Electromagnet Vibration unit using a Neodymium cylinder.

[jhudler]

I have a project creating an epoxy based granite material.
The sphere packing density is quite high with aggregate sizes from 2 to 600 microns.
So that means I'm left with a very thick mix.
Plus the added complication of doing this in a vacuum.

Requirements:

I need to make a shaker/vibration unit (freq from 10 to 2000 Hz).
Using a motor to vibrate has been tried and it overheats... go figure !

Materials:
1 inch diam x 2 inch long Neodymium magnet.
Lots of magnet wire and even a coil winder.

Supposition:
It seems to me that I could mount magnet to base to vibrate, then create a coil around the magnet (with suitable separation) and drive that? I can dissipate the heat with a simple water jacket created from aluminum tubing and encapsulate that with an epoxy mixture designed to conduct heat which would be mounted to the rack in the chamber.

Question:
Is it possible to create a simple voice coil using a cylindrical magnet?
If so given almost 13000 Gauss, how do I calculate the coil parameters (windings,gauge,amps,voltage).

Or do I need to resort to using Axially magnetized magnet setup found in speakers?

Or perhaps there are other ideas.

Thanks,
Jack

 

[MJR2]

Jack,

You leave off a few things.

What deflection do you require?

What is/are the weight(s) of the objects you wish to vibrate?

You are not likely to needs lots of magnet wire as the frequencies you want to operate at and the small permanent magnet will limit your motor.

Have you considered devices that will work near their natural frequency?

Mike

 

[jhudler]

Mike,

Ah yes I did leave that out and one of the things I wanted to get clarified.

Weight can range from 10 lbs to 150 lbs (perhaps 300 depending on density).

Deflection should be in a range of .05 to .25 inches (this is what I was doing with the motor driven version).

I can also supply any power requirements to drive this... if we make it a 4 ohm coil, I could hook it up my 500 watt amp and play AC/DC... err... drive it with a signal generator.

I guess it would be helpful to know what the limits are given a reasonable amount if heat dissipation.

Essentially the mass (epoxy with quartz and ceramics in a mold mounted to a spring loaded table in a vacuum chamber) needs to be vibrated to assist in deairing the mix which then fills in the corners, edges, and sets the final packing density (and all the while maintaining the right vacuum to keep the epoxy from boiling).

Some experimenting is obviously needed to obtain empirical data for just how much we should shake this stuff. In doing so I find the required range of frequency and velocity.

Then I can spend the money (or not) to purchase off the shelf vibration units.

Thanks!
Jack

 

[IRstuff]

I'll take stab and guess that you'll overheat as well.

It's not the type of motor, it's the SIZE of the motor. The vibration testing industry uses shakers that can shake several hundred pounds of equipment, but that'll cost you. The shakers that we use at work are about 2/3 the size of an oil drum:

http://www.qualmark.com/Products/edShakersModels.html

TTFN

FAQ731-376

 

[NickE]

.25" at 2000Hz is going to reqire much more than 500W.

Especially considering that Audio amplifiers are rarely rated to absolute sine wave outputs.

I use a small (bigger than a coffe can, smaller than a 5-gal bucket) shaker that can hit 50lbf, its very big and very heavy. Max current uncooled is 2.9A, 5A air cooled.

Your best bet would be to grossly oversize a normal drive motor. 1HP ~ 760watts (100% duty) so you can determien how big the old motor was, gearing might help too.

good luck.

Nick
I love materials science!

 

[MJR2]

If it is in a vacuum chamber you will need to water cool for heat removal. Maybe that is why your motor overheated.

IRstuff's link looks to be a good direction for you.

Mike

 

[jhudler]

The motor (new) ran perfectly in normal atmosphere.

What about using more four, one one each corner of the frame?

The audio amplifier comment really wasn't a serious one, just thinking out loud.

NickE, how did you determine power requirement of .25" at 2000Hz?

I guess the bottom line question (short of me winding up something and trying it) is; is there any benifit to using a Neodymium cylinder as the core? Or should I stick to modified solenoid?

 

[jhudler]

IRstuff's links do look good, but are rather large.

Even the motorized version took up considerable space.

I can water cool the windings on the coil.

What's driving this is a need for compactness.

Not having any experience with magnetic voice coils is reason I'm asking this question.

Thanks!

 

[badservo]

Can you try a hydraulic motor?

 

[jhudler]

That's an idea... only no hydraulic stuff around here.

 

[EdStainless]

Hydraulic is my thought also. Either oil driven motors or cylinders and control valves.
With whatever driver you use you should have a system with adjustable weights so that you can take advantage of system harmonics.

= = = = = = = = = = = = = = = = = = = =
Still trying to help you stop corrosion.
formerly Trent Tube, now Plymouth Tube
eblessman@plymouth.com
or edstainless@earthlink.net

 

[NickE]

.25" at 2000Hz, lets use a middling mass of oh 100 _lbs.


Note: I'm not particularly good at this, I just happen to run a shaker motor for certian types of testing.

1/2000s is the total time to complete one cycle. (you dont say wether you need +/-0.25" so we'll go with total travel of .25" so +/-.125")

Dist= .021ft
TSD calcs: Time=.0005sec Speed(Avg)=41.7Ft/sec Dist=.021ft

using some basic guesses WTR max velocity and the shape of a sine wave we can say that max velocity is ~1.414*Avg Velocity....

So the mass must acclerate to 60ft/sec in .00025sec.
giving Acceleration as: 235855ft/sec^2 (or approximately 7000*G)

Requiring aproximately 2.35x10^7lbf*Ft/sec^2 or 5896ft-lb/sec or approximately 8000watts. If you can get better than 50% efficiency out of a voice coil I'd be reallly suprised. About a 120Hp or bigger motor should do the trick.

Now lets do the same math for 2Hz.
Time=0.5sec Speed(Avg)=0.042ft/sec Dist=0.021ft
Vmax=0.060ft/sec
Acceleration=.237552ft/sec^2
23.8lbf*ft/sec^2 or 5.938ft-lbf/sec or only around 8watts.

At 20 Hz (the official lowest threshold of heard sound) we should see around 80watts required.

Note: I dont really think that I am right with these calculations, however you can see what happens when you increase the frequency by steps of ten. The power required also goes up as quickly.

(I'm also an audio engineer, my biggest system has a power output btw 200-2000hz of ~1200watts, the speaker cones dont move anywhere near 0.25", Heck I dont think the 1500watts btw 60-200Hz even get that much movement, the 1400watts at 35-80Hz just might, but they're not horn loaded so they arn't pushing 100lbm back and forth.)

Nick
I love materials science!

 

[jhudler]

Well I'm trying to gently vibrate an epoxy aggregate mix, not experiment in interstitial fluid boundaries from cavitations of a dense polymer matrix.

It's reasonable to move a 1/4 inch (that's sloshing) at very low frequencies; quite another higher up.

Using your calculations @ 1/64 of an inch it's ~500 watts, again that is still too much movement and probably too high a frequency, but the power cost is in the reasonable zone for experimentation.


Thanks Nick!

Now how to make an electormagnet that will deliver this movement?

 

[syapnh]

Several years ago I was involved in a project concerning the mixture used to make electrodes for aluminum hot electrolysis. The mixture was very thick and the only way to obtain a compact filling of the form was to generate shock waves: a profiled eccenter lifted the filled form which then falled on a table.
A specially designed servo hydraulic generator (sinus accelerations) did not bring the expected results.
On top the form a "hammer" helped to obtain the compressive force. When the combination between frequency and amplitude was tuned so that the down acceleration became > g and the filled form met the free falling mass at the upstroke the results became better since the acceleration of the falling mass upwards lead to high forces on the contact area.

I think, based on this experience, that it could be better to use a movement profile generating high accelerations as for instance a triangle. For such a profile the power requirements are very high since the mass has to stopped and accelerated in the contrary direction within a short laps of time. I assume it is better to have acceleration peaks than high frequency since the product viscosity will absorb the energy and may be not fill totally the form.

 

[jhudler]

Solved this problem by not using magnetics. We cored the bottom of the vacuum chamber and placed a thick rubber disk with compression seals. Used it as a diaphragm and pushed on it with a eccentric cam on the outside and attach the inside via plate and rod to the table.

This is not for UHV and therefore don't care about permeability of the rubber.