RE: LV2000 demands

At 03:16 PM 11/13/98 -0500, Jeff Anderson wrote:
>
>
>That is an excellent improvement, unfortunately the LV2000 does not take
>the transistors out of the loop, that IS the main weakness. The only
>reason the parts on the board burn up is because of failures of the
>transistors, so in actuality all the LV2000 does is provide a means of
>shutdown when the transistor fails to protect itself. It does not solve
>the weak link, it just covers for it..

Here you are talking about the main amplifier stage, which is built up from
discrete transistors in a differential pair configuration along with a
current source.
It is true that this part of the WG deflection design is not infallible. It
does however work if the power source is kept in check. The "overvoltage"
input protection, of the later model WG deflection PCBs, helps with this
problem. It, like the LV2000, covers for the "weak link" as you put it.
If the original low power section gets fried, there is no protection from
over current and the amplifiers could get damaged. If, however, you have
both this input protection and the power protection of the LV2000, there is
little chance of frying the transistors in the amplifier stage. The LV2000
provides over current protection by the vary nature of the regulators. Thus
a working WG deflection PCB will continue to function with an LV2000 installed.
i.e., the LV2000 "just covers for it" very well.

>Even if a transistor does go on a non-LV2k monitor you have less than
>$2.00 worth of parts to replace IF it takes out the parts instead of the
>fuse... and the regularity of such an event is very uncommon after the
>monitor has been rebuilt and properly adjusted.

The facts are that this WILL happen at some point in the WG deflection PCB.
It's a matter of how much run time your monitor gets.

>If there was a cost effective solution to take the trannys out of the loop
>you still have the deflection transistors to deal with. The only solution
>is to design an entirely new deflection board or monitor or double-up the
>transistors with a means of indicating when one of them blew.. And the
>costs of doing either still does not offset the money or the time involved
>in nursing a straight WG.. Which is simple and cheap.
>

Designing a new deflection board is something that I've been thinking about
for some time. I have two basic designs that could work. One is basically
replacing the amplifier stage with more robust protection circuits, and
high speed opamps to drive output power transistors. This would replace
your deflection PCB, but not the power transistors. This one I know will
work. Another design uses hi power opamps from Apex (Clay did the research
here) to remove not only the deflection PCB but also the output power
transistors all together. The problem with these two solutions is that they
would be very expensive.
It is very possible to produce a new deflection PCB, however, I might agree
with you on the expense vs nursing point. I still need to tweak a few
things on my design, and figure out if it would be cost effective. I don't
believe it will be, but we'll see.
Producing the LV2000 was a much easier decision. Since it does not replace
the deflection PCB, you don't lose the "originality" of the game. Merely
what you loose is having to replace the low power section every time you
get an over current situation there. Plus the LV2000 is relatively
inexpensive, since when the original low power section fries, you not only
have to replace those parts, but also the parts that caused the problem.
For me, and for many others, it's not fun anymore to repair that low power
section when you've done it many times over. Just slap in the LV2000 and be
done with it.
It's all a matter of your perspective and an individual decision. If you
like to repair the low power section, and replace a few transistors in the
amplifier ever six months to a year, that's cool. If not, then you have the
option of building your own LV2000 (schematics, bom, instructions, etc are
all on-line), or you can buy a pre-fabbed LV2000 from me, in kit form or
assembled.

-Anders.
Received on Fri Nov 13 15:35:15 1998