How much Detail in the Patent ?

Did you invent some electronic circuit?

If yes, then this is what you are patenting. The actual unit you made is "one embodiment" not what you are patenting.

You should include the details needed to explain your idea. Basically for a patent to be any good some other engineer should be able to understand what you are patenting. An example circuit that does what you suggest is a good way to explain things.

It is likely that you are providing too much detail about your specific embodiment if you go this far. Remember it is the idea you are patenting not the specific case of using it.

Start here:

-- Joe

Nope. Spend the money in Las Vegas instead.

Your return will be ridiculously higher.

Maybe so, but obviously the long-term return is of secondary importance for gamblers, otherwise Las Vegas would go out of business. I suspect the same is true for many patent-holders. So, go for it - you''re unlikely to get a patent if you don't try.

-- Joe

PhilJ wrote:

In the US, all that is required is sufficent detail so that one SKILLED IN THE ART can make the device (whatever it is). That does not mean that the information is sufficent for sucessful or complete or useful implimentation. Drawings, descriptions, etc are "window dressing" of the patent; what is covered is in the claims (the legal part). I will make up a ridiculous example. Say your item is a method to use relays to amplify microwaves, and you found that certain patterns do that function and that there is a commonality in those patterns that allow it to work, but you are not positive other patterns would work or work better. So you can make a simplified description of one of those patterns (not the best one, naturally) and make some citation concerning possible alterations of that pattern that makes it *possible* for someone skilled in the art to deduce a few of the other patterns. Your descriptions per se are not covered; it is what you say / describe in the claims that are the legal pins of coverage. In this case, one could describe the implimenation with latching relays, when simple reed relays may be a better medium. Mixing of relay types is "obviously" allowed, but you carefully do not hint of that fact in the descriptions ("window dressing"). The wording of the claims should say that any type of relay or switching device can be used = = broad claim, and other claims say 1) latching relays, 2) reed relays, 3) stepping relays, .... n) switching device similar to a relay (tube, transistor, etc) = = each one a narrow claim.

If i remember right, once applied for here, one has a year to do a pro-forma european application.

...and pay thru the nose, *especially* of that patent attorney has no electronic background. To keep costs down, do your own patent research and read patents closest to what you are doing (descriptions and claims). Create your own claims; be imaginative, go wild - it is only paper until it is filed. Whittle, prune and polish as needed, but be careful that some crazy claim that might be polished to useability does not get tossed out just because it looks nuts. If you are extremely lucky, a number of those patent claims may discuss "Z" that also does "Y" with the emphasis on "Z" because they were focused on "Z". But (say) with your approach, focusing the claims on "Y" where "Z" is a side-effect (that cannot be avoided) has implicit advantages. *Those* claims are unique and un-assailable and kick ass, nevermind that with present techniques the practical results may be identical. Look at my oil lifter patent and notice that reversal. Have fun.

No, not really. It takes a few kilobucks to file, and can take numerous megabucks to defend.

It also must describe the "best mode", I.e. the best known way to design/make the device at the time of the filing. The "teachings" must actually teach what you know.

Drawings cost $$ too.

You must show the best one, known at the time of the filing. The words can allude to better possibilities, though. The problem is that someone else may patent the better implementation, later, keeping you from using it.

Yes, what you *claim* is what is protected. What is "taught" is not. That's where many people get hung up in patents; thinking that the obvious stuff in the teachings is the patent. Much of it is often background to help explain the idea.

If you know that a reed relay is the better solution, you must teach it. Though you don't have to mention that it is the best.

Right! Try to protect all sorts of combinations; claim everything. Always try to throw as large a net out as the patent office will allow.

There is also a "bar date" issue. In the US, once an invention has been published (or any "commercial advantage" taken), a one-year clock that starts. After that year, one cannot file a patent. Europeans have no such clock. Once published you are barred from filing a patent.

On Jun 21, 8:38 pm, krw wrote: [.. patents ...]

You can make them yourself in something like Qcad and then print them on a laser printer. They need to be clear and fairly neat.

In the US, you usually have a prefered embodiment. You can show a nonprefered embodiment "to make the explanation clear". Explaining a digital design with an analog circuit is an example of this.

A good patent lawyer will make sure that you say you know of many variations on the design.

My point is that the patent office charges for them, no matter how they're drawn. Sure, anything that works. The lawyer took my crude FrameMaker drawings for one. He also lifted 90% of my words. I wish I got paid 9x what the lawyer did.

Absolutely. The point is that a "preferred embodiment" must be shown if there is one. The inventor must disclose the "best mode" of the invention.

A good lawyer will help expo lore alternatives that the inventor might not think of.