Tempest/LV2000 Problem/Question

Hello, this is my first time posting to this list, but I've been watching
since October... Jeff Hendrix suggested I raise my question with the
vectorlist:

(This is a very long message, as it's a copy of my posts to RGVAC, but I
wanted to make this as complete as I could)

------------------------------
(Here's the original message I sent him)

I've recently installed your LV2000 into my Tempest monitor, along w/ a
cap-kit. However, I'm having some problems, and am hoping you might have
some ideas.
 
When I do your test procedure, w/ nothing connected but the deflector board,
both diodes light up fine. Plus, I've adjusted the pots to +/- 26 volts per
the instructions.
 
However, when I connect up everything else, and start it up, the following
happens.
 
1) Both diodes come on, and so does the spot killer.
2) Spot killer turns off after 2 seconds.
3) The right diode (+26V?) turns off after 5 seconds, and the spot killer
remains off.
4) A small, but very erratic picture does show up in the lower-left hand
corner (monitor layed horizontal on my table). It's about 3" wide x 2"
high. I can see red lines, blue lines, and yellow lines. (Oh, and the game
does play blind)
5) Turn off the machine, the left diode turns off, the right turns on, and
so does the spot killer. (The two fade out slowly)
 
The only up-side to all this is I didn't have *any* picture before I started
this, as the picture had began to fade out in quadrant's, then completely
gone.
 
------------------------------
(Here is his reply)

It sounds like the LV2000 is working the way it was designed to. Instead of
components burning up, it just shuts itself down.

Since you're seeing a picture, you know the HV unit is fine. If you have an
oscilloscope, you can use it to check the output of your game board to
isolate the problem to either the vector generator on the game board or the
deflection board.

------------------------------

Unfortunately, I don't have an oscilloscope, only a DMM. I did post this
question to the RGVAC newsgroup and got a bit of responses, of which is the
following, that I also posted to RGVAC:

First, the biggest questions were about the deflection transistors. All 6
were replaced during the Zanen cap-kit, along w/ replacing transistor
sockets and insulators, per suggestion from Rob Walters at ArcadeObsession:
    http://www.arcadeobsession.com/capkit6100.html
The new TO-3 power transistor sockets are definitely nicer than the old
sockets. (I got mine from Mouser)
 
Also, I replaced the socket and insulator (not the transistor) on the
transistor located on the HV cage. Plus, the plastic insulator circle
around the small transistor by the focus knob was falling off, so I removed,
cleaned, and reapplied it w/ silicon. I also checked for no continuity on
both of these. I wanted to mention this in case it was of any importance.

I did a continuity check on each of the 6 transistors, and none are shorting
to ground, either from the cap itself, or from the screws bolted in. I did
learn though, you have to do this check w/ each pair's respective connector
removed, otherwise the DMM resistance values jump around (like from 2k to
2M).
 
Ok, some questions: What's the proper way to make sure a deflector is good,
and not blown? And can I hook up a single pair of deflectors, and turn the
machine on, w/out consequences? What I'm wanting to do is hook up different
components one at a time, and see when the LV2000 light drops out. **But is
there a risk to this?** If so, please tell me.
 
When I first saw this problem I thought deflection problems as well, except
for one thing: The image wasn't centered, it was placed in the lower left
corner (monitor layed horizontal on my table). Now that seems to me to be
deflecting fully to 2 of 4 edges. I got to wondering if the problem isn't
so much the deflection circuitry, as the circuitry that amplifies
(stretches, multiplies, etc) the image to full screen. Do the deflection
circuits handle this as well? Or is done elsewhere?
 
But I still have the problem w/ the right LED going out after 5 seconds. It
seems something begins to mess w/ the +26V to the point the LV2000 shuts
down, and since it happens only when the HV, deflectors, and Neckboard are
plugged in, I'm assuming one of those 3 is causing it. Can anyone tell me
how to rule out any of these options?
 
Now, does anyone know if the LV2000 problem could be causing the small image
problem? Or, am I dealing w/ 2 separate issues here? (As in, the LV2k
problem is causing the erratic image, but not size problem)
 
------------------------------

I've been doing research and found this site useful:
   http://www.bithose.com/serfaq/REPAIR/F_VAtari7.html
And, in particular:

10.2.7) Display bad only in lower right quadrant
Distorted image which exists ONLY in the lower right quadrant of the
display.
I have seen this several times and the problem has always been a bad ZD101.
Much of the time this causes R101/Q101 to fry and spot killer comes on.

Now if this is described as the monitor in the cab, then this matches
exactly what I'm seeing. Except the ZD101 was removed when the LV2000 was
added. So what would I check now?

If nothing else, this leads me to believe that if I can fix why the LV2000
shuts its +26V off, it'll fix the small erratic image problem as well. Or,
whatever would cause the ZD101 to blow, is probably why the LV2000
shutsdown?

Also, from the same website:
10.2.5) R101 burns on power up
...(snip)...

This caught my attention. My R101 was still working (before adding the
LV2000, of course), but it apparently got very hot, as it burnt the PCB
underneath it. Again, noting this in case it's a clue. I only read this
this morning so I haven't had a chance to look at R903 yet, which is the
suggested culprit.

------------------------------
I apologize for this very large message, but I'm hoping someone here can
help me out. The only two clues I've found so far is a bad ZD101, which is
now gone. And a possibly bad HV, which Jeff believes to be good.
Any comments/suggestions?

Thanks in advance,
--Eric

---------------------------------------------------------------------------
** To UNSUBSCRIBE from vectorlist, send a message with "UNSUBSCRIBE" in the
** message body to vectorlist-request@synthcom.com. Please direct other
** questions, comments, or problems to neil@synthcom.com.
Received on Thu May 17 14:56:25 2001