Vibration Generation advice needed

[GoodVibe]

Ok, call me unbalanced, but I need some advice on generating a vibration at between 60 and 80 hz. I'm thinking an Electromagnetic shaker might be the way to go, but I am open to any and all suggestions.

The key is variable intensity or eccentricity at a fixed frequency. Linear oscillation is preferred, but if there were a reliable way of varying rotational oscillations, I would be very interested.

If the device could generate oscillations/vibrations at different wave patterns, that would also be excellent. The ability to switch between a basic sine wave and something more square edge would be very cool.

A final requirement is size. Ideally, the solution would be able to fit in a cylinder 1" or less in diameter, with the oscillations induced perpendicular to the length of the cylinder.

So, any suggestions?

 

[GregLocock]

Electromgnetic is the obvious one. You'll have to pay to get a 1 inch diameter, but it is feasible.

You haven't specified what force magnitude you need.

Incidentally, because you have failed to specify so many things I can satisfy your spec with a 3V DC electric motor out of a toy.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GoodVibe]

Greg,

We're a little unsure on the exact force magnitude. In general, with a variable force of displacement device, we would look for a displacement of around 1-2mm, which would give a acceleration (rms in g's) of 5-10 at 60hz, or .13 to .26 m/s velocity rms).

The small dc motor toy shakers work except for one thing, the force is locked to the rotational speed. We want to be able to vary the force/displacement at the same rpm or hz. That's the tricky part.

Is there any other detail I can provide?

 

[GregLocock]

You still haven't specified a force.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GoodVibe]

Right you are.

I think somewhere in the range of .1 to .5 newtons would do the trick, if that is possible.

Sorry about that.

 

[GregLocock]

OK.

Have a look at the small B&K acclerometer calibrator.

http://www.bksv.com/pdf/BP0184.pdf

It will shake a 70g accelerometer at 1g, so that is 0.7N

It is a little larger than you specified, but the electrical and magnetic circuit is about right to give you some ideas.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GoodVibe]

Looks like the force requirements would be taken care of, however, the displacement of 10?m is quite a bit less than the 1-2mm we need. It sounds like it is using a piezo drive, which we had looked at but were unsure if they could deliver the displacement requirements.

Hmmmmmm....

 

[GregLocock]

In that case you'll have to go for a voice coil system I think.

Your 1" OD restricts you to a fairly small magnet and coil.

designing an idiot proof suspension is the hard part.

Most small shakers are voice coils, in fact I've only ever used a couple that weren't.

What are you trying to shake?



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GoodVibe]

Greg,

Voice coil is looking better and better. The 1" limitation does make things tricky, but the nature of the product dictates it. I wasn't sure whether to mention the application given the forum, but what the heck. The application is adult in nature and female oriented (if that's to obscure, message me and I'll send you some more details). When analyzing the products on the market, the primary source of vibration is an unbalanced dc motor. The problem with such devices is that the frequency of the female response is very narrow, (between 60 and 80 hz if you haven't guessed) but individual preference dictates variable power. 99% of the devices do this by increasing RPM, usually right past the 80hz sweet spot. The one device that has a solution to the problem is an oscillating electric toothbrush like device that is extremely (and I emphasize extremely) effective.

There are many other engineering challenges, such as battery life, waterproof container, durability, etc... but the big one is variable power at a fixed hz.

I guess it is up to the reader whether this problem is as sexy as a camshaft, but it sure keeps me up nights.

 

[GregLocock]

Ah, I suspect you are actually aiming for a surface vibration level rather than a force then. Should be easy to measure.

What do mobile phones use as vibrating alerts? I'm guessing unbalanced motors. They must be tiny, so you could just have 4 (or whatever gradation of force level is required) of them, phase locked somehow.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GoodVibe]

Now that is interesting. I've though of phase locking two motors, but wasn't sure how practical it was without using stepper motors. The phase lock doesn't do much without being absolutely sure of the position.

If I had an accelerometer handy, the target surface vibration would be very easy to measure. Maybe it's ebay time.

 

[GregLocock]

Maybe some sort of optical device? one led one receptor and a lot of logic?

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[electricpete]

Did you volunteer to assist in the field trials of your product?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[GoodVibe]

But of course (or at the very least my wife and some female friends will, though not all at the same time). Product testing is the best part of the business.

Testing the competition has been interesting. Only one has actually caught on fire, but a few have started smoking.

On the original subject though, it looks like a voice coil might be the way to go. I'll be pulling apart a few hard drives to do some experimenting. The acceleration and frequency response on those puppies are pretty impressive, and might even be worth rethinking the form factor.

Also, if your wives or girlfriends ever give you any funny looks when you're trying something new in the bedroom, just tell them you are attempting to hit 80hz.

 

[Qmavam]

I'm curious how you arrived at 80hz as being the sweet spot. (Maybe I should say how accurate is your research and did it cover a wide demographic?) I was thinking variable frequency from 25hz to 300hz, with variable power. I also wonder about sine, square and triangle wave drive. I've experimented with strong magnets and coils wound on a (round solenoid style) bobbin. Heat is the first problem that crops up. I thought about two coils wired with opposite phase pushing against each other.
I have a low wattage audio amp for power and a variable
oscillator to drive the coil. I need a higher wattage amp and more turns on the coil.
Anyway, I'm interested in your research!

 

[GoodVibe]

Qmavam,

Please forgive my short sentences, broken collarbone (and no, not from product testing)

The 60-80 hz sweet spot has been verified in several university studies and is pretty well known in the industry.

We absolutely want variable frequency with power though, as i suspect each user will vary, some quite a bit. I also wonder if there are biological harmonics that can be tapped in to.

25 to 300 hz would be a minimum range to shoot for. If it can handle the heat, I will shoot for 5=500. Wave shape has been the big breakthrough. I plan on building a 1-5 watt audio amp and driving it with standard audio circuitry and inputs. Think ipod jack. That allows all sorts of interesting complex waveforms. Maybe a high power 10hz combined with a low power 80hz wave?

As far as form, the voice coils I pulled this week from hard drives are looking very interesting. I have a form worked out that is non-cylindrical but a contoured triangle. Will work up some drawings as soon as I figure out how to draw with my left hand. The bobbin arrangement you described is also a very strong possibility. My concerns are heat, and whither mechanical dampening would need to be used.

 

[rorschach]

Is heat such a bad thing in this application?=D I would assume that we are not talking about continuous duty here (several hours). for safety, it should be possible to thermally limit the device to lower power operation if overheating is an issue.

 

[rorschach]

BTW, DAMN I envy your job.....

 

[IRstuff]

There seems to be a disconnect between your desired amplitude and g-level.

amp*(2*pi*f)²=max g-level for sinusoidal vibration.

An amplitude of 2 mm at 60 Hz requires 29 g acceleration, conversely, a 10-g acceleration at 60 Hz will only produce a 0.7 mm displacement.

Your stated velocity targets are more in line 10-g acceleration, since amp*2*pi*f = 0.26 m/s with f=60 Hz and amp=0.7 mm

TTFN

 

[Qmavam]

Regarding complex waveforms; I question how much
effect they would have. My reasoning is, the coil (inductor) is trying to smooth the waveform that you apply and the mass you are moving will be slow to respond to force and continue movement when force is removed.
You mentioned a 1 to 5 watt amplifier, I suggest starting
a little larger, maybe 10 watts. You can always downsize
if you find you don't need the power.
Are you shooting for battery power? Might need to look for an efficient amplifier later.
I would like to see ideas in drawings.