[BC] A fun night..

[Alan Alsobrook]

I know these things happen more often than I stop to write about them, 
but this semi quick story has some technical pointers in it so I'll drop 
it in here.

I guess things really started in the morning when I got a call from one 
of my stations telling me that the right channel sounded really bad, 
"like a broken speaker". Upon hearing my groggy voice that had just 
gotten into bed from another all nighter reprogramming an automation 
system, he asked if he should run to the transmitter and switch to Mono 
mode. Knowing that the Mono mode on the Orban 8100 goes to the left 
channel I said "that should work". And it did, were were in mono but 
sounded just fine. (yes the GM knows where the transmitter is and even 
has keys to it!)

The evening started off well enough, Got a phone call from "some guy" in 
Arizona wanting me to do a few things on the Radio Guide AM RF class 
coming up in Austin. Hung up with him and the phone rang again. Tis time 
it was a frantic call from the station that I had reprogrammed the 
automation for, stating that the Phantom was going NUTZ! I knew 
immediately what was wrong, I hadn't told the day staff to regenerate 
the log from 2100-2359 to finalize the changes I made. OOPS!! so I 
dashed out the door and and arrived a few moments later, and quickly 
regenerated the log and was able to get that under control. The really 
bad part is that during that time we are originating a new nationally 
syndicated show, not something we normally do a small station.

With that all running smoothly It seemed it was time to go start poking 
into the audio problem with the other station. This station uses a dual 
hop STL, the first being on ma bell copper for about 1500' and then 
going to a Mosley DSP6000 Digital encoder set up for 3 channels L,R, and 
Aux 1 (used for remote control).

I started at the transmitter just to make sure it wasn't there I 
switched back into stereo mode and sure enough the right channel sounded 
terrible. A quick swap of the audio cables on the rear indicated that 
the problem stayed with the right channel of the decoder.

I figured it was time to head to the Bell tower (the center location) 
and see what I could find there. With the station back in mono I headed 
down town. After arriving I swapped the audio channels on the encoder 
and the stations audio remained clean. So much for my theory of a 
problem with Ma Bells wiring, this problem was all mine and somewhere 
between the input jack on the encoder and the output jack on the decoder.

At each location I had rebooted the DSP's (several times actually) but I 
had no way of knowing if rebooting the encoder had made any difference, 
since I was only able to listen to the left channel in mono. So back to 
the transmitter. Well what would you know, I got there and found the 
Right audio was still trashed. Now it was think time,0100, do you go for 
it or not? Ok, I'll go for it. I snatched the Decoder from the rack and 
headed back to the bell tower to get the encoder, to bring them both to 
the shop. I discounted the RF portion of the link figuring that even 
though the bands were really bad when the problem started, I don't think 
that their is any way to have an RF problem that only affects the data 
for one channel.

Now I have a decoder and the encoder on the bench, hook up my test leads 
connect the two together and plug them in.. POOF!!! smoke cloud from the 
encoder, not exactly what I was expecting. At least that's a place to 
start. Got inside and found a shorted bridge rectifier for the +-15V. I 
replaced that, and the unit came up. Shew.. Now to see about the audio, 
  check the left channel 0db in 0 db out, and 0.5% THD, that looks good. 
now for the right, 0db in 0db out, but with 38% THD. I think we have a 
problem.

Since the encoder was already open it was checked first, this proved to 
save some time since it was easy to see a very distorted sine wave on 
the O-scope at the R Channel analog test point. looking at the schematic 
  I see 3 NE5532's in the circuit, all on sockets too! Time for a quick 
shotgun blast! 3 new IC's and nothing changes. Hummm, I guess I'm 
actually going to have to figure out what's wrong with this thing.

Some more scope probing shows that one of the 5532's is not getting its 
+15 supply, which comes through a 10 ohm 1/4 watt resistor. Now look at 
that the resistor is burnt up! I don't know how I missed that on the 
visual but sure enough it was black. Now that I was concentrating on the 
area I could see that the .1uF decoupling cap for the IC was also black 
on one end. An Ohm meter check confirmed it was shorted, and most likely 
the culprit, behind all the damage. With both the resistor and cap 
replaced, a recheck showed all good!

Ran around reinstalling everything starting at the transmitter (since 
it's close to the shop), ending up at the bell tower, once all was 
connected and turned on, I hear the digital system sync up and lock in 
at 0559..

Isn't Radio Engineering fun!!!

-- 
Alan Alsobrook CSRE AMD CBNT
St. Augustine Fl. 32086 904-829-8885
aalso at Bellsouth.net