A few questions

[Sho'Nuff]

Hey all, this might not be a "highly" technical question by today's standards, so bear with me.

Being a patent attorney, I often have ideas for various inventions. Most of those ideas I can't pursue, either because they are derivatives of a client's work, or because they have already been invented (that happens more often then you would think). However, I now have an idea for a product that doesn't appear to be barred by either of those issues. I am attempting to build a prototype of the product at home, and am running into some difficulty.

Basically, I am looking to induce a 12-30 inch long blade like object (e.g., a thin stainless steel or aluminum blade) to vibrate at a specific frequency or range of frequencies. The closest examples I can think of are: 1) a tuning fork, where when the fork is struck, it vibrates at a particular frequency, thereby producing an audible response; and 2) the vibrating blades used in modern multi-blade razors, e.g., the vibrating razors produced by Gillette.

Any ideas on how one could go about accomplishing this? Is there a way to calculate what frequency an object will vibrate at when exposed to different types of energy? It seems to me such a calculation is possible. In fact, I am pretty sure my first year physics class may have discussed it when we talked about harmonic vibration and the Tacoma-Narrows Bridge, but it has been almost 15 years since I took that class.

Thanks for your input in advance.

 

[leadpaint]

Well, this is coming from a 15 year old, so this is pretty much redneck style.

Assuming you already have this object, just clamp it down somehow (I'm assuming only at one end) and start hitting it. You could probably quite easily figure out its resonant frequency by hitting it at different frequencies for a while.

If you want to calculate this, then I can't offer any help. Best of luck to you.

 

[PottedMeat]

A tuning fork and vibrating razor are two different things - the fork vibrates at its fundamental frequency based on its dimensions/material. The razor I believe is forced by a electromagnet - 60Hz AC goes into the coil, the loose razor floats above and follows along.

How far do you want it to oscillate? What frequencies? How heavy is the metal?

 

[CycloWizard]

Such an invention does already exist, though with much smaller blades. It's called a vibratome (not sure if that is the trade name or the general name) and is used for sectioning tissue.

That said, it is possible to compute the fundamental frequencies of objects to an arbitrary degree of precision using finite element analysis. There are a few important factors when one is generating the model: the mechanical properties (elastic modulus and Poisson ratio, if it is perfectly elastic), the geometry, and the boundary conditions (i.e. how it is held down on any side).

 

[Sho'Nuff]

Originally posted by: CycloWizard
Such an invention does already exist, though with much smaller blades. It's called a vibratome (not sure if that is the trade name or the general name) and is used for sectioning tissue.

That said, it is possible to compute the fundamental frequencies of objects to an arbitrary degree of precision using finite element analysis. There are a few important factors when one is generating the model: the mechanical properties (elastic modulus and Poisson ratio, if it is perfectly elastic), the geometry, and the boundary conditions (i.e. how it is held down on any side).

Interesting, the vibratome is somewhat like what I am thinking, but as you said, with a much smaller blade. My idea also relates to a completely different application then tissue slicing.

I'm sorry to be so vague, but this is my first really decent idea for a new product (provided it works), and I may want to patent it eventually. If I describe what the invention is, I risk losing the ability to patent it at some point.

 

[jagec]

Ultrasonic transducer bolted to a blade.

Feed it different frequencies of current to alter the frequency of vibration. Obviously you'll have complications due to the natural frequency of the blade, but that shouldn't really matter.

 

[wwswimming]

modal analysis like this is easy with ANSYS or NASTRAN. i haven't used
COSMOS, which is a lot cheaper. Finite Element Software.

it takes some skill to model what you think you're modelling. you might
represent a clamped geometry as a welded joint, whereas a professional
physicist might model it as 2 groups of pinned nodes.

anyway, if you can get a 30 day demo or COSMOS or ALGOR (yes, pronounced
just like Al Gore), it would give you a chance to model the geometry you're
imagining.

 

[Rubycon]

Originally posted by: wwswimming
modal analysis like this is easy with ANSYS or NASTRAN. i haven't used
COSMOS, which is a lot cheaper. Finite Element Software.

it takes some skill to model what you think you're modelling. you might
represent a clamped geometry as a welded joint, whereas a professional
physicist might model it as 2 groups of pinned nodes.

anyway, if you can get a 30 day demo or COSMOS or ALGOR (yes, pronounced
just like Al Gore), it would give you a chance to model the geometry you're
imagining.

ALGOR is awesome! :thumbsup:

 

[PCBeal23]

My solution to this would be Quartz, though the frequencies they resonate at when given a charge might be too high and possibly melt the material they are attached to.

Also, for an FEA, I recommend Inventor's FEA, as they were recently (in '07) augmented by the formulas from a company called Plassotech (which I used to work for).

Autodesk bought Plassotech shortly after I left the company and integrated their algorithms and functionality into inventor. Plasso's originaly FEA softwre was called 3G and by far one of the industry leaders in FEA. Unfortunately, I did not pursue my degree in Mech. Eng., so I wasn't able to play with the software that much.

 

[TuxDave]

Well, if you're using a uniform length of something, can't you just use speed of sound through that particular material, its length and whether or not you're clamping one end or both to determine valid standing waves in it?

If that type of information is not available or it's irregularly distributed, I would just do the whole "hit it with a hammer" and record what set of frequencies persist the longest.