[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Super Staccato MkII VTTC Controller



Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>" <jimlux-at-earthlink-dot-net>

Have you considered using something like a Microchip PIC processor  instead
of the 556s etc.  I've found that once you get above 2 or 3 timers, going to
a little chip processor is worth it.  Fewer wires, easier to debug, etc.
And, you could potentially make it adjustable from a PC.  The chips run
about $5 (the EEPROM ones are bit more, say $10), Microchip sells a dev kit
with programmer for $100, although you can build a programmer for <$10-20,
and the development software is free.

I've found that the PIC is no more or less sensitive to ESD than the 556s.

I used to build all sorts of staged timer stuff with a combination of
relays, time delay relays, 555 circuits, counters and decoders, etc.  I
finally bit the bullet and started using PICs and I am glad I did (it's like
giving up discrete logic and using PALs, except for PALs the programmers are
more expensive!).

The only better solution is to use the small industrial PLC (programmable
logic (or ladder) controllers) from Allen Bradley and the like.  These are
really slick, are bullet proof, and fairly inexpensive ($100 ish), but
depending on the make, the development software will be expensive.  $100 may
seem like a lot, but by the time you buy all the IC's, drive transistors,
relays and power supplies, make a prototype board with wirewrap or point to
point or etching, put it in a package with connectors, etc., the $100 starts
to be a really good deal.

The only problem with the cheap PLC's is that for fast events (microseconds
and a few millisecond time scales) they don't work as well as the PIC
solution.  They are really designed for things that need to be switched in
10s of milliseconds.  The "big iron" PLCs (like the SLC 5 from AB) have
modules that can do fast timing, but you're looking at serious $$ then.

----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Saturday, January 20, 2001 9:24 AM
Subject: Super Staccato MkII VTTC Controller


> Original poster: "David Sharpe by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <sccr4us-at-erols-dot-com>
>
> Terry, John, David, Carl, Tesla List
>
> I'm working on a new VTTC electronic controller for medium to high power
> VTTC's.  It will incorporate a true "Super-Staccato" controller which will
> allow
> 1,2 or 4 consecutive pulses with adjustable rest periods. It will also
> incorporate
> forced filament heating time delay before HV application is allowed and
> real-time
> electronic over current protection of peak plate and grid current.  It is
> my hope
> and intent to develop a flexible "all in one" PCB solution that can be
> built and used for
> VTTC systems from 0.5-10kW or more.
>
> There are several issues with the simple circuit we posted in the past.
> Carl touched on it previously that the reset of the Schmidt trigger under
> certain
> situations may result in the cathode modulation pulse not being
synchronous
> with the
> AC utility wave.  After receiving a very powerful PSpice based electronics
> simulation
> package at work, I began testing of the earlier circuit and this new
> proposed solution.
> The bottom line is if the AC wave is below 1/3VCC and the reset of the
> Schmidt is
> low, output is off and everything is synchronized.  However, if the reset
> of the Schmidt
> goes high with the input trigger voltage below 1/3Vcc will result in a
> "runt" SCR trigger
> pulse which does not coincide with 0 crossing, but is delayed variably
> during the
> positive anode conduction wave of the VT.  One simple solution would be a
> hardwired switch on the PRF controller timer and fixed RC time constants
to
> force the
> reset timing to be completed while the VT is in the negative utility
> alternation under all
> conditions.  Another solution which is more complicated but appears to
> solve this
> situation under all conditions is:
> 1.  Pass the AC waveform through an improved Schmidt Trigger which is not
reset
>      controlled.
> 2.  This square wave but 50/60hz synchronized signal passes through a
> differentiator
> 3.  Another 555 timer is used as a 7.48ms SCR trigger monostable that is
in
> fact reset
>      controlled as the original circuit, the critical difference in this
> case is this monostable
>      is edge triggered.
> 4.  Output from circuit 2 passes into a edge triggered timing selected
> monostable which
>      delays PRF reset control for time to allow 1, 2, or 4 consecutive SCR
> output trigger
>      pulses.
> 5.  Output from 4 passes into another differentiator as 2 above.
> 6.  Output from 5 passes into edge  triggered PRF monostable control
timer.
>  Once
>       timing is started SCR trigger monostable (3) is reset and held reset
> until timing
>      is completed.  Once output goes low, reset on 3 is allowed to go high
> and monostable
>      3 will output high with next edge trigger.  IF an edge trigger is
> coincident with reset
>      there may be some "jitter" but the effective SCR triggering will
> remain synchronized
>      to within 0.5 ms of AC utility 0 crossing.
>
> Even though this circuit sounds complex, the actual "Super-Staccato"
> controls only requires
> 2 556 timers, 1 2N3904 output transistor, and 1 4N37 optocoupler to
isolate
> SCR drive
> from timers.  I intend to post timing diagrams, schematics, description of
> operation  within
> next week or two at Terry's (thank you!) graciously provided posting
ground.
>
> Since the electronics "infrastructure" has increased, consideration of
> monitoring plate, grid
> and filament conditions and operating external relays for HV control made
> sense.  These
> circuits are still very much in development but are for tube protection at
> this time.
>
> Basically operating a relay through a 741 comparator if peak plate or
> control grid currents
> exceed a set point value, and opening a relay if filament current loss is
> seen, and guaranteeing
> adequate filament heating time before HV is allowed to be turned on. It
may
> sound like
> extra "fluff" for a small coil, but paying hard earned cash to watch a
tube
> arc over and crash
> at $100-$1000 or more a pop versus $20-30 of electronics and relays seems
> like a good
> investment of time and materials to me.
>
> Comments and input are welcomed.
>
> Regards
> Dave Sharpe, TCBOR
> Chesterfield, VA. USA
>
>
>
>