Basic Four quadrant bidirectional switch patented

Can you find an example of this being disclosed prior to the patent date?

That is not an issued patent, it's a patent application. Under the new rules adopted several years ago, applications are published with some delay after filing. I did a very quick search at the USPTO using the first inventor's name (Fornage) and this does not seem to have issued, at least to date. The same inventor has a decent number of recently-issued patents in the energy field.

That's not an issued patent, it's just a published application. Applications are now published with some delay (I don't remember how much) after filing. A quick search of the USPTO using the first inventor's name shows this is not (yet) an issued patent. The same inventor has a number of issued patents in the field of power electronics.

It could be a troll, or just a dummy.

That patent issue that I was fighting (as expert witness) last year, a troll calling a few vague sketches with no meat under them a patent, was just thrown out by the judge... who has now asked us to submit our costs :-) ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Doesn't prior art get that tossed summarily ?

How do you figure? Looks like they're using some kind of dual gate depletio n mode like structure in addition to what you consider an obvious analog sw itch configuration of FETs. Just saying something is insane doesn't cut it, show where they're plagiarizing prior art in any form whatsoever.

Is it form or function that cna be patented ? The thing doesn't look like i t does anything undone before. If it outperforms the preceding devices, doe s that make it patentable ? This thing looks like it does about what an inc andescent light dimmer does. If the new process can make that faster so it can do more, but the basic parameters are damnear what was already had, are they patenting something other than an epitaxial diffused or whatever bipo lar ?

Maybe you folks consider thisa a stupid question, if so, sorry. But it seem s to me this has been done. If a device can do it better, but it still does the same thing, what gives anyone the right to patent it moreso than say, a better OPAMP design ?

Can someone explain to me just what this thing is supposed to do unlike som e things that are already out there ? There are countless typres sof thrysi stors out there, and they become the subject of discussion once in a while because it is not easy to be up on all the different types. Now they did no t all come out at the same time. Did anyone patent the SCR ? The GCS ? The triac ?

That is my question, where do we draw the line. Should I have not posted my thing about the "Single Ended Push Pull Amplifier" ? Is tha tpossibly pate ntable ? I deemed it not, did I cheat myself ?

it does anything undone before. If it outperforms the preceding devices, d oes that make it patentable ? This thing looks like it does about what an i ncandescent light dimmer does. If the new process can make that faster so i t can do more, but the basic parameters are damnear what was already had, a re they patenting something other than an epitaxial diffused or whatever bi polar ?

ems to me this has been done. If a device can do it better, but it still do es the same thing, what gives anyone the right to patent it moreso than say , a better OPAMP design ?

ome things that are already out there ? There are countless typres sof thry sistors out there, and they become the subject of discussion once in a whil e because it is not easy to be up on all the different types. Now they did not all come out at the same time. Did anyone patent the SCR ? The GCS ? Th e triac ?

my thing about the "Single Ended Push Pull Amplifier" ? Is tha tpossibly pa tentable ? I deemed it not, did I cheat myself ?

That's explained in background:

"Fully controlled bidirectional power switches may be utilized in devices s uch as AC-AC matrix converters to provide the converters with higher operat ing efficiencies as compared to conventional DC-linked AC-AC converters.

A fully controlled four quadrant bidirectional switch can be implemented, f or example, using two insulated gate bipolar transistors (IGBTs) in antipar allel along with two diodes in series, or utilizing two source-connected hi gh-voltage metal-oxide-semiconductor field-effect transistors (MOSFETS). Ho wever, such solutions suffer from drawbacks such as high conduction loss an d may be limited to relatively low frequencies (e.g., less than 50 kHz).

Therefore, there is a need in the art for an efficient fully controlled, fo ur quadrant bidirectional switch."

It serves a useful engineering purpose (AC-AC matrix converters). It is a s ignificant improvement over existing implementations (efficiency and high f requency operation). It is a non-obvious in that nothing else like this has been done before. That pretty much gets it.

Basically, what you patent is a device or method which is novel, useful, and non-obvious.

There are "design" patents which cover a specific form, but the more useful ones are "utility" patents, which cover *how* something is done.

For the purposes of patenting, "novel" and "useful" and "non-obvious" are all terms of art - they have a particular range of meanings to the Patent Office.

It isn't necessary to solve an unsolved problem, in order for an invention to be patentable.

It's perfectly possible to patent a new (never-before-disclosed, and "not obvious") method of solving a problem, even if there are 10,000 other known ways of solving the very same problem.

It isn't even necessary that it outperform the preceding devices. Equivalent, or even poorer performance doesn't rule it out.

It is necessary that it be *different* from earlier solutions. The difference has to be significant, and not just "colorable" (that is, if somebody patented an invention before, you can't just make a trivial change like coloring yours orange and claim it's different).

Quite a few successful patents have come along because somebody had patented one solution to a problem (e.g. light dimmer), other people wanted to compete, and they discovered alternative ways of solving the same problem and patented those methods.

Sometimes you'll look at a commercial device, study how it works, and ask yourself "Why the *bleep* did they do it in that particularly obscure and indirect fashion?". In a lot of cases, the answer was "They could avoid an existing patent that way, and could get a patent of their own on their method." Their solution wasn't necessarily better... but they could implement and sell it legally.

So - these guys claim to have figured out a way of implementing a four-quadrant bidirectional switch which is significantly different than any method previously disclosed, and is useful, and that their new method is "not obvious".

Whether that's true enough to result in a patent will be a decision for the patent examiner. Whether it's *better* enough to be a commercial success, if patented, will be up to the marketplace to decide.

Novelty in patent law means no prior art well known to the public prior to the patent application.

Are you trying to tell me that the relay and/or toggle switch is NOT "prior art"?

Unfortunately many things well known to those skilled in the art are evidently not known to patent examiners. Although these "inventions" are often fully and completely described in expired patents, for some reason the new applicants are frequently so inconsiderate that they choose a different title for their "new" "invention" from the title of the previous patent(s) for the same thing. Sometimes they are even rude enough to use different words in the description and claims. The most thorough examinations for which time is available, which might even go so far as to include a google search, are therefore unsuccessful.

If the examiner is particularly determined and does actually look for prior art, you can always make sure it won't apply if you replace the vacuum tubes of the textbook circuit by MOSFETs, and prefix each claim with "An integrated circuit including ", which will of course make it novel again.

If all of the above fails, then make sure you add some numerical values that nobody in the past could be bothered to measure. That way you will have made an innovation. For example the bread refreshing device used in US 6080436 differs from a toaster in that the temperature of the element is between 2500F and 4500F, whereas a lowly toaster's element is just "red hot" which is clearly far worse. Also whilst a mere toaster pops up the bread at some unspecified time when it is nicely brown, the proper bread refreshing device toasts it for between 3 and 90 seconds, which is much more specific. Of course this makes the new toasting process novel, because it must be if these same numerical values were not described in written accouts of past feats of toasting.

Another way to make your patent valid enough to be granted is to change the title around and submit it multiple times so that it goes to multiple examiners. Some examiners are so competent (or overworked?) that they can tell that an application is valid without even reading it. If you submit enough copies of your application (perhaps with different titles), sooner or later its inherent validity will be noticed by a sufficiently gifted or tired examiner. I suggest that for the more adventurous patent applications, five copies will usually be sufficient.