Maker Pro
Maker Pro

Hey, Win! CCFL Inverter "self-learning" resonant frequency?

J

Jim Thompson

Win,

What do you know about CCFL inverters that claim "self-learning" of
the transformer resonant frequency?

Thanks!

...Jim Thompson
 
F

Fred Bartoli

Jim Thompson said:
Win,

What do you know about CCFL inverters that claim "self-learning" of
the transformer resonant frequency?

Yes, how do they have the capacity to induce such a frequency?
 
J

Joerg

Hello Jim,
Win,

What do you know about CCFL inverters that claim "self-learning" of
the transformer resonant frequency?

I am not Win and also not the expert on this but AFAIK the frequency is
swept during start-up and upon strike is being locked into current maximum.

However, I am only familiar with series resonant conversion and that's
how it is often done there, in cases where max power transfer is desired
instead of regulation. When regulating it is customary to slide up and
down the lower frequency slope (with SRCs).

If you are thinking about doing this in a chip, the method might be
patent protected for CCFL. Not sure but a quick Google found this one:
http://www.freshpatents.com/High-ef...t20050210ptan20050030776.php?type=description

Regards, Joerg
 
W

Winfield Hill

Jim Thompson wrote...
Win,

What do you know about CCFL inverters that claim
"self-learning" of the transformer resonant frequency?

Thanks!

That they've been listening to too much Rush Limbaugh?
 
W

Winfield Hill

Jim Thompson wrote...
Win,

What do you know about CCFL inverters that claim
"self-learning" of the transformer resonant frequency?

Thanks!

Getting involved in patent cases again? I see a patent
application "High-efficiency adaptive dc/ac converter,"
is that the one? "... the circuit is self-learning
and is adapted to determine the optimum operating
frequency for the circuit with a given load"
 
S

Spehro Pefhany

Jim Thompson wrote...

Getting involved in patent cases again? I see a patent
application "High-efficiency adaptive dc/ac converter,"
is that the one? "... the circuit is self-learning
and is adapted to determine the optimum operating
frequency for the circuit with a given load"

I'd dither, synchronously demodulate and integrate.


Best regards,
Spehro Pefhany
 
J

Jim Thompson

Jim Thompson wrote...

Getting involved in patent cases again? I see a patent
application "High-efficiency adaptive dc/ac converter,"
is that the one? "... the circuit is self-learning
and is adapted to determine the optimum operating
frequency for the circuit with a given load"

There are three patents, actually ;-)

Already scheduled to court.

Just poking around to see what's in the prior art.

...Jim Thompson
 
W

Winfield Hill

Jim Thompson wrote...
There are three patents, actually ;-)
Already scheduled to court.

I know about US6259615 and US6396722, what's the third?
Same guy, Yung-Lin Lin? All dating to Nov 9, 1999?
 
J

Joerg

Hello Jim,
Just poking around to see what's in the prior art.

Doesn't the common oscillator count? After all it 'self-learns' the
resonance of the LC circuit connected to it :-D

That would put prior art back into the days of Heinrich Rudolph Hertz
who passed away in 1894. As for switch mode controllers a frequency
adaptive scheme was easily done with chips such as Unitrode's UC3860.
Those are really old as well.

Regards, Joerg
 
J

Jim Thompson

Hello Jim,

Doesn't the common oscillator count? After all it 'self-learns' the
resonance of the LC circuit connected to it :-D

That would put prior art back into the days of Heinrich Rudolph Hertz
who passed away in 1894. As for switch mode controllers a frequency
adaptive scheme was easily done with chips such as Unitrode's UC3860.
Those are really old as well.

Regards, Joerg

Thanks, Joerg! That's the kind of lead I'm looking for... really old
prior art.

I'm seeing an awful lot of issued patents these days that would
indicate that the examiners aren't doing proper prior art searches...
issuing patents on stuff us old farts knew about when we were kids ;-)

...Jim Thompson
 
Jim said:
Win,

What do you know about CCFL inverters that claim "self-learning" of
the transformer resonant frequency?

Have a look at Baxandall's Class-D oscillator

Baxandall, P.J, Proc I.E.E 106, B, page 748 (1959)

The oscillator self-oscillates at the transformer resonant frequency.

Jim William's used it in his classic application notes on CCFL drivers

http://www.linear-tech.com/pdf/an55fa.pdf


http://www.linear-tech.com/pdf/an65f.pdf

though he calls it a Royer inverter, which is pretty strange for a
variety of reasons (Fred Bloggs begged to differ in our little spat on
the 29th July 2002).

Baxandall didn't develop the circuit explicitly to drive CCFLs - he
needed to generate high voltages for other purposes - but it was fairly
clearly intended to deal with the relatively high winding capacitances
of transformers with a high step-up ratio.

Pity that Jim has got me kill-filed - he'll never read this ...
 
J

Joel Kolstad

Jim Thompson said:
I'm seeing an awful lot of issued patents these days that would
indicate that the examiners aren't doing proper prior art searches...
issuing patents on stuff us old farts knew about when we were kids ;-)

As far as I can tell, while the patent office may well not be doing as much of
a prior search as they should, these days every Tom, Dick, and Harry who comes
up with a circuit that wasn't on one of their college homework assignments
figures it ought to be patentable and the business people running many
companies will do nothing but encourage them to do so -- prior art be damned.
 
W

Winfield Hill

[email protected] wrote...
Have a look at Baxandall's Class-D oscillator

The immediate issue isn't to find a workaround, it's to find
a way to invalidate a patent whose owners assert infringement
by products using their patented scheme. Sadly, despite even
good evidence of obviousness, it's very hard to get a jury to
overturn a patent that enjoys a good presumption of validity,
having passed careful review by patent examiners --- Who are
amateur jurors to overturn a patent professional? To succeed
one has to find something serous, like real evidence of fraud
(applicant intentionally didn't mention significant prior art),
or gross incompetence (the examiner failed to find or respect
significant prior art), that forces the jury to feel compelled
to over-rule the examiner.

In other words, one must find EXACT instances of prior art.

Either that, or the expert witness and defense lawyer (and his
entire team) need to have *extraordinary* skill with the jury,
so the jury feels the patent owner is a skank, and are looking
for good excuses to do the right thing as citizens, and overturn
his patent. That doesn't happen very often.
 
W

Walter Harley

Jim Thompson said:
[...]
I'm seeing an awful lot of issued patents these days that would
indicate that the examiners aren't doing proper prior art searches...
issuing patents on stuff us old farts knew about when we were kids ;-)

One of the reasons I'm so depressed about patents lately is that whenever I
go looking at patents to figure out whether I'm going to be sued for some
obvious idea I've used, I see all kinds of other obvious things being
patented. And I'm anything but a clever engineer, as Fred Bloggs has
correctly observed on occasion, so if something's obvious to me it's just
absurd that it's patented.

Here's an example, patent 6,372,976, filed in 1998 and granted in 2002. "A
pickup for an electric guitar includes a housing made to fill an existing
cavity in an electric guitar body originally used for a dual-coil humbucking
pickup. The pickup also has a single pickup coil mounted in the housing."

In other words: they've patented the idea of putting something small into an
enclosure, so that it fits a bigger existing hole.

Is it conceivable that solving this problem took an engineer more than 30 or
40 seconds of careful thought? Is it even conceivable that anyone could
come up with a more obvious solution to the problem?

I suppose now if I patent the idea of stuffing Kleenex into the end of
hand-me-down shoes to make them fit, I'll have to cite this patent as prior
art.
 
T

Tony Williams

Thanks, Joerg! That's the kind of lead I'm looking for... really
old prior art.

In that case perhaps look at the spark gap transmitters
developed by Marconi or Poulsen. AFAIR they used the negative
resistance of an arc in association with a self-tuned LC tank.
 
F

Fred Bartoli

Tony Williams said:
In that case perhaps look at the spark gap transmitters
developed by Marconi or Poulsen. AFAIR they used the negative
resistance of an arc in association with a self-tuned LC tank.

Plus a spark gap arc is a kind of CCFL, no?
 
T

Tony Williams

Plus a spark gap arc is a kind of CCFL, no?

Sort of. It's a discharge lamp, which could become an
actual arc if the current was allowed to rise too high.
 
Top