[TenTec] 9420 Power Supply - Update

[Dr. Gerald N. Johnson]

On Sat, 2008-11-01 at 22:44 -0500, Tom Macon wrote:
> Thank you to all those who have responded - I certainly appreciate it.  Here's some general comments in response to suggestions that I've received.
> 
> I normally run the Omni on a 961 supply that I picked up a couple years ago.   When I started having Herc problems, I tried putting the Omni back on the 9420 and found that it no longer worked, as I described in the original post.
> 
> Here's a little more detail on the voltage trip problem.  When the Herc is switched on, the Overload light comes on immediately and the relays for the 4 supplies do not stay closed.  The overload light and VCC fade away after 4-5 seconds as the caps discharge.  The over-voltage SCR on the Herc's distribution board remains tripped (since it's powered by the always-on supply) but since the Overload light is powered by the switched voltage, it does not stay on.
> 
> The voltage trip problem started with occasional trips while I was talking.  They became more and more frequent, and now it does it almost every time I switch on the Herc. I'm guessing one of the supplies is going unstable, but so far haven't been able to tell which one.  
> 
> With a digital scope, I have succeeded in capturing a momentary spike of about 20v on the switched voltage line that occurs at switch-on.  It seems to decay toward the normal supply voltage but, since the whole thing shuts down, it's hard to tell.  I wish I could scope the voltage of each p.s. while the Herc is running but I can't if it won't stay on.  I think testing the individual supplies is the next step - I just gotta figure out a suitable way to do it.
> 
> TU again for your help!
> 
>  - Tom, K9BTQ

Sounds to me like you have a melted pass transistor. Standard supply
failure mode and why there is a crowbar. Check the resistances from
collector to emitter. One will be very low.

Can't say it couldn't be a driver transistor or a control IC, but the
pass transistors are the most prone to failure because they run hottest.

73, Jerry, K0CQ