"Fresh" patent granted for JFET linearization technique?

[stm]

Am I missing something, or the mega-old JFET resistance linearizing trick used in MXR's Phase 45 (and also presented in National Semiconductor's Application article AN-129 from the 80's) has "just" been patented as of February 19th 2002?

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=6348834.PN.&OS=PN/6348834&RS=PN/6348834

[MartyMart]

I'm no patent expert Sebastian but that doesn't seem "patentable" to me !!
If it's an "app note" circuit detail anyway, how the heck was it patented ?
I dont see how you can patent a circuit to effect the performance of a known "device"
anyway ! .... it's not a newly designed "product" but just an application of something that
exists.
Isn't it like a patent for a "new improved" stawberry lollipop, which is better tasting and a
more consistant "thickness" due to an additional manufacturing process called "super lick"   

,,, can you reinvent the bias/performance details of a device a second time ........

[stm]

After reading and re-reading the patent text I don't see anything new with respect to the original circuit.  I was hoping that someone told me "no, here's a difference and/or improvement".  I agree with you, Marty, it should have been considered prior art, unless this fact escaped from the people in charge to verify this.

What I find really amazing is that someone considered it was profitable at this time in history to patent such thing.  I understand filing a patent may cost over $10k, if I'm not mistaken, so it makes no sense unless you are expecting some big revenue from this.

[d95err]

I understand filing a patent may cost over $10k, if I'm not mistaken, so it makes no sense unless you are expecting some big revenue from this.

...and the problem for everyone else is that if the patent owner sues somebody for using this technique, it will cost a lot more to defend themselves in court, even though the patent is false and borderline fraudulent.

[db]

This technique is nothing new.  As well as AN-129, try looking in any good electronics text book e.g. Horowitz and Hill Chapter 3 section 10 p139.

No-one would have to defend themselves in court I would have thought.  I haven't read the full patent and some of the diagrams are hard to read on a monitor (extremely large for some reason) but it does seem that the jist of the "discovery" is as we think.

Perhaps someone should point this out to the author and/or the patent office?

[Sir H C]

Do these other processes modulate the gate voltage on the FET?  It seems they are attempting to allow the linear range for the FET to get larger by modulating the gate-source with a bit of incoming signal so that the non-linearities of the fet-resistor are cancelled.  I don't know how well such a technique will work in the real world though.

[Ben N]

It costs money to prosecute a patent, too. Even if the patent examiner couldn't see the obvious, the attorneys the patent owner hires to prosecute it will., when they consider the chances of prevailing on such a weak claim.

Ben

[Paul Perry (Frostwave)]

Once upon a time - long ago, maybe 20 years - governments employed patent inspectors to check whether stuff was 'new'. Einstein used to do this when he was young.
Today, the approach is to let anyone patent anything & let the courts fight it out. A guy here patented the wheel: http://edition.cnn.com/2001/WORLD/asiapcf/auspac/07/02/australia.wheel/
As you would expect this leads to a lawyer's picnic.
Nothing makes lawyers happier than to fight.
In fact there are companies - themselves owned by lawyers - who do nothing but buy up patents in the hope of tracking down people who might be infringing them & sueing them. Since thety are getting the legal costs at wholesale rates, they are in a position to grind anyone down. It's a real mess.

www.tinaja.com/glib/casagpat.pdf for some advice on patents from an electronic guru.

[Sir H C]

I have a friend who is a patent inspector.  They are improving on the problems that plagued the system in the past.  He has a PhD in microbiology and works on patents in that area only.  THey have specialists in all these different areas to try to keep the people in check.  I still am not sure if this is not original, but I don't think it is that useful as the techniques used look like they would require tuning and trimming to keep everything working right.  There are other solutions that work better, so this could well be someone who has something original that has been leapfrogged by everything else. 

Remember it is not using a FET as a resistor but keeping the range that the resistor is linear much greater than seen in Phase 90s and the like.

[Ben N]

Nothing makes lawyers happier than to fight.

Hey! I resemble that remark! 

[stm]

Remember it is not using a FET as a resistor but keeping the range that the resistor is linear much greater than seen in Phase 90s and the like.

I disagree with "the like" options.  The Phase 45 does employ the linearizing technique for the FETs--is this pedal from the 70's?.  I bet MXR didn't do it on the Phase 90 to reduce parts count and cut costs down.

[Sir H C]

What linearizing techniques does the phase 90 use?  LFO goes to the fets, right, what other signal.  The patent talks about putting a resistor across the FET and the detriment to dynamic range of the fet by doing this.  They are doing an active technique with the input signal modulating the gates to keep the resistor linear.

[analogguru]

This technique really isn´t new.  Have a look at the Urei 1176 or another Fet-Compressor/Limiter.
Also the Ibanez Jetlyzer or Roland AP-2/AP-5/AP-7 used this technique.  Even the Orange squeezer has this feedback network (R-C to gate) to get a bigger headroom.

For economical reasons you will not find it on every FET-phaser. For what ?  this (additional) little distortion no one will listen on an (most of the time) already distorted signal.

analogguru

[brett]

In fact there are companies - themselves owned by lawyers - who do nothing but buy up patents in the hope of tracking down people who might be infringing them & sueing them.

One Australian company (who basically invented modern power steering) makes as much money out of lawsuits as it makes out of licensing fees.  Apparently, many auto companies don't know (or pretend not to know) that power steering is patented.

On the other hand, a friend has an engineering business where he is scanning patents all of the time, and using them.  He makes more money than anyone I know.  If he makes one of this, and two of that, he might never get busted. 
cheers

[analogguru]

You will not believe it, but in 1996 a pure tube expert from Erlangen in Germany patented an Tube-preamplifier:

Patent DE19647875 (short)

Patent DE19647875 (full)

even if this technology is well-known since more than 50 years:

Patent CH266522A (full)
Patent DE848661B (full)
Patent FR1015580A (full)
Patent GB484102A (full)

I don´t know why, but it remembers me a little bit to the tube stuff of Electro-Harmonix.

analogguru

[puretube]

You will not believe it, but in 1996 a pure tube expert from Erlangen in Germany patented an Tube-preamplifier:

Patent DE19647875 (short)

Patent DE19647875 (full)

even if this technology is well-known since more than 50 years:

Patent CH266522A (full)
Patent DE848661B (full)
Patent FR1015580A (full)
Patent GB484102A (full)

I don´t know why, but it remembers me a little bit to the tube stuff of Electro-Harmonix.
analogguru

Funny (?) thing is, that those who only superficially look at those schematics and those claims
indeed might think they find prior art from those old patents (the examiner held against that newer one)
in that invention...

It`s the DETAILS, that count...

[George Giblet]

There are problems with the usual linearization with resistors around the JFET but the methods where the feedback is applied from the output of an opamp instead of the drain has been known for many years eg.

http://sound.westhost.com/project67.htm