How much compression do commercial FM stations use?

To me that is kind of a tough one to answer.  Different stations/formats use different processing techniques.

FROM ANOTHER SITE: "Very few radio stations have sound quality as the primary requirment for getting listeners. A rare few will make an effort in that regard, if they have an engineer who is also an audiophile. The effort is put into being the LOUDEST station on the dial, the reason being that as a listener scans by the LOUD station, they'll stop and listen more frequently than the will a softly-playing but higher-quality station. And a fringe-area listener might choose the LOUD station because it will cut better through the noise. So the audio get steamrollered, making the "loudness war" even worse than the source CD. Folks like Optimod, who makes the hardware compressors, do a great job in keeping fidelity intact, but it's ultimately up to the station's engineer to decide how that station should sound. Classical, jazz stations, etc. know that their listeners DO care about sound quality, so they make the fidelity effort that commercial stations don't. Radio is all about having listeners, nothing more than that. I haven't worked in radio for almost 20 years, but I'm sure things haven't changed all that much. And your tuner plays a large part of that, too. Dedicated tuners are a rarity nowdays, the FM section of most receivers is simply an afterthough, like AM used to be. They just don't spend the pennies to get a good sound. But with online streaming, satellites, digital stations, etc. the days of broadcast radio are certainly numbered. One station I worked at was AM stereo. Believe it or not, it acutally sounded pretty good."

In my opinon, LESS is MORE!

Sorry I don't have any numbers data about the observations I am about to share.

Maybe later I can look under the hood and at least find out about the compression/limiting I am doing with my station.

But first let's talk about other stations in the market.

There is a satellite sports station in town at 1490kHz which sounds like the loudest station in town, but the trouble is, it's seriously distorted. It's been distorted everyday since the sports format began a year ago. Before that, when it's programming was local, it had clean audio at a lower level.

A classical group worked out a deal where they are leasing an HD channel to put a shoe in the door as a legitimate broadcaster, which qualifies them to lease an analog 250 Watt translater for classical music. From day one the audio was amazing on the analog... no distortion, loud passages nicely held within safe limits, but especially quiet sections raised smoothly so that they can be heard above traffic noise in a car.

For my overall station audio I run the audio through Stereotools, a beautifully written computer utility that matches professional hardware performance, the work of Hans Van Zutphen in the Netherlands. I am using one of the many included presets, in this case an AM radio preset, which pushes my audio way up into the upper range of the VU meter without overmodulation and with a very natural sound; no distortion.

My voice recordings, which I do daily, are mastered using SampleWrench, a little known but excellent (now free) software that records very clean audio. The master wav file gets edited (my mistakes repaired) in the free Audacity audio software, available from Sourceforge.org, and the last step is to add a preset compression which raises the quieter portion for a louder overall sound without "pumping" or "breathing" artifacts. I spent a long time finding a ratio that I still like after using it for several years.

Today I will look inside the windows and see if I can pick out some numbers to report.

Oh, one more experience... for years I was an audio cassette duplicator, and I tried several hardwares in a search for a reliable way of leveling the audio and was totally pleased by the Aphex Compeller, which I still have in the rack (not being used anymore).

The compression schemes can vary quite a bit depending on the signal path being fed...

KRNI began this thread expressly asking about FM compression, but I am widening the subject by discussing several compression environments.

One fact I can now report is that the compression ratio I use for voice recordings is 1.5:1.  My initial monitoring during editing is done using a Wholehouse 2.0 FM transmitter to a C.Crane Radio Plus, but I am NOT limiting for FM, but for general audio on the recording itself.

My radio playlist output goes through its chain: Winamp Playlist, DSP Stereotools, AM chain output via USB Audio to AMT5000, Stream Encoder to Icecast and Shoutcast Servers...

Since Stereotools is a single utility per computer, I use its North American AM Transmitter Preset, which of course favors the AMT5000 AM transmitter in my chain, but I also find that AM Compression works very well for internet listeners listening on their computers.

The Stereotools Control Panel is an entire universe of options and settings, and I am so-far unable to find a simple description of my compression/limit ratio, but I will read the vast Manual and search for a way of finding out.

Stereotools includes an FM Transmitter section which can determine the pre-emphasis curve (75 us), but if used the FM transmitter's internal preemphasis must be OFF, otherwise there would be double pre-emphasis, which would create trouble.

Stereotools includes a AM Transmitter section in addition to the North American AM Transmitter Preset, and this panel allows adding NRSC AM Preemphasis (also 75 us), but in my case I have the AMT5000 transmitter jumpered for its internal NRSC, and cannot use the Steretools NRSC because it would drive the streaming servers with preemphasis which is not desirable at such a high level (75 us).

Interestingly, for online streaming Stereotools offers a 15 us preemphasis, but I cannot use it because it would increase the NRSC all the way to 90 us, which would exceeed the AM standard and cause modulation problems.

Today I have learned many things, such as, there's more I don't know than I didn't know before.

Hey, I found a mistake that I never would have noticed if it weren't for this thread on Compression!

Because I have Stereotools set for the North American AM Transmitter plugin, it automatically had itself checked as "Downmix Stereo to Mono," which means that my recent change to stereo streaming online was not stereo at all and I didn't realize!

The repair was right there, in the highly intelligent Stereotools panel, I clicked "CQUAM Stereo" and all is well.

For years I'd been sending mono over the internet as a bandwidth savings, but with many high profile symphony series I decided I should present them in stereo, so I over-rode Winamps "Force mono" and also reset the Stream Encoders to put out a stereo stream instead of mono.

Now I think I'm really in stereo! Thanks everybody!

To bring you an important message about Stereotools by Hans.

In a previous thread John WDCX reported Asymetric AM Modulation supported by Stereotools, the plugin for computor audio processing.

My immediate exploration (at that time) of the DSP, VST and Standalone Versions of Stereotools found, it turns out mistakenly, that only the Standalone Version offered asymetric AM modulation. WRONG!

Today's Stereotools review has discovered an ADVANCED user panel that opens up more sophisticated sections, and HEY, the DSP Plugin Version DOES, yes DUZ have Asymetric AM Settings!!

You mean, I've been having 125% modulation and didn't even know it?

This week is starting well.

What other surprises lie in store. Lay ahead. Await.

You are truly blessed.

It's not just the ratio, there's also the threshold, attack and release times. I have a Alesis nanocompressor and found the right combination to get almost to the volume of the typical station. I am using under $100 worth of processing compared to $1000s worth of audio prosessing for the commercial stations. The compression kinda makes it sound cool on the air with a song with a drum intro. and the little reverb effect it creates.

The commercial stations probably use more to get so loud sounding without going into distortion with overmodulation. But they have a whole lot more high end equipment than we use with a part 15 station.

Mark

My opinion is that if you're an audiophile, you're probably not going to be listening to a radio station to satisfy that need, either AM or FM.  Similarly, you're not going to be listening to an Internet stream at a relatively low bitrate.

The most critical aspect of most Part 15 over-the-air stations is range.  Loudness, achieved via compression, will increase the listenable range of your station.

I generally use 4:1 compression.  Given that I broadcast mostly mono vintage Jazz, obscure 50's & 60's pop and 80's new wave music, I'm not worrying too much about the loss of dynamic range.

Yeah sad thing is without compression I'm already getting louder audio than commercial stations. Than when I run it through my behringher peak leveller to get it high as I can without over modulating its as loud at level 6 on my grundig as commercial stations on 11 or 12 volume.

Speaking of that by the way does anyone know if the Insignia FM HD Radio stations also use compression?

Does too much sound like a fair answer?

I know the commercial FM station in my area makes operating a Part 15 FM undesirable (especially as most people have but lower end receivers). It seems that they modulate the carrier of weak stations (even down on shortwave bands in some instances).

I have even heard of them going as far as tracking down Part 15 stations and having a NOUO sent to them... as if they have any potential of reaching more than a few houses away... All the while that same commercial FM causes havoc by riding the carrier of distant stations and weak signals across the FM band... and as mentioned earlier... sometimes out of band.

This occurrence isn't always limited to lower end receivers.
So I don't have the number, nor do I have an answer, but the local commercial FM uses too much. It is hard to hear the local NPR at times. It may be of interest that both the NPR and the commercial FM towers are at nearly equal distances from me, and unless my memory is failing, the NPR is actually running more power than the commercial.

Sorry about the rant.
Too much is too much, and there should be a limit on how much interference a large commercial FM station can cause.
Maybe the commercial FM should be issued a NOUO... Seeing how they have reportedly had them issued to several Part 15 FM operators in the area.

I know the original post is looking for an answer, so I'll sit back now.
My apologies for the disruption.

The reason for compression is to raise the lowest sounds up higher without making the loudest sounds get louder.

My reason for wanting to compress is that I know for a fact that human lsiteners have noise around them in their environment.

Driving has traffic noise, outdoors has aircraft, power lawn equipment, dogs barking.

Indoors has fans, airconditioners, maybe somebody has TV sound truned on.

If recordings are played in their natural state without adding compression everything might be fine, because most pop music is recorded loud from start to finish. But classical music can get very silent and suddenly get very loud.

Limiting is needed to keep the loudest signals from overmodulating. Limiting must be set very exactly so that the top of the volume touches but does not exceed the maximum allowed modulation.

I'm really, really, REALLY glad this subject came up. I've been inside the Stereo Tools Control Panel ever since, learning how to mess things up.

That's the "error" part of trial and error.

Next comes the straightening of the mess, and that's when we finally learn something.

Sorry we haven't heard a report on compression ratios used by commercial FM stations, which was the question at the start of this thread... maybe if we wait awhile longer it will come.

For now I'll abbreviate by saying that today when I turned the station on for the day the audio levels were all goofed up.

So I undid all the changes I'd made yesterday in Stereo Tools (an audio processing software) and got back to where things were before I tinkered.

Then I re-tinkered and figured out why everything changed, and can now take the rest of the day off because I'm so much smarter.

Tonight if things are slow on the site I'll go into more detail than you need to know.

"Sorry we haven't heard a report on compression ratios used by commercial FM stations, which was the question at the start of this thread"

That is because there no standard compression.  Depends on music & time of day, and listeners.