TenTec
[Top] [All Lists]

Re: [TenTec] Troubleshooting Power amps in general

To: <geraldj@weather.net>, "'Discussion of Ten-Tec Equipment'" <tentec@contesting.com>
Subject: Re: [TenTec] Troubleshooting Power amps in general
From: "Art Trampler" <atrampler@att.net>
Reply-to: atrampler@att.net, Discussion of Ten-Tec Equipment <tentec@contesting.com>
Date: Mon, 22 Mar 2010 19:56:13 -0500
List-post: <tentec@contesting.com">mailto:tentec@contesting.com>
Jerry,

Thanks for the overview on how you would approach this.  

I knew I could count on this group for sound direction.  I won't work with
the HV exposed; not my skill level.  I will discharge, and discharge again.

Art

-----Original Message-----
From: tentec-bounces@contesting.com [mailto:tentec-bounces@contesting.com]
On Behalf Of Dr. Gerald N. Johnson
Sent: Monday, March 22, 2010 6:57 PM
To: tentec@contesting.com
Subject: Re: [TenTec] Troubleshooting Power amps in general

There are several possible failures that will blow the fuses. Blowing 
more fuses won't fix the failure and may damage more parts.

The failure could easily be a shorted rectifier and filter capacitor. 
One caused the other, but its never clear which went first. If a diode 
shorts from a transient voltage, that puts AC on the electrolytic filter 
capacitor which can destroy the capacitor. If the filter capacitor 
failed shorted, that excess current can melt the rectifier destroying 
the junction making the rectifier into a short.

Loss of protective bias can let the PA tubes draw excess current. If 
that only blows the fuses you are in luck, but before you apply plate 
voltage you need to know that didn't damage the tubes, like burning 
loose a grid wire to flop over inside the place to make a short. Or a 
filament strand that can break and short the grid, when then makes the 
tube draw excess current and damages cascade. Its possible for a tube to 
fail without outside influence other than a poor weld at manufacturing 
time, or too much shock somewhere in shipping or operation.

Or like a certain series of Ameritron amps, the PA could go into a LF 
parasitic oscillation somewhere below 100 kHz, that wasn't bypassed 
adequately and got into the plate transformer high voltage winding and 
broke down the insulation because of a matching resonance in that winding.

We would wish that the fuse would blow before other parts are damaged. 
Unfortunately that's inconvenient because there are legitimate turn on 
surge currents that the fuse has to carry. Then we wish for the fuse to 
blow quickly, but the slow blow fuse that allows charging the filter 
capacitors can't be a fast acting sensitive fuse. Investigating the 
choices of fuses for a 1 KW AC solid servo amp about 1965, I came to the 
conclusion that even the fastest available solid state rated fuses (very 
expensive because the fuse wire is solid silver) were still only 
reliable indicators of a failed semiconductor device. That's because the 
fuse took longer to blow than the power transistor.

To trouble shoot, I would remove the tubes first. Check them for shorts 
between filament and grid and plate. Its least likely there will be a 
filament to plate short, but it not impossible even without showing a 
connect to the grid also.

Then a classic troubleshooting technique is to apply power through a 
current limiting resistor. For a power amp, a 1 KW incandescent lamp 
would be suitable. It limits the current and glows brightly until the 
short is found and cured. But that exposes you the high voltage which 
can do more than hurt you. It can do permanent damage including kill 
you. Its a bit safer to apply 5 of 10% of rated line voltage. The only 
circuits that won't work at low voltage are voltage regulators and the 
tube filaments. All the other circuits will rectify and make DC voltages 
but at about the same 5 or 10% of normal operating voltage. When all the 
bias and supply circuits show the proper fraction, you can increase the 
voltage with a variac but then the lamp is a very good protector (AND 
VIVID indicator). Shorted filter capacitors and rectifiers and maybe 
transformers will show their failure at the low applied voltage, though 
there is some chance they won't break down until nearly all the normal 
voltage is applied. That makes the current limiting lamp vital.

Electrolytic capacitors don't store charge long, but you need to be sure 
they are discharged before you get body parts across them by shorting 
them with a sturdy and well insulated screwdriver or purpose made ground 
stick. If there is charge you may well make divots in the screwdriver 
and whatever components you shorted. After the flash, shove the 
screwdriver back in contact again to be SURE the discharge is complete.

These are the techniques I'd use to trouble shoot, I keep a bench supply 
with fixed 6.3 and variable 6.3 volt outputs that are handy for such 
tests. The lamp I don't use as much as I should.

73, Jerry, K0CQ

On 3/22/2010 3:50 PM, Art Trampler wrote:
> The responses have confirmed what I was already thinking/my gut was
telling me.
>
> I guess I was hoping some fount of wisdom would show me why replacing the
fuses at this point made some sense!
>
> Stuart--thank you for the thought on making a diagram of anything I
disconnect to isolate a component.  So maybe this weekend I'll set up a
suitable workspace and begin the process...safely.  I don't know how long
those capacitors could hold a charge, but I won't be taking any chances with
them.
>
> Doug...sorry about your experience.  Ouch doesn't begin to describe it.
>
> I just hate it that I'd left the shack at that point, though in all
probability it would not have told me much to have been there.
>
> 73,
> Art
>
>
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
No virus found in this incoming message.
Checked by AVG - www.avg.com 
Version: 9.0.791 / Virus Database: 271.1.1/2763 - Release Date: 03/22/10
14:44:00

_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec

<Prev in Thread] Current Thread [Next in Thread>