I think I found the problem...

A case of PEBKAC (Problem Exists Between Keyboard and Chair)? Tell us about your war stories!
User avatar
Deep Thought
Posts: 3211
Joined: Thu Mar 20, 2008 9:23 am
Location: La Grange, IL
Contact:

I think I found the problem...

Post by Deep Thought » Wed Sep 03, 2008 5:53 pm

Symptom: AM 10 KW DA, 2 towers, running at 6 KW with 2.5 KW reflected power

Victim: 0.004 uF G4B, rated at 20 Amps, part of input leg of ATU tee network. You'll also notice three blackened turns on 20 amp coil which is in series with the capacitor.

Engineer on site swears he checked the ATU input impedances prior to returning the station to full power.

Pathology:
0821081023-00.jpg
Cause: 2.5 -j15 ohm input impedance at 5 KW power for this tower. That's 44.7 amps for those without a calculator handy.

Solution: $700 part rushed from Kintronic Labs, next day air. In the meantime, 1/4 power with cobbled up temporary replacement. 5 minutes with impedance bridge to set the network back to 50 j0.

Moral of story: Don't skip verification of proper setup before lighting the big flame.
Mark Mueller • Mueller Broadcast Design • La Grange, IL • http://www.muellerbroadcastdesign.com

User avatar
BroadcastDoc
Site Admin
Posts: 2729
Joined: Tue Nov 13, 2007 3:34 pm
Location: Milwaukee, WI
Contact:

Re: I think I found the problem...

Post by BroadcastDoc » Wed Sep 03, 2008 6:46 pm

To quote the Mythbutsters:

Well *that's* your problem right there... :)

Can't say that I've experienced that failure yet, though I've heard stories!
Christopher "Doc" Tarr CSRE, DRB, AMD, CBNE
Help support the Virtual Engineer, use our 1&1 Affiliate link if you need good, cheap hosting.
Virtual Engineer. The Broadcast Engineering discussion forum

User avatar
RGORJANCE
Posts: 1387
Joined: Mon Dec 31, 2007 7:48 am
Location: RACINE, WI

Re: I think I found the problem...

Post by RGORJANCE » Thu Sep 04, 2008 7:55 am

I have seen this many times, however I am impressed by the cleaner gray color on the floor, and the white residue on the panel.

Whatever happened to the black sticky stuff that they used to encapsulate???

Also saw in lots of locations, all kinds of shrapnel all over the floor after a big arc.

Just had a very interesting one here in Racine. We lost metering reliability on the remote base current. Went to the doghouse and observed a huge black smear on the toroid, and a metallic spray pattern on the insulator supporting the J-plug. Closer inspection revealed the epoxy .01 8kv cap split all the way around and three additional hairline cracks about half way around. The bigger crack was a good 1/8 inch wide. The xmttr did not show any indication of VSWR and the station was still delivering power. Never saw that kind of capacitor damage without serious xmttr reaction.

Replaced the cap and didn't notice any parameter change.

Bob

User avatar
Deep Thought
Posts: 3211
Joined: Thu Mar 20, 2008 9:23 am
Location: La Grange, IL
Contact:

Re: I think I found the problem...

Post by Deep Thought » Thu Sep 04, 2008 9:06 am

0.01 uF at a relatively high frequency (say, 1460) is only about 11 ohms reactance. If that's on the output leg it's probably a small percentage of the total for the leg. A short would not change things much.

This particular cap is from the mid 80's and is (was) full of wax instead of tar. It just overheated, which melted the wax, which squirted out of the seam between the ceramic and a metal end cap, until the metal inside finally touched.
0821081024-01.jpg
You can see from this (blurry) photo where it escaped. The final gasp had some fireworks but not enough to rupture the ceramic, as evident by the blackness near the end bell.
Mark Mueller • Mueller Broadcast Design • La Grange, IL • http://www.muellerbroadcastdesign.com

w4cl
Posts: 479
Joined: Wed Apr 09, 2008 3:04 pm
Location: 1 3/4 meters

Re: I think I found the problem...

Post by w4cl » Thu Sep 04, 2008 6:59 pm

I have been out of AM radio too long. I forgot they EVEN made caps like that. [shakes head in disbelief as the alzhiemers sets in]

User avatar
K9EZ
Site Admin
Posts: 2986
Joined: Wed Nov 14, 2007 3:16 pm
Location: Raleigh, NC
Contact:

Re: I think I found the problem...

Post by K9EZ » Fri Sep 05, 2008 9:10 am

I am going to have to pull the pics of when WISN blew out a cap in the tuning network on its Nautel Amphet 50.
Kent Winrich
Owner, Consultant, Floor Sweeper and Official Thread Hijacker
Broadcast and Technology Blog http://sceotech.com/blog/
Raleigh, NC

boiseengineer(old)

Re: I think I found the problem...

Post by boiseengineer(old) » Tue Nov 25, 2008 10:08 pm

Have to share this pic from a friend.
A little under rated for 50 KW.
DAY Match.jpg
Last edited by boiseengineer(old) on Tue Apr 07, 2009 8:51 pm, edited 1 time in total.

User avatar
Deep Thought
Posts: 3211
Joined: Thu Mar 20, 2008 9:23 am
Location: La Grange, IL
Contact:

Re: I think I found the problem...

Post by Deep Thought » Tue Nov 25, 2008 11:25 pm

Nice fuse.

Is that in a diplexing trap?
Mark Mueller • Mueller Broadcast Design • La Grange, IL • http://www.muellerbroadcastdesign.com

w4cl
Posts: 479
Joined: Wed Apr 09, 2008 3:04 pm
Location: 1 3/4 meters

Re: I think I found the problem...

Post by w4cl » Wed Nov 26, 2008 6:34 am

That isn't an old Lawrence Behr is it? :D

boiseengineer(old)

Re: I think I found the problem...

Post by boiseengineer(old) » Wed Nov 26, 2008 7:14 pm

Home brew five tower, two station array, and that was the DAY Common Point Match.

VXO
Posts: 45
Joined: Thu Nov 15, 2007 2:18 pm
Location: Miami, FL
Contact:

Re: I think I found the problem...

Post by VXO » Tue Feb 03, 2009 10:06 pm

This somewhat reminds me of the large electrolytic capacitor I once witnessed fail with a distinct "spewing" sound. I wish I had a recording of that, as it is the most hilarious sound I have ever heard.

Spttthhhhthewthewthewthhttthhththththttthhhhpoop!
A K T I V - S C H A U M

K4WRF
Posts: 90
Joined: Sun Jun 08, 2008 8:32 pm

Re: I think I found the problem...

Post by K4WRF » Tue Apr 07, 2009 7:26 pm

I honestly believe I prefer these types of gross identifiable failures as opposed to the one out of 22 that randomly change value by 20% on warmer days with no visible or consistent pattern. Makes one grow old, mean and cynical in very short time period.

w/

User avatar
RGORJANCE
Posts: 1387
Joined: Mon Dec 31, 2007 7:48 am
Location: RACINE, WI

Re: I think I found the problem...

Post by RGORJANCE » Tue May 05, 2009 10:49 am

Gross failures rarely require more than a look at the destroyed part. It's a real challenge to have a trouble dumped on you that has over 80 man hours of effort with no fix. Happened to me twice working on crypto equipment in the 60's over in Germany.

It is a great feeling when you pull the rabbit and the whole rest of the bunny family out of the hat after only a few hours of troubleshooting!

HOWEVER, there is a huge downside to getting the tough ones fixed. The bosses think you're brilliant and you instantly inherit everything else that nobody could fix. Bummers! I'll take the easy ones anytime, except those that throw melted tar over everything nearby.

Fossil

User avatar
Deep Thought
Posts: 3211
Joined: Thu Mar 20, 2008 9:23 am
Location: La Grange, IL
Contact:

Re: I think I found the problem...

Post by Deep Thought » Tue Sep 01, 2009 10:20 pm

Nope. 100 amps will not go lightly through a 20 amp coil...
KKAD Tower 3 power before.jpg
IMG_5425.JPG
...the roller and roller bar was completely destroyed.
Mark Mueller • Mueller Broadcast Design • La Grange, IL • http://www.muellerbroadcastdesign.com

boiseengineer(old)

Re: I think I found the problem...

Post by boiseengineer(old) » Thu Apr 08, 2010 10:02 pm

With the freaky slow moving thunderstorms in the East the last few day I bet there'll be a lot of the aforementioned failures. How's everyone holding out?

Stay safe!

Post Reply