Re: yoke resonant frequency

At 12:19 PM 9/3/97 PDT, you wrote:
>At 01:50 PM 9/3/97 -0500, you wrote:
>>> I've been lurking on this subject for a while and have an armchair
>>> engineering question. Has anyone hooked up a spectrum analyzer to see
>>> what the spectral content is of the vector signals? With all this talk
>>> about hooking up audio amps makes me wonder if this will filter off some
>>> of the high end of the signal.
>>
>>Of course it should be measured at the output of the driver because the
>>yoke itself will cut out some high frequency stuff. Which brings me to
>>another question:
>> Aren't deflection coils current controled? i.e. isn't the deflection
>>proportional to the current and NOT the voltage across the coils? I know
>>this is a really basic question, but I've never dug into the schematics
>>really good and they look like an audio amp at first glance. BTW, I've
>>often wondered if speakers should be current driven for similar reasons.
>
>They are current controlled, if they were simply resistive it wouldn't make
>any difference whether you looked for current or voltage. They're inductive
>and the amount of deflection depends on the amount of current flow, not the
>voltage across the yoke.
>

I agree, the signal coming out of the deflection transistor is in the
current. When I first looked at the schematics for the deflection board
when I was fixing my Tempest, the following tidbit leaped out at me. From
the differential stage to the deflection output, the schematic looks *very
much* like an Operational Transconductance Amplifier that I designed in
grad school. The big difference between my project and this one is that
mine was a VLSI mask and this one is discrete components. So I drew a big
box around it the OTA part and sure enough, there is a feedback resistor from
the deflection transistors back to the differential gain stage!

So from the big picture, each axis can be viewed as an op-amp configured as an
inverter driving the coils. (Note: This is from year old memory, it may be
non-inverting. I'm going to have to dig out the schematics back out.)
Of course, there is also the ancillary spot killer stuff, but that is an
add on. This design should be somewhat linear, depending on the overall
OTA gain.

(I really wish I had the lab resources and time to dig into this project. I
really like this analog stuff. :-) )

>From the armchair,
Chris Moore
cmoore@heartlab.com

--
Christopher V. Moore
Heartlab, Inc. - 101 Airport Rd - Westerly, RI 02891
Phone: (401) 596-0592 - Fax: (401) 596-8562 - Email: cmoore@heartlab.com