewtopic.php?f=13&t=2655

2. Always use DIGU or DIGL (almost always DIGU except for you RTTY types ). This will automatically remove all of the audio processing stages from the audio path, thereby ensuring a clean and un-distorted digi mode waveform.

3. Keep your ALC < 0dB. -1dB is fine. The ALC is an awesome, look-ahead, intelligent soft limiting algorithm, but you don't want or need any compression of your digi mode signal, as in this case compression = distortion. This is as opposed to phone operations, where the compression offered by the ALC may be seen as a desirable feature.

4. Contrary to what is recommended for conventional radios, always set your audio drive level from your digi mode software (WSJT-X, Fldigi, etc.) to -1dB, then set your desired RF output power level with the Drive slider. This is backwards from conventional rigs because in the ANAN series radios the gain in the PA and driver sections is fixed and therefore the entire dynamic range of the amp is always available. The Drive slider merely controls the output from the DAC into the fixed gain of the driver amp and PA sections.

5. Use the Single Calibrate mode of PureSignal to set your linearization prior to commencing operations. I.e. once you have your RF drive where you want it (using "tune" in WSJT-X or Fldigi, for example), use the two-tone feature (set to use drive power) and click the Single Calibrate button a few times until you get a good solution. Be sure to use DUP mode on the panadapter so you can see what is actually being transmitted for IMD. Note that the reason you can't use the automatic mode is because you are not hitting 0dB ALC where PureSignal kicks off automatic measurements, and also because the digi mode waveforms are not kind to the automatic algorithm.

6. If you are not working an entire sub-band a la FT8, i.e. if you know you are always going to center up on the 1500Hz sweet spot and then work a PSK, Olivia or RTTY signal or the like, then set your transmit passband to match your transmitted waveform bandwidth.

Do all of the above and you'll have a squeaky clean digi mode signal.

Special note: you will see noise "bursts" on the waterfall when digital modulations switch instantaneously from one tone to the next. This is normal and is a side effect of the FFT process used to display the panadapter and waterfall. These are particularly evident and noticeable when using one-tone-at-a-time modes like JT and FT modes. There is generally no need for concern about this.

[w9mdb]

Thanks a lot for posting this....

Couple questions
#1 Where is the ALC setting?
#2 Where is the "Single Calibrate mode of PureSignal"?

I'm probably blind but can't find either of these.

de Mike W9MDB

[w-u-2-o]

Thanks a lot for posting this....

Couple questions
#1 Where is the ALC setting?

There is no ALC control per se. Switch the meter to ALC during transmit. Adjust the audio drive level from your digi mode program upwards until you see -1dB. There are many ways to adjust the audio drive level. In some digi mode programs there is an adjustment right in the program. Depending on what sort of virtual audio cable software you are using there may be adjustments in that software and the Windows Mixer may also be involved. In PowerSDR you can adjust VAC TX gain. My recommendation is to set everything to 0dB (usually the maximum setting), then back off the digi mode program control, if available, by 1dB. If not available there, then back off VAC TX gain by 1dB. This keeps things obvious and avoids potential confusion because it is easy to forget about hidden Windows and VAC software settings.

#2 Where is the "Single Calibrate mode of PureSignal"?

When you click on the Linearity menu the PureSignal control window opens. You will see a button that says "Single Calibrate". If you hit that button PureSignal makes a single calibration and the PS-A button goes dark. To get back to automatic use the PS-A button.

[w9mdb]

Couple more questions.
What is a "good solution" for Single Calibrate? How does one know?

And I guess the "Correcting" means Pure SIgnal is doing it's thing?

Is this appropriate then for all digital modes?

I'm showing the highest harmonic 50dB down on an FT8 tranmissions which sounds pretty good.

de Mike W9MDB

[w-u-2-o]

Couple more questions.
What is a "good solution" for Single Calibrate? How does one know?

In DUP mode look at your IMD products. If the close in ones are down -50dB or more, then it is good.

And I guess the "Correcting" means Pure SIgnal is doing it's thing?

Yes.

Is this appropriate then for all digital modes?

IMHO, yes.

I'm showing the highest harmonic 50dB down on an FT8 tranmissions which sounds pretty good.

If that's in DUP display mode, then it sure does.

[Tony EI7BMB]

Just curious as to what guys generally set the level of RX1 AF to when using WSJT X for best signal to noise ratio on receive ?

[w-u-2-o]

Great question, Tony, but perhaps worthy of it's own topic, since this one is about transmitted signal quality, not received signal quality. Would you mind starting another topic, please?

[Tony EI7BMB]

You are right Scott, it does not sit well with this topic on reflection. I'll start a new one.

[W2QO-Jamie]

1. What should the transmit bandwidth be set for in FT-8 using WSJT-X?

2. I have Show mode/filter on VFOs turned on (in Thetis) and when set up for FT8 and VFO A displays DIGU and Var 1. That seems right but VFO B is showing LSB 2.9k right now. Should it be showing DIGU? and if so, how do I change it? (The 2.9k setting is the subject of the question above.)


Thanks for any help,

Jamie, W2QO

[w-u-2-o]

For FT8 I would use a bandwidth that is commensurate with the entire sub-band, which is usually about 3KHz. There is really no advantage on the TX side in going smaller than this, and it allows you to select your precise TX frequency in WSJT-X without fussing with Thetis.

Not exactly sure what your question is regarding VFO B. If you are using RX2, then RX2 will be in whatever mode you set it in. If you are using MultiRX VFO B will be in whatever mode VFO A is in.

[K9RX]

I recently read the section of the WSJT manual regarding the measurement tools. There is a "linearization tool' there... and it is stated one can use this to get this correct OR if one has the ability to do Linear Phase on TX (vs Low Latency) that you are set to go and need not worry about this. I don't know how much difference it makes but I've since changed my DSP OPTIONS setting for DIG to be Linear Phase on both RX and TX.

Also although this question is regarding Xmit - on RX IF you are going to use 60M you will either need to have 2 different TX profiles (read: PITA) or just create one at 2.8Khz width and use it on all DIG bands. It would be nice if this were automatic so I could use 3Khz (what I had used before I got in to 60M operation) on all bands but 60 and there, KNOWING that it is restricted, it would automatically be set/limited it to 2.8Khz ... but it is what it is. For the most part what this means is that if you approach the end of the passband, 2.8Khz, on TX the signal will be attenuated (lower amplitude) as it goes down the skirt. TX 100hz away seems fine and for the most part covers most bases since the majority of users, for whatever reason, limit themselves even more.

Gary
K9RX

[W2QO-Jamie]

Hi Scott,

Thanks for the response on the transmit bandwidth.

Back to my second question. I am NOT using MultiRX and NOT using RX2. I am simply in ordinary split mode. I am pointing out that VFO B is showing LSB as its mode while VFO A is showing DIGU as its mode. Is that expected behavior in split mode with Thetis?

I would have thought in split mode that VFO A and VFO B would both be showing DIGU as the mode.

Thanks again,

Jamie, W2QO

[w9mdb]

What option do you have checked for WSJT-X in the Settings/Radio/Mode?
I use USB and have Thetis CAT Control "DigL/U Returns LSB/USB" is checked.
That works here
de Mike W9MDB

[w-u-2-o]

If I'm in DIGU mode and manually select split, I see no problems.

I've never used split with WSJT-X and actually wouldn't know why you would or how! Sorry!

[w9mdb]

Split allows you to actually transmit outside your bandpass by shifting the carrier freq.

So if you have somebody at, say 2400 and that's at the edge of your bandpass you've set up, split will move that 2400 offset into the 1500-2000 region and bump up the carrier 1kHz to do it.

Mike

[W2QO-Jamie]

Hi Scott,

If you check the box to turn on the Display,General, Other setting to Small mode/filter on VFOs when you are in DIGU mode with split on what do you see in the VFO B box for the mode and the filter width?

Here is what I see:

DIGU Capture.PNG (359.43 KiB) Viewed 54877 times

It has nothing to do with WSJX-X at all which I had turned off for this screen shot and I re-started Thetis as well.

In fact upon further experimentation, no matter what mode I have VFO A in, VFO B always shows LSB and no matter what transmit bandwidth I set, VFO B always show 2.9k. (This is with RX 2 off and MultiRX off)

Perhaps this is a glitch in Thetis. It does not appear to affect the operation at all. It's just something I noticed that seemed odd.

Thanks,

Jamie, W2QO

[w-u-2-o]

I see nothing, because...wait for it...I do not have Setup > Display > General > Small mode/filter on VFOs turned on.

So that screen shot was very valuable, thank you.

Edited: I was playing with this some more, and I was able to get the mode to split, and get stuck, between VFO-A and VFO-B. In my case in the exact opposite way. I now have DIGU stuck on VFO-B.

BREAK

@ramdor ... Richie, I can't give you the exact steps to reproduce, but in switching between modes, band stacks, and split mode, I managed to get the mode annunciator to show different things for VFO-A and VFO-B and now can't get them to show the same thing. It's like the VFO-B mode annunciator (or worse, the actual mode) is stuck.

Thanks,

Scott

[w9mdb]

I turned that on the VFO display on my system and VFOB is showing LSB but is actually transmitting USB.
So me thinkst it's just a display problem.
The TS2000 MD command sets the mode for both VFOs and apparently that's not being done and the mode on VFOB doesn't matter.

de Mike W9MDB

[W2QO-Jamie]

It's not just the MODE that is incorrect. The BANDWIDTH display is also incorrect.

As mentioned, the actual operation is okay.

Jamie, W2QO

[w9mdb]

Actually testing does also show the "wrong" bandwidth 2900 instead of the 100-3260 I have set in Var 1.
And the 2900 is the bandwidth that is being used.
I ran a tune signal and it started dropping around 2900.
de Mike W9MDB

[W7GES]

I'm using Thetis v2.8.11 MW0LGE 21k7

I do have Pure Signal working in SSB using the PS-A Automatic mode.

I have a question about Pure Signal in the Single Cal mode for RTTY.

I open the Linearity menu, start a RTTY transmission and press Single Cal a few times. The button lights up green.

Do I need to do anything else to keep the rig / software in Pure Signal mode? I don't notice any correcting indicator, etc.

When switching to another band do I need to press the Single Cal button again, or is this a continuous process after enabling?

Thanks,

George / W7GES

[w-u-2-o]

The Single Cal process results in a single correction that is only valid for the exact conditions that exist at the time the Single Cal button was pressed. Primarily that means the state of your antenna impedance match, amplifier power level, and frequency. If any of those three things are changed then a Single Cal needs to be performed again.

However, you may find that the Single Cal result is reasonably valid over a range of frequencies. This can be determined by looking at the transmitted signal quality in the spectral display using DUP mode. If significant intermodulation products are seen then that would be an indication to redo the Single Cal.

Note that in Single Cal mode the correcting indicator on the spectral display status line is not shown. It is important to ensure that the correcting indicator is green in the linearity window. That is what designates that a Single Cal correction has been determined. Again, this correction is only valid under the conditions for which it was obtained. Changing frequencies, power levels, or bands, will make it less valid, if not completely invalid, even if the indicator remains green.

[W7GES]

Thanks for clearing this up.

The Pure Signal Auto mode works well for me in SSB but I noticed it wasn't working most of the time in RTTY. Once in a while it would randomly show the correcting message. Maybe it was while I was setting sound card levels or it may have been before I tightened up my transmit filters in my RTTY Transmit profile.

I think the Single Cal mode will work alright as long as I know how it works now. I will check out the spectral display while using DUP.

If I'm running I can calibrate after I adjust my rig / amp output power and pick a frequency.

I'll just need to remember this when changing bands / wide frequency changes / output power.