I have been chasing a problem for three years now and I am tired of trying to solve world hunger alone. My streams on both Live365 and Shoutcast sound pretty good to me - wowever, I use a LPFM transmitter for my over the air and in the studio monitors. No matter what I do, on some songs I hear a weird "swish" up around (maybe) 12Khz-ish during some parts of the songs. I have tried (3) different transmitters (from 3 different manufacturers) and even chased the potential pre-emphasis issue (like 75 versus 50) - I am not overly compressed AT ALL. Again, the internet stream sounds FINE but the FM broadcast, on ANY receiver, occasionally has this "swish." Do you think it's overmodulation?
-
Get involved.
We want your input!
Apply for Membership and join the conversations about everything related to broadcasting.
After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
If you use a disposable or false email address, your registration will be rejected.
After your membership is approved, please take a minute to tell us a little bit about yourself.
https://www.radiodiscussions.com/forums/introduce-yourself.1088/
Thanks in advance and have fun!
RadioDiscussions Administrators
novice question
- Thread starter shawncochran
- Start date
Sounds like it might be overmodulation at high frequencies. Reduce the modulation and see if the problem clears.
If you're hearing a splatter sound during certain high frequencies, then as Frank mentioned, it could be your Low Pass Filter in the stereo generator is 'ringing', or you are simply over-modulating. This would be especially true if you're noticing the splatter at 12kHz.
"Swishing" at higher frequencies can also be attributed to what's known as dueling algorithms of audio codecs. This is especially true if you play MP-3 encoded files through a cheap digital limiter (Behringer) or digital audio processor. If you are airing a web stream, then there's no way to know what the form of compression (codec) was used to create that particular song file. The more lossy the compression, the more swishy it will sound when converted to analog.
"Swishing" at higher frequencies can also be attributed to what's known as dueling algorithms of audio codecs. This is especially true if you play MP-3 encoded files through a cheap digital limiter (Behringer) or digital audio processor. If you are airing a web stream, then there's no way to know what the form of compression (codec) was used to create that particular song file. The more lossy the compression, the more swishy it will sound when converted to analog.
Thank you very much. I can reduce the modulation and play with that tomorrow. The output from the compressor/limiter may be the issue here though. While the internet stream is solid, the over-the-air is occasionally 'swishy' and that digital-to-analog conversion makes so much sense to me. Yes, the audio files are MP3s (higher bitrates) but since I am using a cheap limiter, a Behringer in fact, I think you've led me down the "primrose path" so to speak.
Unfortunately there is no benefit to running higher bit rate on an MP3 file. Its just a tightly compressed, lossy file format and doesn't do well when played digital to analog. That's why stations with HD signals, should never play MP3's. The compression algorithms between the two formats cause really weird artifacts.
.wav or AC3 is about the only files that are acceptable for transmission.
So I have been digging since you started helping - buried in my automation software I found a compressor/limiter and AGC that were squeezing the life out of the files to begin with (think mid 80s loudness war on steroids) - I turned that OFF, and am using just the H/W compressor/limiter and I tweaked some settings in there. 98% removal of the occasional 'swish' I was hearing - so there must be a plethora of issues with MP3s being compressed then compressed then broadcast. I am considering coming out of the music PC prior to the LPFM through a piece of software called "stereotool" but I haven't convinced myself there will be any additional improvements.
Audio compression (like in your automation software) and digital file compression (like MP3's) are two completely different things. One has nothing to do with the other.
You need an FM audio processor to feed your transmitter. This will prevent over deviation and the high frequency splattering that you are hearing.
You need an FM audio processor to feed your transmitter. This will prevent over deviation and the high frequency splattering that you are hearing.
Frank makes a good point. There are two different references to "compression" that you're dealing with here. One is the type of file format compression, where a much larger files is compressed using a mathematical algorithm (MP2, MP3, .wav, AC3, etc). The other is audio dynamic range compression, where the peaks of the audio waveform are reduced or compressed to prevent peak energy over-deviating your transmitter.
Compressing larger audio files create artifacts which are noticeable if converted to analog, depending on the compression format.
Compressing larger audio files create artifacts which are noticeable if converted to analog, depending on the compression format.
I should have mentioned that I do have hardware processing in chain before the transmitter. The last item in the chain is a compressor/limiter prior to signal injection. Of course my microphones have compressor/limiters are well (one is a DBX 286s, the other a straight DBX 286) -- I have chased * a lot * of things I did wrong here. I have made a lot of incremental improvements ... including the fact that when I was following *my own stupid wiring diagram!* I noticed I accidentally had one place where balanced and unbalanced were mixed; correct that mistake and there was a pretty hefty improvement!
I put an 8khz test tone through all board inputs and made sure I had consistent balance, that helped too. I also found some input and output lines that benefited greatly from a ferrite bead being added.
I pulled the output from the H/W audio chain to about 95% to avoid 'splatter' ... and I relocated the transmission line to run 90deg to everything else leaving the studio.
Lastly, the post-processed audio was, in my opinion, being overdriven and likely drove some of that splatter.
Now, given ALL MY ERRORS (!) there is ZERO doubt that going lossless on the audio would be, once again, a whole MAGNITUDE of improvement! It seems my self-inflicted wounds are remedied; to go further requires me forking out a chunk for improved music library files.
I have one continuing thought to ask though... my FM signal does not have RDS right now ... I was wondering if using stereotool to create the RDS signal at the PC level would allow it to display on FM tuners (or is there additional H/W required prior to my transmitter) - in other words is the data buried in the signal or H/W injected? I suppose my question would also be applicable to HD broadcasting - is it buried in the signal or is it H/W injected?
Sorry for all the novice questions -- this forum has been A BOON to my knowledge! Thank you so much so far.
I put an 8khz test tone through all board inputs and made sure I had consistent balance, that helped too. I also found some input and output lines that benefited greatly from a ferrite bead being added.
I pulled the output from the H/W audio chain to about 95% to avoid 'splatter' ... and I relocated the transmission line to run 90deg to everything else leaving the studio.
Lastly, the post-processed audio was, in my opinion, being overdriven and likely drove some of that splatter.
Now, given ALL MY ERRORS (!) there is ZERO doubt that going lossless on the audio would be, once again, a whole MAGNITUDE of improvement! It seems my self-inflicted wounds are remedied; to go further requires me forking out a chunk for improved music library files.
I have one continuing thought to ask though... my FM signal does not have RDS right now ... I was wondering if using stereotool to create the RDS signal at the PC level would allow it to display on FM tuners (or is there additional H/W required prior to my transmitter) - in other words is the data buried in the signal or H/W injected? I suppose my question would also be applicable to HD broadcasting - is it buried in the signal or is it H/W injected?
Sorry for all the novice questions -- this forum has been A BOON to my knowledge! Thank you so much so far.
- Status