My Audio Setup

To start off, let me say that I was a touring soundman and a front of house mixing engineer in clubs full time for a number of years. The only reason I play with the equipment I do is simply because I enjoy the gear and tinkering with processing. 

There is absolutely no reason any ham should go down this rabbit hole unless they find it enjoyable. I've seen many people get frustrated in this area and spend money that they really should not have spent in the endeavor.

That said, I'll start with some fundamentals that any ham can follow to improve the quality of their audio and the cleanliness of the signal they are transmitting.

What is my secret?

There is no secret. What you hear is a combination of the hardware and equipment for sure, but this is very important, it is also the result of my experience in using all of it, the endless combinations possible and having spent hours playing with it...even with the experience I have.

Mic choice, use of processing and EQ and how to apply all of that simply takes time and experimentation. Wedging that result into a narrow 100Hz to 3kHz pass band for SSB just complicates things even more. The biggest improvement in my signal is the proper application of compression, downward expander (noise gate) and EQ, in that order. This is why my audio character stands out from a stock microphone.

This will take time and patience to learn.

The equipment I have kicking around

• Microphones

  • Shure SM-57 - I just got this one not too long ago, a long time classic. It's a dynamic mic and you'll want the Shure A2WS windscreen for it. It's really good on my voice as it isn't super heavy on the bass/low-mid frequencies. It may be the easiest one yet to get tuned up for my voice and has really great noise rejection. It really needs a pre-amp, if you want to run it into a rig directly you will want a Cloudlifter or similar as a pre-amp before the radio.

  • Earthworks Ethos - This mic is pretty incredible, not really as far as 3kHz of amateur radio use goes but in general. This is a small diaphragm condenser mic, based on a reference microphone design that Earthworks adapted for close talking, voice use for broadcast. Extremely natural and flat response with extremely smooth high frequency response with no built in boost. It handles plosives very well and for this type of condenser, it doesn't overload when worked up close. The sE8 has a similar response but overloads easily and has zero plosive protection where this one plugs all those holes, albeit at a higher price point.  I have a powerful baritone voice that overloads many mics and microphones with peak boosts in 4-6kHz range make my voice sound really gritty and this mic is a really nice combo with my voice.  I just dip some low-mid and a few room boom mid-range areas and it's all set to go, no unnatural deep carving needed.

  • SE Electronics sE8 - A small diaphragm pencil condenser microphone. This mic is at this point the closest runner up to the SM-57 for being the best balanced for my voice. Being a pencil condenser mic, it is very sensitive so I run this mic with the 80Hz high pass and the -10db pad enabled, both work very well. These mics need pop filters and are really meant to be worked 8" or so away vs close work. I find 8" the sweet spot to manage plosives and also to avoid too much proximity effect that exaggerates the low and low-mid frequencies. Like any condenser mic, controlling the input gain and background noise is key, they will pick up great detail in your voice but they will also pick up a conversation across the house.

  • Electrovoice RE-50b - This is a dynamic, omnidirectional pattern mic that is used commonly for news interviewing. The omni pattern makes the coverage very forgiving allowing for moving around in a big bubble while operating. The pattern also eliminates proximity effect, where a mic loads up and gets too muddy and thick when you speak right into the grill. For my voice, proximity effect is not needed as I have too much bass as it is. This is a very good mic for speech articulation, and the pop filter is about the best I've ever used, no windscreen needed at all with this mic. Excellent mic if you have a bass or baritone voice in particular. The omni pattern does mean it will pick up the room ambience around you.

  • Heil PR-30 (cardioid pattern, large diaphragm dynamic studio mic).  This mic handles plosives better than the MD-421, has less loading via proximity effect and has high output but a bit lower input sensitivity which is handy to help mitigate background noise. The pickup pattern is narrower and more focused than the MD-421, but not so tight as to be awkward to work on the air. The PR-30 is smoother in the upper frequencies I have overtones in my voice at 3k-4k and this mic is smoother in that range where the Sennheiser was a bit gritty for my voice.

  • Sennheiser MD-421 II (cardioid large diaphragm dynamic studio mic). It can be balanced well for my voice and does a fair job at background noise rejection. This mic pretty much requires the use of a foam wind screen (get a high quality one) or a pop filter screen if you work it any closer than maybe 8".  It has a pretty highly exaggerated presence rise that is an advantage for radio but can be gritty on some voices. The hi-cut variable filter is very effective as well but, in my case, in order to drop enough bottom end off the mic, it disturbed the low-mid sound quality. An excellent mic, an industry legend for a reason.

  • Audix OM2 (hypercardioid dynamic vocal mic) - Great in highly noisy locations but you need to work this mic right on top of it and right on center and only an inch or so away for it to sound its best. If you turn your head, you drop way off. Best if you have a very noisy shack situation.

  • Shure Beta 87a - This one has been proving to be pretty good. If you have a husky voice as I do, with a lot of bass and low mid-range, this mic doesn't load up on the bottom like say the Sennheiser, where I have to dump out a lot of bottom with the high pass filter. It's a condenser mic but relatively low output level so it isn't as prone to background noise but still has the crisp definition and presence of a condenser mic. *This mic was a fail ultimately as it was picking up heavy RF on 40m only oddly, so it's back in the stockpile again*

  • Audiotechnica BPHS1 wide range dynamic element headset - I like dynamic mics because they are more easily controlled, less sensitive and don't bring in as much background noise with them. They keep the voice more distinct from other noises in the room. This one is pretty heavily mid and upper mid forward and takes a lot of EQ craft to smooth and balance the sound. It has a built in character that maximizes communication grade audio. For a headset, it's pretty good but needs EQ.

  • CAD M177 (studio condenser mic) - A lower sensitivity compared to many studio condensers that makes it more usable for radio but I still don't use it often because it will pick up a noise from the other side of the house and requires very specific setup. To sound right, a studio condenser mic requires a noise gate and compressor, no matter what anyone says. Has a pretty sterile sound.

  • Various PC headsets like the Gamecomm with a PC electret mic element - These can sound really good and are relatively easy to make an adapter for (they need voltage supplied on the audio pin from the mic jack).

• Outboard processors

  • Symetrix 528e - on my boom mic * RIP, the old stalwart is just slowly dying so I've retired it * I wish they still made them.

  • Aphex Channel - Not worth the money for radio use but it does work. I use this to feed the TS-590sg

  • Joe Meek 3Q (Sounds great but no noise gate so I rarely ever use it now)

• PC equipment and software

  • M-Audio Fast Track pre-amp USB interface (feeds balanced audio into the PC for processing) Only down side is it doesn't really handle line level input very well.

  • Focusrite Scarlett 8i6 3rd gen - USB interface/soundcard. - I've not been impressed with the drivers for this interface. I've had very inconsistent buffer/latency performance even at large buffer sizes at just 48k sample rate. I also had RF issues with this interface, more sensitive than the Motu.

  • Motu M2 - Currently using this one, it's more stable, with far less latency that was immediately noticeable compared to the Focusrite. It's also proven to be more RF resistant for me.

  • Voicemeeter for mixing, Reaper for plugins and processing. - Fun to experiment with but it's a whole 'nother layer of complexity. If you're willing to go through the trouble and deal with the bugs and glitches, this is by far the cheapest way to go deeply down the rabbit hole of advanced processing.

• An Edirol 10 channel compact mixer - My PC, and all my radios feed audio outputs into this mixer, the output of the mixer goes to a small class D amplifier into an old set of Polk RM3000 satellite/subwoofer speakers I've had for years. A nice set of powered speakers really make the RX audio sing on radios. A small line mixer makes it really nice to combine the audio outputs of all your gear in the shack.

The very basics for most hams

The most common mistake people make is not properly adjusting the mic gain level on their rigs. Typically, people have it adjusted far too hot trying to squeeze every last watt from their 100W rig. This comes from a fundamental misunderstanding of how SSB works as a mode.

SSB mode, considering most rigs have average power meters will only indicate about 30W or so while speaking if the mic gain is properly configured. The mistake people make is trying to see 100W output on an average reading (as opposed to a peak reading) meter and they proceed to max the mic gain and processor drive levels to do it. 

The result of this is horrible sounding audio and a very dirty signal that will interfere with other users on adjacent frequencies, not to mention that it stresses the RF deck in the radios.

Step by step setup

• Switch your rig to your dummy load (A dummy load is a 50 Ohm power soak that allows you to test and transmit without bothering other users or emitting noise on the bands. You plug it in as an antenna replacement for tuning up and testing. Every single ham should have a dummy load capable of handling the most power output they plan to generate. A basic 100W or even a 1500W "Cantenna" is not expensive at all and is as basic as a rig, antenna and microphone to operating. Get one. Use it. You'll have it forever.

• Ok, so you are on your dummy load right? You aren't using your antenna and transmitting unnecessary noise on the bands, right?

• Now set your rig for 100W output, find where your mic gain setting is and put your power meter into ALC mode (check your manual).

• Key up your rig, hold the microphone 3-4 inches away, directly on center to your mouth and speak in your normal voice level you will operate with.

• Adjust your mic gain up or down until you see the ALC meter reading about 1/3 to 1/2 of its scale range only on the peaks while you are talking. You shouldn't see any more ALC indication than that.

  •  ALC stands for Automatic Level Control. An oversimplified way to understand this is that it is controlling the maximum level of drive to keep the output of your rig in a linear range. It takes the peaks and levels them, squashing them down and creating an averaged level output. It allows for your loudest peaks to be controlled while also letting your softer words be heard.

  • ALC levels hitting the end of or beyond the range, will make your audio sound harsh and unnatural and with many rigs you can overshoot the ability of the ALC circuit to control your signal and result in splattering and a non-linear output. Aside from RF in the shack issues, this is the top reason for dirty signals on the band.

• If you set your mic gain for proper (within range) ALC deflection, you have already done far more than most on the air towards getting a clean and natural sounding signal.

• The next step is to adjust the speech processor level (proc button). Many rigs, particularly older models just have proc on or off with no adjustment. If the mic gain is set for proper ALC, you should be all set. 

• The speech processor will often narrow up your audio and will further compress it, making your speech louder and more penetrating which is useful when there are noisy conditions or your signal is weak to the other station. Processing will help to get you heard though it's not the most pleasant sound to listen to.

  • If your rig has an adjustable proc level, you want to set the level again as you are talking, for proper ALC deflection.

  • If your newer rig has proc In and proc Out levels, you will want to use your TX Monitor function and some headphones and adjust the proc In level first until you can just hear it change the "color" of your audio. It will start to sound boxy or hard. Just when you hear this begin, stop there.  For proc Out level, that is the one you set to then drive the resulting signal until ALC is showing proper 1/3-1/2 range deflection on the peaks as you speak in a normal tone of voice.

    • Proc In is how much signal is being bumped against the "limiter" of the processor, this narrows the difference between the softest and loudest voice peaks.  Too much proc In level will make it sound very harsh, drive a lot of room background noise in and gives that horrible Darth Vader sound that people have.

    • Proc Out takes the result of that and then drives it out of the rig on transmit.

    • You want fairly minimal Proc In to help boost the sharpness of your audio and then Proc Out set with enough drive for proper ALC indication.

If you made it this far you should be in very good shape for basic operation and sound pretty good. The ham radio world would be a better place if everyone just took a few minutes to understand this setup process.

A further comment on ALC. Some rigs will start to splatter with even a very small amount of ALC being indicated. I suspect poor quality control and improper bias voltages on pre-drivers and final transistors is to blame. Even if you follow the instructions in the manual, you can still have some IMD issues. The FT-991 series from Yaesu is an example of a rig that gets really dirty fast if you are hitting the ALC hard. The Icom 7300 in contrast, you have to be just smashing your ALC to have significant issues with IMD. Icom did a great job with the 7300.

While monitoring on a second receiver you will notice that each rig has a point with the mic gain where you can't hear yourself getting any louder as you increase the mic gain. You should stop increasing mic gain right about where you hear this threshold cross. This generally will prevent IMD issues, regardless of what your rig's ALC meter is telling you. On my 590 as an example, mic gain at 35 (with my external audio chain driving) is where this threshold exists. I can turn it to 99 without gaining any improvement over the air but it drastically increases IMD and splatter up and down the band from the actual transmitted audio.

Further complications...and improvements

Rigs started to have adjustable TX bandwidth back in the 90s or so. It was mostly the top end rigs that began to add the feature and now, even the entry level rigs have menu adjustments to set this. TX Bandwidth and also in turn RX Bandwidth controls determine the range of audio frequencies that you will transmit or receive.  

This is the overall width of your SSB signal on the spectrum and can make a huge difference in the characteristic sound that you transmit or hear while operating.

You will have to consult your manual to figure out if this is a feature your radio has and how to configure it.

I will use my Kenwood TS-590SG as an example. In my menus I have a high pass (this is the lowest speech frequency it will transmit before it is filtered out) and a low pass setting (this is the highest frequency of speech that it will transmit). I can set the high pass as low as 10Hz which will pass through a lot of rumbly bass output and as high as 3kHz which has the definition or "shine" on the sound of speech.  

I recommend that you set your high pass to 100Hz to avoid puffing sounds and wasted energy in your signal. Too much bass will make you sound muddy, waste transmit power and make you harder to understand. It is another layer of complexity to try and EQ and process the rumbly bass below 100Hz.

Properly managed, a low bottom end gives energy and a presence to your audio but you typically only want to spend transmit power there in strong signal and quiet band conditions and it's one of the trickiest things to get right. If you get it wrong you sound like you are talking through a pillow.

On the low pass (high frequency limit), in typical conditions you want this to be about 3kHz. In crowded bands like 20m you may want to turn it back to 2.8, 2.6 or even 2.4 to help punch your signal through. There are rigs that can do 4kHz and beyond, but the wider you are, the more distance you need from any adjacent conversations or you will interfere with them.  The higher frequencies add articulation, clarity and a more natural and open sound quality.

Unfortunately a lot of people are running 4 and even 5kHz wide while only being 3kHz away from other people having a conversation. They are actively interfering with the communications of others by overlapping their transmissions in the passband of the other users. If you take one thing away from this, please understand this concept and have some courtesy for other operators on the bands when you choose your operating parameters.

Wider sounds more natural when properly balanced, but it's also less efficient so it requires more power to get the same level of effectiveness from your signal. Go narrower to leave more room and to make a punchier and more effective signal if you have low power for the noise level of the propagation or band conditions.

Some rigs have very advanced equalizer capabilities now as well. Some have basic bass and treble controls that are fixed frequency (IC-7300 as an example), some have graphic EQs that you use software on a PC to configure, like the ARCP 590G software Kenwood provides and some have 11 band graphic EQs built into the rig itself. There are even rigs like several from Yaesu (991 as an example) that have parametric EQs in them.

To sum this section up

• Set your high pass to 100Hz - I say this because without other processing, most rigs don't handle low frequency EQ and dynamics well, below 100 Hz. It will make your life simpler.

• Set your low pass to 3kHz for general operation, or as close to it as you can get - It's the higher frequency response that makes your audio sound clearer and more natural.

• If you want punchier audio for noisy conditions, set your high pass to 200 or even 300Hz but you will sound very "low-fi" like an old telephone at these settings. Or enable your "proc" which generally does this for you.

• The wider you are, the more the power is spread out, you may need to re-adjust mic gain to bring up ALC if you widen up.  If you narrow up, you may not need as much mic gain. Always re-check your ALC deflection after making bandwidth or EQ changes.

• When applying an EQ, always begin by reducing frequencies to see if you can get the results you want, before you start boosting. Reducing will keep your signal cleaner, where boosting will introduce noise and possibly distortion to your signal. I like this quote "with an EQ, when you are cutting frequencies you are cleaning up or improving the original signal, when you are boosting the signal you are changing how the signal sounds."

• To hear yourself "over the air" some rigs have a TX monitor you can use with headphones but they aren't generally a very good representation. Ideally you transmit into a dummy load and monitor on another receiver. 

• If you have or can borrow an SDR receiver they are very handy to have for this. The SDRPlay RSP series are an amazing tool to have for testing and a lot of fun as receivers in general. You just need a small antenna plugged into the SDR to pick your signal up from the dummy load generally.  For their capability they are an incredible bargain.

• You can't hear yourself accurately until you record yourself and play it back because your own head will dampen bass response while you are speaking and you will almost certainly turn up the bass frequencies too high to compensate.

My favorite system to monitor myself

I put the radio I'm setting up on a multi-port switch.

I set the output of the switch to my dummy load.

I put the receiver I'm monitoring myself with on one of the grounded switch ports. This will give a good S9+20 signal to monitor with. My preference is an SDR Play using SDRConsole with the RX EQ disabled. SDRConsole allows for very easy recording as well.

Always take care that the receiver you are using has a wide enough, unconstrained frequency filter to hear the full signal you are testing. Disable all RX EQ and any noise reduction or other filters on the receiver you are using to test. You may need to experiment with the AGC on the receiver as well to be sure you are hearing things as they are.

Some audio file samples

A full range recording before any filtering is applied 

A 100-3000Hz bandpass filter, typical for SSB

300-2400 band-pass with heavy compression for DX 

You should notice the difference in low frequency and high frequency response. These files are not recorded after transmit, they are recorded on my PC. In transmit, through the radio, applying these filters will have a more audible and obvious effect.

Misc issues, other things to know

As you make your station more complex, you are without a doubt inviting more issues. Consider the notes here before you start adding complexity. As I've mentioned a bit, this is something that should be fun or interesting to pursue or it will result in a lot of possible frustrations.

I recently had the specter of RF in the shack pop up. In my case it was just a single band, either 40m with one antenna configuration or 160m with another. There are a number of causes for this, antenna imbalances (my likely cause using balanced feeder), equipment issues, grounding issues, improperly seated connectors, common mode currents, running power with the antenna too close to the shack (sitting in the RF field) as examples.

My fix in this case was to locate where the RF was getting in. In my case I changed microphones and the issue was exaggerated so it occurred to me to swap in other mics to narrow this down. After I realized it was one of my microphones being more sensitive to 160m RF, I decided to apply a few things to mitigate the issue.

First I picked up an assortment of FairRite snap on ferrites from Mouser. I picked Mix 31 compound to be most effective with 160m, some quite large snap on beads and some smaller. This way I could isolate my XLR audio cable right at the microphone and also the smaller ones for my headphones and other audio wires. I put 5 turns of the mic cable through the large toroid right at the back of the mic. I also added a Jensen ISO Max MI-XX isolation transformer placed right at the input of my Symetrix 528e processor on the mic input line. This damped the RFI enough to solve my issue and should serve me well as I change my configurations around. 

When troubleshooting RF in the shack issues, start by disconnecting everything you can, just a rig, a hand mic and an antenna/tuner.  This eliminates more possible entry points. If this works, start to add back your connections, CAT control, PTT links to amplifiers, audio outputs, external microphones, audio chain etc etc.  Check and re-check across multiple bands until you find when the RF issue appears again. Generally whatever you last added is the issue.

Before you go too crazy tearing apart the shack, do a connector inspection first.  Spin and re-seat any RCA, TRS, mic and coax connectors. Be sure they are all seated and making clean contact. Often a bit of corrosion from sitting or things getting tugged on will loosen things and this can easily introduce RFI issues.

You can address this by mitigating and filtering as I did or by correcting the actual cause which is generally an antenna system issue. The latter is preferred as ideally you don't have RF in the shack at all but sometimes it's simpler to use a band-aid fix or maybe you can't relocate your antenna.

Another tip with RF in the shack. If you run powered speakers, headphone amps and other complications, you will at some point get RF into all of that. It makes a big difference how you route your cabling. Visualize from left to right, all your receive audio gear on the far left, the transmitting RF emitting gear in the center and antennas and amplifers on the opposite side. This helps to keep higher power RF away from the receiving audio equipment. 

Keeping RF, audio and power cables somewhat bundled and separated will help you quite a bit. When it's all tangled in a spider's web, you are inviting issues. With all the tinkering I do, I have to take time about three or four times a year to unplug and re-route it all after things get all tangled from plugging and un plugging as I mess around.

More advanced use of processing concepts and gear...to be continued.

I'm running an Apache Labs Anan 7000DLE MKII SDR transceiver as well as a SunSDR2 DX from Expert Electronics. In both cases my audio is routed as mic -> outboard pre-amp processors -> USB sound interface (Motu M2) -> Windows 10 or Ubuntu -> SDR software via the ASIO audio or ALSA (linux) transport layer.

Both of these SDR rigs have a complete suite of processing tools, EQ, gate, compressors, clippers. They have everything you need to get almost any microphone dialed in to the best it can be. The two rigs differ in the tools, but they can achieve similar results in the end.

Using Voicemeeter Potato beta release with the built-in processing

A recent (11/2022) change I've been trying with great results is running the Voicemeeter Potato beta release stand-alone, no DAW involved. They added a six band parametric EQ, a compressor, a gate, a denoiser filter, a nice reverb and delay plugin. The only missing link is a de-esser. I'm very pleased with how well it all works.

The advantage is that it's just one less layer of software audio to glitch.

At this point, I run virtual audio through Voicemeeter for the Anan, the Sun and the TS-590sg as my normal daily routine.

Using Voicemeeter Potato and the Reaper DAW with plugins

So I got bored again and decided to give it another try running a DAW for audio processing.

A DAW or digital audio workstation is studio recording software that runs on a PC or Mac and takes the input typically of a USB soundcard interface (A Motu M2 in my case) and records and alters tracks for mixing music typically.

I use Voicemeeter Potato, which is a digital mixer. The point of Voicemeeter is to take my input devices, the signal from my Motu and mix and route the audio for that device to things like my radios or for use with chat and video conferencing apps. Voicemeeter also serves as a handful of virtual audio cables that allow interconnecting audio between applications. Running digi modes with fldigi as an example, I tie one of the busses in Voicemeeter to the SDR rig I'm running for sound input and output.

Once the audio is now into the PC and routable, the next component is the processing to apply. In my case I'm running the Reaper DAW, there are many others to look at like Audio Mulch, Pro Tools, VST Host etc. Most of the DAW packages include a number of VST or other type "plugins" and these plugins are basically the software equivalent of rack gear.

The complexity and learning curve of getting all of this installed, configured, tweaked and stabilized (relatively) is a bit daunting and it will go wonky fairly often requiring a dance of shut down and bring up of the software and generally gritting teeth until all the sample rates settle in and stabilize.

The major advantage is, this offers limitless processing capabilities and for vastly less money than using hardware rack gear.  I think, all in, Voicemeeter, Reaper and a few added plugins I paid for, I'm in for maybe $200.  It would be a few grand in physical gear to equal the capability I have.

A few notable plugins I'm using. 

• Waves NS1, it's a noise reducer, not a gate, but a filter that removes background room boom and fan noise. It's a bit smoother sounding than a gate. 

• Standard Clip from SIR - It's a mastering clipper, basically makes the final mix louder  I found this unnecessary with the Reaper multi-band comp called Xcomp as an effective substitute with the Reaper peak limiter in place.

• Stereo Tool VST plugin - It's a suite of functions, probably handier for people without as much understanding of individual components, but I found it redundant with the rest I have available.

• Reaper VST plugin collection - Reaper ships with a complete array of plugins and I'm using a pile of them. The multi-band EQ, the reverb, the multi-band compressor and the hard limiter.

Pros

• Infinite flexibility

• Results you just can't get otherwise unless you have a wall of studio rack processing gear

• Bang for the buck is very strong

Cons

• Complexity, when something is acting up, you've just piled a bunch of layers into the mix

• Instability - low latency audio processing is very unforgiving of other things running on your system. When apps that use audio come and go, they can "upset" the muxed connections and it can be irritating to get it all to settle down again

• Technical ability needed - It's a learning curve to learn about sample rates, buffer sizes, OS tweaks etc. 

Summary

I recommend doing this if you are patient and enjoy tinkering with software, but if you are not, just find another way.

A cool alternative

The Behringer Xair and the Midas sister to it is a hardware appliance that has digital processing built into it. This takes your PC out of the loop and the software/firmware runs more stably inside the hardware appliance.  If I get mad enough at my software DAW setup, I'll be trying one of the Xair 18 or Midas 18 models as my next experiment.

EQ some tips to make sense of it and how to learn to use one.

Equalizing audio takes time and patience to "master", there is no formula to follow and it has many variables.

• Your voice. You run what you brung and with all sound processing, "you can't create what isn't there" is an important thing to remember.

• Your microphone and how you work it, how close you speak to it, are you on center, how loud you speak. All of this can have major impacts.

• The processing tools you have available, outboard, your rig, rig control software etc. So many options.

• The rig itself, op amps, how you feed the audio, impedance matching, level matching. Each rig varies and a setup that sounds great going into one rig will likely require some tweaking for a different model.

Different types of EQ

• Graphic EQ - These are the EQs with the fixed frequency sliders or knobs. They can be 3 band, 5 band, 11 up to 31 band for 1/3 octave splits. Each slider is a filter of a fixed width, generally designed to interact with the adjacent frequencies. The sliders are gain controls, up for boosting, down to reduce.

• Parametric EQ - Found in outboard gear, software DAW processing and notably in recent vintage Yaesu rigs. These are the most powerful and flexible but are also the most confusing and take more practice to apply. 

  • Each band of a parametric EQ has three factors, the center frequency of the filter, the width/shape of the filter (is it a notch, narrow or is it a wide and smooth shape), and the gain for boost or cut. 

  • When you pick the frequency it's similar to choosing which slider to use on a graphic EQ. 

  • When you pick the width or "Q" you are determining if it's a narrow notch, single frequency or if it's wider as if you moved multiple sliders on a graphic EQ. Narrowest is single slider, slightly wider would make a "V" of one slider and the ones on either side, soft and very wide would create an arc centered on a frequency with a rounded shape including multiple sliders on either side.

How you learn where to start with an EQ

You need a way to listen to yourself. The worst option is the "monitor" function in most radios using headphones plugged into the jack. It is better than nothing but they aren't known to be very accurate. A much better option is a second receiver. Transmit into a dummy load through a switch, use one of the other non selected switch ports to feed a second receiver and listen with headphones on that receiver. You only need a few watts to get a good 20 over 9 signal generally.

I use a convoluted method of a dummy load with an RF coupler on the line to the load and feed the output of that coupler into an SDR receiver and use software that allows me to not only hear myself but record and play back as well.

A critical part to understand is that the receiver you use has a receive filter and you need to understand how wide that filter is, if your speakers or headphones reproduce things well etc. Many receivers do not open up down to 0Hz so you can have unintended spikes in low frequency and not realize it if testing with the wrong receiver or filter width. The same goes with higher frequencies. Icom rigs as an example are very limited, 100Hz to 2.9kHz is about all they allow in some cases so you won't have a good idea of anything that is sent below or above those limits.

Take your EQ and set it to zero all the way across.

Now while talking, one frequency at a time, take the frequency and boost the heck out of it. Take notice of where in your speech range that frequency has an effect and notice what that effect is.  Do this one frequency at a time, returning each one to zero as you go through and learn what part of your speech each frequency range affects.

What you want to listen for are "hot spots", there are frequencies that for each voice + mic + rig will be too hot and they "ring" when you find them by boosting the EQ. "Ringing" means they are much hotter when boosted than other ranges. Generally a hot spot is a problem frequency that is out of balance and could use some cutting or attenuation.  Hot spots make you sound unnatural.

Mud muddy muddier muddiest, you sound like you are talking through a pillow.

The number one issue are "mud frequencies", they sap efficiency and obscure clarity and make it hard to understand people. The cure for mud is choosing a high pass/low cut frequency that works for you. Typically I recommend starting with 100Hz for high pass to make your life simpler. Rolling off below 100Hz solves a lot of issues for people and makes your signal more power efficient as well.

Below 100Hz is rumble, vibration, an anchor energy.  A kick drum beat in music is 63-80Hz the punch you feel from it. For voice it's the semi-artificial Howard Stern DJ rumble range.  It uses a lot of power and most hams do not have speakers sufficient to hear it. You can wind up with far too much here because what you listen with or others aren't actually picking up how much energy you are wasting here.

The low-mid 100-200Hz depending on your voice is a back of throat, guttural sound like a low pitched "ooooooh" "new" "you", it's a warmth frequency but this is the most common range 120Hz-160Hz where most people need to cut quite a lot for radio use, like -5db and deeper. If you are too hot here, you will sound mushy.

You almost never want to boost the bottom end for radio. If your voice doesn't have bottom end, boosting there will just pull in room rumble and waste power. You almost always want to cut these frequencies, which in turn will make your signal sound like you boosted the highs.

If people say you need more highs, it almost always means you need to cut the low-mid a bit more.

Now if you find where the low-mid needs to be cut for your voice, you can bet that the next harmonic (if you find it's 140Hz, roughly 280Hz is the next harmonic) can use a dip too. I generally find that if you drop say 6db at 140, that 280 will be a bit less, like 4db-5db.  This range is warmth to boxiness. If you say the words "wuh" "muh" "wah" it's that lower part of the mid and start of boxiness that you hear. It makes things sound unnecessarily "thick".

All these ranges vary with voices, mics, rigs etc. there is no magic formula and you really can't share settings with anyone unless all conditions are identical for mic, rig and then the voices will vary between ops.

The next range is the harder boxiness range. 400Hz to about 800 or 900Hz. The words "box" "ox" "octo" "ock ock ock". This range gets aggravated with speech processors commonly or if you are hitting the ALC too hard. It just creates an unpleasant character. You want to beware that if you dip this too much, you will lose some important energy for understanding words. Slight dips here to level things are typical, like -3db or so and only if you need them.

This boxy range can be detected when esses sound like you are saying "ex", if it's too hot here it will interfere with hearing "x" vs "s" or if you say "x-ray" where the "ex" sound is exaggerated.

Up from there is the hard edge of the top of mid-range, the 1.2 or 1.4kHz area if you make sounds like "ink" "ank" "tank" it's the harder edge of those and similar to the last they tend to get hot with heavy compression, too hot mic gain, heavy processing.  Go easy here as this is the meat of articulation, but too hot here makes you sound like you are on an old telephone.

The upper mid to what are the "highs" for radio 2-3kHz are sounds like "eek" while pinching it in your throat "squeak", to exaggerate the eeeee part is about 2.5kHz. This puts a metallic harshness if boosted to far but this is the area that helps cut through noisy band condx too so cut here only in odd conditions.  Typically this area, a filter centered at 2.5-3kHz depending with a slight boost of 3-4db will put some "shine" on audio for most people.

99% of the time, if you clean up your high pass to 100Hz, the low-mid just above that and the next harmonic and you get that just right, people will start to comment about how nice your audio sounds. 

The Icom IC-7300 has the easiest setup of any rig I know of. Using the stock hand mic, set TBW to WID, set Bass to -2, Treble to +3, Compressor to 1 and mic gain around 30-40 depending on how loud you speak and you are just good to go. The simplest rig with the nicest baseline audio that has been put on the market.