Heathkit H89 Video Question

From: Tony Duell <ard_at_p850ug1.demon.co.uk>
Date: Sun Aug 31 12:01:25 2003

> OK, I'm working in unfamiliar territory... and since there seem to be a lot
> of video experts of late, this seems like the time and place to ask...
>
> I'm working on an H-89 that seems to have a problem in the horizontal drive
> on the video board. The unit powers up from cold to a good display,

OK...

> properly sized, and stable. After the unit operates for 10-15 minutes, the
> video breaks down, first by jittering slightly for 30-60 seconds, then it
> goes to a single vertical line on screen that fades to black. At this time,
> the filament on the CRT also stops glowing. If I leave the unit off for
> 20-30 minutes, it comes back with good video for a bit, then repeats the
> above. If I turn the unit back on right away without a cool down, there's
> no display/no filament glow.
>
> After tracing the (good) horizontal output signal from the terminal logic
> board into various part of the video board, and checking voltages where I
> felt I could reach safely in a hot set, I discovered that if I shot
> horizontal output transistor Q217 with cold spray I'd get video back right
> away.

OK, it sounds like the HOT is getting too hot :-). This could be due to
one of 3 things :

1) The HOT (Q217) is faulty.

2) The load is too heavy. This could be the flyback (but I doubt it -- a
shorted turn in the flyback is going to have more serious effects that
this), it could be something powered from the flyback (again unlikely)

3) The drive waveform is wrong. I don't have the H89 schematics to hand,
but there's often a low-value electrolytic capacitor (a few uF) in the
base circuit, particularly if it's transformer-driven. If this
deteriorates, the drive waveform is incorrect, the HOT turns off too
slowly and disipates more power than it should.

> 1) Although the flyback _appears_ OK, could the leak have caused some unseen
> damage that is now causing excessive current through Q217 and shutting it

I think this is unlikely.

> down? My limited understanding of flyback construction is that they are
> sealed units. It takes several minutes (10-15) before the failure is
> observed. Could a flyback failure allow operation for this long? How do I
> test this thing?

The only real test is a ringing test (you resonate one of the windings
with a capacitor, apply a voltage pulse and observe the decaying
oscillations on a 'scope). Problem is, this doesn't work too well on
intermittant faults!

>
> 2) If I proceed under the assumption that the flyback is OK and Q217 is
> not... Q217 is a BU500: 1500V, 6A, 75W, SI-NPN. I normally buy from Mouser,
> DigiKey, and Jameco, but none of them seems to have this part, and I wasn't
> able to turn up anything useful Googling. A possible dealer or two in the
> UK, but for a $2-3 part I was hoping for a more-readily-available source or

Can't help you, I'm in the UK, and it's not hard to get the BU500 series
hre.

> an equivalent. The closest I can find in the family is BU508A, which is
> 1500V, 7-8A, 120-125W (depending on whose specs you believe). Is that OK or
> too far out of spec? Does anyone know of US source for the real McCoy?

Specs are not normally of much use in choosing a HOT, because it works
under such strange conditions. But the BU508 is also a HOT so it might
well work.

But check the base network first.

-tony
Received on Sun Aug 31 2003 - 12:01:25 BST