• 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

On Air Distorition

  • Thread starter janedoahlovesbrentlee
  • Start date

J

janedoahlovesbrentlee

Guest
We have an optimod 8100A ( I think? Its the last big blue one before you get into the series with latency issues). We have a CHR with a direct format competitor in market. So, the processing has to be aggressive. The problem is that if the jock isn't watching the levels on the console, you'll hear distortion on the air. Other wise, if they keep it below 0, it sounds very clean, and loud. Our engineer is a RF guy, and not a sound guy, so he's out of ideas. Is there something in the optimod I can change to fix that, or is there a unit I can install inline to correct the problem? A compressor or something to stop the problem before it gets to the optimod?

Also, I have a Gentner TS-612 and a station I worked at before had a "night ring" with this system. When the station went completely automated, the legal ID at the beginning of the first hour of automation had a PIP that fired either the switcher or a relay. When this would happen the incoming studio lines would get rolled over to the in-house phone voicemail. Then when the station went live again there was another PIP that turned the phones back on. How do I make that happen?
 
It sounds to me as if you're running out of headroom in some device between the console and the processor... and the processor you have is the 8200, not the 8100.

I would do an end to end from the console to the input of the processor and check each piece of gear that passes audio to make sure it is not being overdriven. Take the station OFF THE AIR on an overnight to do this and listen to the output of each device. Take the time and your station will sound better.

As for the Gentner, I wouldn't know because I have never worked with this unit. There is documentation buried on the Gentner site, or what has become of Gentner site for some of their products.
 
Thanks for the help! What if the first device in line is the optimod? Or if it's something else, how do I pad it down?
 
Check input levels to the 8200, and recalibrate the input of the unit if necessary. I have used a compellor with the 8200 before to protect the STL from being overdriven. Putting one in would possibly help you here also help to maintain the levels if you are trying to compete with someone else in town.
 
Actually, you may have an 8100, we had a blue one at one station. It was provided by BE, who either repainted the front panel or had Orban paint them blue. Perhaps Mr. Orban can help us as to who turned them blue?
Anyway, if it's got analog meters on the front, it's an 8100. If it has a jog wheel and LCD display, it's an 8200. Which one ya got?
 
Yeah it's the 8200. The 8100 is our backup one. I got them confused.
 
The 8200 has a built in Auto Gain Control feature that automatically adjusts the volume before it enters the processing stage. If you have defeated the AGC, then the incoming audio enters the Optimod too hot if the jocks are pegging the meters on the console.

R
 
I totally agree with taking the time and checking EVERYTHING. The problem could be well before the Optimod. Start by checking the output of the console - if its the console causing the problem then there is nothing the Optimod can do if its getting distortion.

What console are you using?
 
>Yeah it's the 8200. The 8100 is our backup one. I got them confused.

OK. I defer to the 8200 gurus. We've one in backup service at one plant, it is very seldom used, so I know only the basics about it.
 
littlejohn said:
>Yeah it's the 8200. The 8100 is our backup one. I got them confused.

OK. I defer to the 8200 gurus. We've one in backup service at one plant, it is very seldom used, so I know only the basics about it.

OK is your seldom used backup plant the one giving you trouble? If not, after you check everything else and all of it checks out OK, you might swap the two 8200's around and see what happens.

R
 
janedoahlovesbrentlee said:
We have an optimod 8100A ( I think? Its the last big blue one before you get into the series with latency issues). We have a CHR with a direct format competitor in market. So, the processing has to be aggressive. The problem is that if the jock isn't watching the levels on the console, you'll hear distortion on the air. Other wise, if they keep it below 0, it sounds very clean, and loud. Our engineer is a RF guy, and not a sound guy, so he's out of ideas. Is there something in the optimod I can change to fix that, or is there a unit I can install inline to correct the problem? A compressor or something to stop the problem before it gets to the optimod?

Also, I have a Gentner TS-612 and a station I worked at before had a "night ring" with this system. When the station went completely automated, the legal ID at the beginning of the first hour of automation had a PIP that fired either the switcher or a relay. When this would happen the incoming studio lines would get rolled over to the in-house phone voicemail. Then when the station went live again there was another PIP that turned the phones back on. How do I make that happen?

Since we made it clear it is an 8200, check your IO settings.....Your clipping input should be referenced to +4dbu.....NOT -10 or something way off....Turn on your AGC and set it to a 10-20db......watch your INPUT levels on the meters and make sure it stays below max....(clipping)...Then set your AGC to a medium fast setting (about 1-3 sec release)..

Watch your output level driving the STL...do NOT overmod on RF OR T1.......it will cause clipping (personally I think the best audio out of an 8200 is done with the unit at the tower site and using the composite out to the transmitter. You can get loud and cleaner that way....or if you use the 8200 at the studio driving a T1 STL, you BETTER have a good clipper at the transmitter site and allow it to run the mod UP to 101+ % on peaks..but make sure you dont overdrive the T1....
I have heard this setup sound good ONCE (KZPS in Dallas on its old setup) but I think the engineer there was heavily clipping in the exciter....otherwise, he was often accused of overmodulating......but the meters never showed that.

As for switching the phones, that depends on what phone system you have....a relay out of any automation system could do the job...but some phone systems are hard to do it with than others.....What is your phone system/voicemail make? How many lines do you have ringing to the CR you want going to the VM?
 
janedoahlovesbrentlee said:
We have an optimod 8100A ( I think? Its the last big blue one before you get into the series with latency issues). We have a CHR with a direct format competitor in market. So, the processing has to be aggressive. The problem is that if the jock isn't watching the levels on the console, you'll hear distortion on the air. Other wise, if they keep it below 0, it sounds very clean, and loud. Our engineer is a RF guy, and not a sound guy, so he's out of ideas. Is there something in the optimod I can change to fix that, or is there a unit I can install inline to correct the problem? A compressor or something to stop the problem before it gets to the optimod?
<snip>

The first step in eliminating distortion is isolating it, that is, find out where it is occurring.

Unless you have a broken piece of gear, chances are that the overload is just mis-aligned levels, where one piece of equipment's output is hotter than needed by the following piece, whose input overloads.

If this is the case it's really easy to fix, especially if you have a 'scope (any production or audio guy worth his soap should know how to do this anyway, as a matter of good engineering practice)

The idea is to align the clipping points of all equipment in the complete chain, from console out to processor in, using a test tone. This way they all overload at the same point, and no single piece is the weak link in the chain. You can then be assured that the signal-to-noise has been set to be as good as that particular equipment combination will allow.

You'll have to take the station off the air , or at least subject your listeners to test tones, which my not be a good idea during afternoon drive. ;-) I myself prefer late nights with lots of coffee.

You will need a tone generator, and a means to monitor: either an oscilloscope, a set of powered speakers with bridging inputs or, better yet, both. A calibrated audio meter will come in handy, but you may not need it.

Connect your monitor (speakers, scope, etc.) directly paralleled to the console output, leaving the console's output still connected to the following equipment.
Send a sine wave (1 kHz will do) from the tone generator into a mono console fader at normal operating level. Set the channel balance (if available) so both left and right are matched in level. Monitoring the audio at the console's output, turn up the fader level until you can just barely hear (or see, with a scope) the tone clipping [at this point the console meters may be pegging. Ignore them :) ]. If it does not clip, send more level from the generator. Just be careful that the generator itself is not clipping.

If you can measure the console output level with a meter, the difference in dB between "operating level" and "clipping level" is your "headroom." It should be, at a bare minimum, 15 dB; 20 or more if you can get it. Headroom is your 'insurance' against casual overload by sloppy board operators.

If you have a calibrated audio meter, you can measure the actual peak level where clipping occurs. In a 'standard' professional analog audio environment, if the overload point is not well over +20 dBm, you may have found your problem. :)

The exact level where the audio clips will probably be fairly audible, and visible on the waveform of the 'scope if you are using one. Back off the fader level very slightly until it sounds clean again, and leave it there. If the console has clipping indicator LEDs, now would be a good time to calibrate them. If there are master output level controls on the console, set them to the identical output level.

You should observe whether one channel clips at a noticeably different level than the other. This would indicate a problem in the lower-clipping channel, and should be addressed.

Now, hang your monitor/speaker/scope on the NEXT piece's OUTPUT, and (leaving the console levels exactly where they were), reduce or increase that next piece's own INPUT controls until its output is also just below clipping (with some pieces of equipment it may be necessary to do a similar stage-to-stage level adjustment to prevent clipping internally). Once again, be careful to keep both channels at the same level. The idea is to prevent clipping inside that piece of gear, distortion that would be passed onward, while keeping levels as high as possible.

You may note that the clipping levels for the input and outputs of any particular piece of equipment should be matched as closely as you can get them. Do this by initially reducing output level well below its amplifier's clipping point so you can calibrate the input level; once the input levels are calibrated, do the same with the output controls by raising them to just below clipping.

Note that some equipment has amplification -- which may overload-- BEFORE its input level controls. If reducing the input control does not reduce the clipping, you may need to turn down the OUTPUT of the PREVIOUS piece driving that unit.


If your setup sends console output directly to the processor, your job is done. Otherwise, repeat the procedure with every piece of gear all the way through the entire system (leaving the generator sending level to the console input where you originally set it, just below overload), and you will have optimized inter-equipment levels and, hence, signal-to-noise, for the entire chain.

BTW, when you send audio to a dynamic system such as a processor, you may find setting input clipping level to be somewhat tricky, as the processor will counter your increases. Your processor's manual may give you specific advice concerning this. But there should still be an audibly clear point where the audio hits clipping, and you just back down from there, using its input level controls, or the previous equipment's output controls, whichever is appropriate.


I know this procedure sounds lengthy. Hopefully I haven't forgotten anything vital. ;-) But it's pretty simple once you've done it a time or two, and it will give you a peace of mind to know your setup is working at its best.

Kind Regards
David Reaves
 
David Reaves said:
You'll have to take the station off the air , or at least subject your listeners to test tones, which my not be a good idea during afternoon drive. ;-) I myself prefer late nights with lots of coffee.

I had to laugh when I read David's remark. :) It reminds me of whenever the station I work for, has to do proof of performance testing. We usually do the testing in the very late evening and overnight hours. But chances are you'll still have listeners at that time. Which begs the question of should you announce this event beforehand?

I can see it all now...

"Coming up next, we're gonna broadcast a bunch of tones. If you wanna hear a specific tone, call the request line right now!"

;D

R
 
its the all request hour with tones back to back in all in a row - COMMERCIAL FREE.

I rolled when I read you post - if any programmers read this thread it wouldn't surprise me if they tried it. :D
 
... "And this one's going out to Bob and Marylou, by request ... fifteen-hundren cycles!"

<beeeeeeeep>

or with the Bill Drake style, "And the tones just keep on comin' !"

I wish I thought of doing this during proofs or when I did overnight tests for airchecks by the NRC
 
Honestly, I don't think I'd go through a big procedure with an oscilloscope. Assuming your equipment is recent vintage broadcast gear that is correctly connected, it's very likely in the 8200.

The first thing to do is check the Input AGC meter. (I think that's what it's called.) It shouldn't read much over 10 with the console wide open. If it reads higher...that's your problem. If it's too high, go through the "Quick setup" and follow the on-screen instructions. It will tell you to run your console wide open, then set the input levels for "10"

Once setup is correctly done, switch to one of Bob Orban's settings. It won't be loud, but you are trying to determine of the distortion is coming from the processor or something else in the plant. This won't completely settle the issue, for instance, some exciters sound bad when you push them hard. But it's a start.

You can hook up an amplifier on the output of the 8200 and listen with headphones. Or you can run the 8200 into your modulation monitor and monitor off of that. If the 8200 sounds great until you connect it to the exciter...then you just found the problem.

And finally, remember that an 8200 was never that great. You can't compete with an Omnia 6 or an Orban 8400 or 8500 with it. By the time you get the loudness you want, it sounds kinda distorted. That's just an 8200 for you.

If that's the case, find yourself a composite clipper, like the Modulation Sciences CP-803. That will allow you to increase your loudness without driving the 8200 so hard.
 
David Reaves said:
I had to laugh when I read David's remark. :) It reminds me of whenever the station I work for, has to do proof of performance testing. We usually do the testing in the very late evening and overnight hours. But chances are you'll still have listeners at that time. Which begs the question of should you announce this event beforehand?

I can see it all now...

"Coming up next, we're gonna broadcast a bunch of tones. If you wanna hear a specific tone, call the request line right now!"

;D

R

Not too far off the mark. I used to engineer for a 500W east coast daytimer on a Canadian Clear channel and when we'd come on during the "experimental period" (remember that?) in the wee hours, we'd get excited DX'er calls on the studio request lines and cassette airchecks from half way across the country with our test tones, legal ID's and bits of tunes we played to set-up audio processing. If you're on someone will be tuned in!

Don't expect to see much useful with sine waves into an O-scope connected to a processor. First off even when set up properly the sine waves are going to be hammered, secondly sine waves don't come close to the peaks and assymetry of real world audio. This may be a good case of try to isolate the problem and trust your ears. Got a tune that consistently sounds like cr_p? Then wait till late night, load it into the automation and let it repeat while you tap the audio at various points with your "Bit Buddy", monitor amp and favorite headphones.
 
greg.hahn said:
Honestly, I don't think I'd go through a big procedure with an oscilloscope. Assuming your equipment is recent vintage broadcast gear that is correctly connected, it's very likely in the 8200.


TowerLamp said:
Don't expect to see much useful with sine waves into an O-scope connected to a processor. First off even when set up properly the sine waves are going to be hammered, secondly sine waves don't come close to the peaks and assymetry of real world audio. This may be a good case of try to isolate the problem and trust your ears. Got a tune that consistently sounds like cr_p? Then wait till late night, load it into the automation and let it repeat while you tap the audio at various points with your "Bit Buddy", monitor amp and favorite headphones.

After I posted it, I realized that maybe I should not have suggested the option of using a scope. A lot of people don't have them these days, and fewer know how to use them. But then, I also said you could just as well use those well-trained ears. :)

But seriously, checking out and aligning for maximum system headroom doesn't take that long (especially if your 'system' consists of only a console and a processor)!

As for using tones rather than music, there is no purer source to expose distorting equipment than a clean sine wave exactly at peak level! And keep in mind, I am not talking about what goes on inside the processor, rather the (hopefully) linear interconnections between the equipment.

You HAVE to take care of what's coming into it before even beginning to look at the processor!

Kind Regards,
David
 
Actually, I'd rather use squarewaves to quickly test a system, as long as it isn't a dynamic system (Set it to 'proof' mode if it has one). The problem with sinewaves is, using a scope you will not see distortion till it hits the 1 - 2 percent area. Most modern stuff is in the like .002% or so range. Going from .002 to .2 won't be apparent unless you got better eyes than me. Yet, it's two powers of ten change. I'm much more interested in, does the system ring, or roll off lows or highs, have really good transient response, and decent frequency response. These are the things people will hear, and consequently are things to look at very critically. I'm not knocking pure tests, I just think you'll get better 'fast' information using a good squarewave. With a dual trace scope, you can superimpose the generator and output waveforms and see what isn't making the program.
 
littlejohn said:
Actually, I'd rather use squarewaves to quickly test a system, as long as it isn't a dynamic system (Set it to 'proof' mode if it has one). The problem with sinewaves is, using a scope you will not see distortion till it hits the 1 - 2 percent area. Most modern stuff is in the like .002% or so range. Going from .002 to .2 won't be apparent unless you got better eyes than me. Yet, it's two powers of ten change. I'm much more interested in, does the system ring, or roll off lows or highs, have really good transient response, and decent frequency response. These are the things people will hear, and consequently are things to look at very critically. I'm not knocking pure tests, I just think you'll get better 'fast' information using a good squarewave. With a dual trace scope, you can superimpose the generator and output waveforms and see what isn't making the program.

If you are trying to determine and align the overload points in an audio chain, you will not hear it (or see it) with a square wave, because it is already as clipped as it can be. :)

Since the level at which a sine wave clips at is usually pretty abrupt, it is ideal for this purpose, using either a scope or an ear or two.

Square waves are the perfect signal for many tests, but I don't consider level setting and alignment to be one of them.

Kind Regards,
David
 
Status
This thread has been closed due to inactivity. You can create a new thread to discuss this topic.


Back
Top Bottom