My article about sound card levels

Having your levels correct is critical to maximum effectiveness in both TX quality and also for being able to decode signals properly.

The way to think about how this works is, in SSB mode you set your rig to 100W output and then adjust your mic gain and away you go. If you whisper into your mic, you will see very little power output if adjusted properly. As you make your voice louder or get closer to the mic, your voice will drive the rig harder and you will see more output power.

Digital modes, you run the rig RF power level at 100W (Or whatever it's max power output is, 5W, 25W, 200W etc) and use the soundcard device adjustments to drive the rig to the desired maximum level of power output. Use your software to adjust the drive for lower power operation.

In fldigi, for audio OUTPUT drive level, the numbers in the far lower right corner in the TX level attenuator box are where you adjust at the software level, they are next to the AFC toggle, just to the left of it.

The level of your INPUT or receive audio is just as important. With the open source flidigi software as an example, you look at the VU meter in the bottom center. You typically want the VU meter at -50 to -40 area when you are tuned to a passband with no signals coming in.

The diamond indicator should show green a few clicks over black (too little signal) but no more than yellow.

If you see black you have too little audio and if you see red, too much. You want to be just right here for decoding optimally or it will lead to frustration.

In Windows, in the Control Panel -> Hardware and Sound -> Manage Audio Devices it's the easiest place to ensure you don't have your audio devices set to Default (all OS noises will go through your rig if you do) and if you click the devices and go to the Properties tab, you can rename them.

I recommend that the Playback tab device be named to "Rig Drive Level" and the Recording tab device to "Waterfall level". You'll need to re-select the devices in your software setups after renaming, but it makes it a lot easier to remember which line does what.

So this tip you may want to experiment with. Have you ever been in a QSO and had a much stronger signal than the one you were working come in the waterfall and "pump" your signal where you lose decoding? Here's how to avoid that.

When running digital modes, almost always, I turn the AGC on my rig OFF. I run RF gain wide open, set my INPUT audio levels from the rig to the software until proper levels are indicated on the fldigi VU meter. If I get hammered with a very inordinately strong signal that starts to decode poorly, I slowly ride the RF gain back to reduce the signal for processing.

Running the AGC off keeps strong adjacent signals from "pumping" the gain and wiping out weaker signals. It is a massive improvement, particularly if you are working psk-31 or any mode with a lot of close adjacent signals.

If your rig won't allow for AGC to be turned off, if it has an option for fast AGC, set to fast, it will respond better for digi mode use.

Also, double and triple check that you do NOT have any NR or NB filtering enabled as they will really badly affect the ability to decode incoming digital signals.

Ever see someone calling and calling but you can't figure out what mode they are using? Try figuring out MFSK vs Domino vs Thor vs Olivia. After enough time you may be able to, but really it's pretty hard. Also what if you like to play with even less common modes like Feld Hell or Throb? Again in time you may recognize them by eye or ear but most people cannot.

The way you fix that is to enable transmitting the RSID encoding at the start of your transmissions. The RSID is a series of tones that identify what mode of TX is to follow. If you set TxID it will send the RSID for the mode when you TX, if you enable RxID you will see what modes people are calling with.

You don't need to enable RSID with every transmission you make, but for calling CQ it should be sent with every CQ call so that people can figure out the mode you are using. Once engaged in a QSO, you can choose to disable it as it's just added traffic with no purpose.

Again with fldigi (noticing a theme here?) you can toggle both RxID and TxID with toggle buttons at the top right of the app and also you can toggle them in the calling macros that automate what you send for a CQ string in a button push.

The one exception to this is when you are working PSK-31 as it is a mode most people recognize right away and tends to be operated in a given passband with others running the same mode. If you are running QPSK or PSK-63F and others beyond the plain BPSK-31, use the ID so people can figure out what you are running.

If you find your rig jumping modes using fldigi and wonder why, go into Configure -> Config Dialog -> IDs and under RSID check the checkbox for "Notify Only" and it will just pop up a window that you can choose to click or not if someone identifies using a different mode. The search passband option allows you to scan the waterfall width for other signal modes.

In nets, you want the notify only box unchecked and you want the "search passband" also unchecked. This will allow you to follow mode changes on the fixed spot you are tuned to as the net progresses, without getting confused by other adjacent signals.

Frequently, during a QSO the contact gets broken due to nearby signal interference. The AGC tip solves an awful lot for this, but there are still times when nearby signals (The FT8 hordes, the piles of various Winlink on top of PSK sections on 20m as examples) can interfere with your contact. If your rig has filters in it, use them. People seem to forget or don't realize they can enable the filters on their rigs.

In the software you are using, when in a contact, most of the time I recommend using the QSY to center or sweet spot to center your contact on 1500Hz. The reason for this is to optimize for the frequency response of your passband for decoding, but also, because most rigs' filters are centered there or 1kHz. Once you have the signal centered, you can apply your rig's filters if available.

My rig is a TS-590SG and it has DSP filtering that works really well from 5kHz down to 200Hz wide. I typically run the filters 20% or so wider than the mode I'm working, or minimally enough to blank out any adjacent signals. It significantly enhances decoding. As an example, say I'm running Olivia 16/1000 which is roughly 1kHz wide, I'll run the filters centered at about 1500Hz wide. If you go too tight you can negatively affect the decoding. With narrower modes of 500 Hz or 250Hz I'll run 600Hz filter or 400Hz or so.

If you have an older rig with open slots, and do a lot of NBEMS or Olivia work, a 1.8kHz SSB filter and a 500Hz filter are very useful to add. If you can add only one and message traffic isn't something you need a 500Hz is very useful and covers a ton of modes for general chatting.

If you are fishing for contacts, leave the filters wide open to see what is around. If you are calling CQ in a crowded spot, tighten the filters around your signal.

If TX Hangs - Escape!

I'm sure you've hit a macro that went wrong or had some other issue that hangs TX. Just remember this.

Escape! Escape! Escape! The esc key in Fldigi and I believe a bunch of other software will halt TX, it's a good pull cord to have ingrained.

Other Rig Settings

• AGC off as mentioned above - really, try it.

• No TX/RX EQ enabled, can drive you crazy if you forget

• No NR or NB settings enabled, will harm decoding greatly

• No ANF or BC, notch filters, they will hurt decoding

• No speech processor, same reason for RX/TX EQ

• Make sure your RF gain is wide open or that you realize where it is set (AGC off will show no signal level on s-meter)

• Squelch open, on FM in particular for simplex, open the squelch, far better results. With a repeater it doesn't matter much as the assumption is you can hear it anyway if you can get into it.

• Rig set to full power output always, control actual TX power out with your audio drive levels.