-
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
tips for processing with a composite stl
- Thread starter whit979
- Start date
Get a composite clipper and put it at the transmitter site. It doesn't have to be aggressively set if you don't want it to, but it should be there anyway. There is a certain amount of overshoot created in the STL which the composite clipper can get rid of.
Don't overmodulate the transmitter.
Be sure your STL is working properly.
That's about it. When it's working right it's kinda nice having the controls back at the studio.
Don't overmodulate the transmitter.
Be sure your STL is working properly.
That's about it. When it's working right it's kinda nice having the controls back at the studio.
I agree with Greg.Hahn...and be sure not to over modulate the STL TX
What STL?
Don't say Marti 8. another problem. Most people run the audio at 50 percent even with processing.
Composite toys sound better than stations without an stl because of the processing given the stl when it is sent. IMO
Station (No group Manager- Strident) was mad station A sounded better than station B...same ownership, same processing, the class A had an stl and the B didn't.
Don't say Marti 8. another problem. Most people run the audio at 50 percent even with processing.
Composite toys sound better than stations without an stl because of the processing given the stl when it is sent. IMO
Station (No group Manager- Strident) was mad station A sounded better than station B...same ownership, same processing, the class A had an stl and the B didn't.
I agree with the two major points already brought up:
1. Run the modulation on the STL system as low as you can and still make your noise performance. This will preclude getting distortion from running the modulator and detector out of their linear range.
2. Put a clipper between the output of the STL receiver and the exciter input. The detector really isn't that linear on transients - you'll get spikes and junk out of it which the clipper will get rid of.
Whether you run the studio clipper is a matter of opreference... try it both ways and do the one whioch sounds best. I like to turn the studio clipper off, but I've heard plants with it running which sounded great also.
Remember, the more signal you can get at the receiver, the better the noise performance will be and noise is usually the limiting factor on composite systems. Unless it's a short path, I'd use the biggest dish the tower will support and the biggeest feedline I could afford. You'll be on it a long time, maximize it at the outset.
1. Run the modulation on the STL system as low as you can and still make your noise performance. This will preclude getting distortion from running the modulator and detector out of their linear range.
2. Put a clipper between the output of the STL receiver and the exciter input. The detector really isn't that linear on transients - you'll get spikes and junk out of it which the clipper will get rid of.
Whether you run the studio clipper is a matter of opreference... try it both ways and do the one whioch sounds best. I like to turn the studio clipper off, but I've heard plants with it running which sounded great also.
Remember, the more signal you can get at the receiver, the better the noise performance will be and noise is usually the limiting factor on composite systems. Unless it's a short path, I'd use the biggest dish the tower will support and the biggeest feedline I could afford. You'll be on it a long time, maximize it at the outset.
O
OnTheAir123
Guest
I loved the days of composite STL's and having the processing at the studio with a Mod Sciences Clipper at the transmitter site. Adjustments were sooooo easy. Now it's Digital STL's with the processing at the transmitter and a pre-processor at the studios for STL protection.
Still sounds great but you have to drive if you decide that you want to make a tweak.
8)
Still sounds great but you have to drive if you decide that you want to make a tweak.
8)
littlejohn said:
We had the luxury of a two-mile hop with an EIGHT-FOOT dish and 7/8-inch feedline [from a more-distant prior path] looking up to a lower-gain receive dish at only 100-feet on the tower. HUGE RF receive level with minimal receiver-end gain nuked any chance for interference and noise, so we could allow about 5db of transmit headroom on our Mosley 606C. This significantly benefited the STL audio quality over other situations with marginal receive signal and a heavier TX modulation.
The composite proc we chose was from Jim Somich – the “BBE” [BaseBand Enhancer]. Somich was quick to correct one who called it a “composite clipper”! I was a bit puzzled as to its “how and what”; but it effectively eliminated transients, fattened the sound with an increase in density and a slight bass enhancement, and could be driven into nearly-constant activity with no 19kHz pilot ducking. It was well-designed, constructed, and a no-contest winner over the Mod Sciences unit. We purchased it in 1990, so I’m not sure about its availability today.
Processing was the garden-variety Compellor>Prisim>gutted Optimod-8100; but the well-optimized STL system and Somich box were primary factors in an awesome air sound!
The very best method is to take the transmitter to the receive site late one night, and trim the receiver to the transmitter. Usually, if you wrap a piece of wire around the load, it will pick up enough RF to run the receiver into saturation. Then you can touch up the LO for best distortion and noise performance. It >do< makie a difference.
If your composite STL is over 5 years old..
If your comp. STL is over 5 years old, it DOES need rebuilding-especially if its a PCL-606. The electroyltic caps in these units all dry up making the unit unstable and sonically poor. The decline happens over time, so it's not a night and day thing-until you get the units rebuilt.
Also, for some weird reason they make bipolar caps by putting two tantalums in parallel, polarity reversed...it never made sense to me, but if you replace them with REAL nonpolars, it makes a huge difference.
If your comp. STL is over 5 years old, it DOES need rebuilding-especially if its a PCL-606. The electroyltic caps in these units all dry up making the unit unstable and sonically poor. The decline happens over time, so it's not a night and day thing-until you get the units rebuilt.
Also, for some weird reason they make bipolar caps by putting two tantalums in parallel, polarity reversed...it never made sense to me, but if you replace them with REAL nonpolars, it makes a huge difference.
Well, out of Moseley's 606, 6010, 20, and 30; TFT 7200(I think), Moseley 505, Micro Controls composite, and MArti's.... the 606c was the best sounding one, with the Micro Controls box being a very close second. The difference being, with the occasional cap replacement, the 606 will soldier on forever. The Micro Controls boxes had problems with the chip caps in them changing value more or less randomly. But they sounded awesome when they worked.
We currently use Lynx T-1 speread spectrum radios and it's all a bitstream. They're more reliable than the telco T-1s or 1GHz STLs. There isn't any graceful fade to noise though... when they quit they're totally gone.
We currently use Lynx T-1 speread spectrum radios and it's all a bitstream. They're more reliable than the telco T-1s or 1GHz STLs. There isn't any graceful fade to noise though... when they quit they're totally gone.
OnTheAir123 said:
No you don't. Most processors today have remote control capability over a LAN, WAN and in some cases even the Internet.
We used to run our 8200 processor at the studio, but I swear there was a dramatic improvement in the overall sound after it was relocated to the xmtr site.
R
Actually, you can keep the processor at the studio with linear STL's. I have great success with running pre-emphasis AES across a number of Moseley Starlinks. On the other end we use Orban 2300 stand alone generators. We even use the 2300 generator with Omnia 6's at several locations. I have witnessed people driving starlinks at -20 dbfs. Might as well bring back the Marti 10's. With the processors there you can run at -3 dbfs with no problem at all.
OnTheAir123 said:
I notice several manufacturers still recommend put the proc ampat the plant, even when we're doing a bitstream which never returns to analog until it gets to the receiver detector. And I kinda wonder why.
Perhaps because it reduces the chances of noise? Here's our setup:
Studio:
program signal > Aphex Compellor > AD Converter > Harris Digital Microwave STL TX.
Transmitter site:
STL RX > Digital AES/EBU input into Orban 8200.
With this setup, the audio remains truly digital from the time it enters the STL TX, to the tine the Orban gets it.
R
Studio:
program signal > Aphex Compellor > AD Converter > Harris Digital Microwave STL TX.
Transmitter site:
STL RX > Digital AES/EBU input into Orban 8200.
With this setup, the audio remains truly digital from the time it enters the STL TX, to the tine the Orban gets it.
R
The whole idea of digital is there ainm't no noise. Presumeably, and up to the S/N floor. Try this:
Wheatstone G series>dbx agc>program profanity delay>intraplex T1 shelf>Lynx t-1radio stystem>omnia 6HDx>continental 802D.
Note, the only A/D is the mic digitizer.... CDs are loaded at digital. Whole thing runs AES3. It does get reclocked in the profanity delay, there's no help for that. Everyplace else it's running 44.1KHz, and sounds pretty good. Noise isn't a great problem. The HD is all Harris, but again it's bitstream throughout.
I've been unable to demonstrate a difference in jutter between master clocked and individually clocked chains, as long as they've got good equipment in them. The jitter difference in the last plant, all clocked with a Lucid master clock system, and this one was, there wasn't any.
State of the art seems to be, as long as you have that lovely clipper the 802D sports, you can do a real good job in total digital. Where we have Harris Digit exciters, we still feed composite... their clipper still picks.
Wheatstone G series>dbx agc>program profanity delay>intraplex T1 shelf>Lynx t-1radio stystem>omnia 6HDx>continental 802D.
Note, the only A/D is the mic digitizer.... CDs are loaded at digital. Whole thing runs AES3. It does get reclocked in the profanity delay, there's no help for that. Everyplace else it's running 44.1KHz, and sounds pretty good. Noise isn't a great problem. The HD is all Harris, but again it's bitstream throughout.
I've been unable to demonstrate a difference in jutter between master clocked and individually clocked chains, as long as they've got good equipment in them. The jitter difference in the last plant, all clocked with a Lucid master clock system, and this one was, there wasn't any.
State of the art seems to be, as long as you have that lovely clipper the 802D sports, you can do a real good job in total digital. Where we have Harris Digit exciters, we still feed composite... their clipper still picks.
- Status