Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SSS148 on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Seismic Shake Table - Help For A Non Mech Guy 1

Status
Not open for further replies.

sticksandtriangles

Structural
Apr 7, 2015
494
I want to make a semi-pro level seismic shake table, but I have no clue where to even start.

What I what:

- Motion control in the x and y direction​
- Motion control in the vertical direction if at all possible, but very willing to settle with motion in just the x and y​
- I want to be able to program an arduino/raspberry pi type element that can read displacement/velocity/acceleration vs time data and move the shake table accordingly​
- I was thinking a scale of around 4'x4' but willing to work with this dimension. Still not sure the weight that I would be pushing, but the weight of the structure above would probably be 10 lbs and the subassembly being shaken below would be 20 lbs if I had to guess.​

I have started to do some preliminary research on this topic and this website with actuators looks promising:


Speed:
Programmable:

Questions that I have though are the following:

- These actuators do not look to move fast enough to capture the fast ground shaking that an earthquake typically represents (I would love to be able to input an El Centro type ground motion into this thing Are there any actuators that move this quickly?​
- How programmable is this actuator? Can I input lots of data and get this thing to respond properly under crazy earthquake accelerations​
- Is there any other product I should look into? Let's for now assume budget is not a concern... feel free to fire away with thousand dollar ideas​


Appreciate the thoughts!




S&T
 
Replies continue below

Recommended for you

That's a thing, that if it works to normal industry standards, costs about as much as a decent house. I like your actuators. I'd worry about third order harmonics etc, they appear to be a motor and gearbox driving a ball screw.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Can you help me out with the third order harmonics comment? Not sure what that means.

A little bit of googling looks like that this maybe an issue with power supply?

Any other actuators that may perform better than the one I linked? My pockets could go a little deeper than the 70 bucks listed on this website.

S&T
 
So you are building a shaker table and don't know about harmonics? Hmm

When a mechanical system reverses direction it tends to stick slightly at the point of reversal. This adds distortion to your waveform, which comes in as third, fifth etc harmonic of the fundamental tone.

That's why even quite powerful shakers are often built using voice coils, basically loudspeaker coils and magnets. These are able to reverse direction smoothly. Hydraulic shakers are also common but they suffer slightly from distortion. Sometimes people add dither to the input signal to suppress this, with various degrees of success.




Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Thanks for the feedback gentleman.

I am a structural engineer... when things move and ΣF ≠ 0, I struggle to understand these concepts [lol] That's why I am building this shake table after all.

I will look into the voice coil actuators as well to see how much one of these guys would cost.

I sent out some quote requests to geared linear actuator companies with the following criteria:

-Max velocity = 300mm/sec​
-Change directions at a max of around 5 times a second​
-Fully programmable movement​
-Moving about 30lbs of weight​

We will see if any of the geared linear actuators can achieve these sort of parameters. I wonder if they will mention any of the deficiencies you guys have mentioned.





S&T
 
Google "oshpd seismic testing" - to see what California's structural code requires (3 axes testing, simultaneous); using the image search shows the size of the stuff being used by commercial labs. We tested our 5-lb electronics box, which got mounted to a simulated wall and gypsum board, at a certified lab. The shaker table was big enough to park a small car on, and used hydraulic actuators.
 
IRstuff, the 60 lbs force is what I was targeting, assuming I wanted an acceleration around 2*g.

Where did you get the $300 price from? Are you looking at a specific product or speaking from experience? I have one quote in the range of $700.



S&T
 
Not to burst any hobby bubbles, but you might be better off buying a small used commercial shaker than building your own. Like most industrial equipment, they either sell for big bucks or get sold for scrap in which case you'd likely save quite a bit vs buying the steel (read: mass) necessary. I'd keep an eye on the websites of a few of the larger industrial surplus dealers and assume whatever you buy will need a controls upgrade. Be forewarned that most folks in that line of business are somewhere between used car salesmen and scrap dealers, used shaker specific dealers and rebuilders not included of course.
 
We are not going for any commercial sized shaker here.
I am thinking something cool I can take to a middle school to get kids excited about structural engineering.

Maybe I keep an eye out for a used 3d printer'ish type actuator that I can knab for cheap. Prices I have been receiving have not been as bad as I expected either. Most are in the range of $500 - $1000 and I would be more than willing to drop that type of cash for some hobby excitement.

Still debating between which type of actuator I want as well. I have some good fast moving geared actuators and (1) voice coil actuator I have been looking at. I am also debating looking into a linear motor type actuator as well.

Lots of decisions...



S&T
 
It also looks like I need to buy a motor and a means to control the motor. I assumed that the actuator I was going to buy came pre-packaged with this and a programmable interface to control the actuator.


I am looking at this type of motor.

I've reached out to these guys to see how they typically control their motor. I am guessing some arduino type element that tells the motor to give power for x amount of time.

This is starting to be more complicated than what I imagined. No wonder the pre-packaged shake tables are in the range of 20k.

S&T
 
For your purposes I suggest that the best solution is to use a variable speed motor that drives a crank arm connected to a platform. You can only get sinusoidal motion. That will be far simpler, cheaper and more robust, and adequate for showing resonant frequencies.
 
I was really liking this actuator:
www.thomsonlinear.com

Needs a motor to be driven and means of controlling the output.

You might be right compositepro, this may be too complex and expensive to be worth the trouble.

I really wanted to input El Centro into this device though [cry]

BTW the voice coil actuator quotes are coming back really high, in the range of 4k-8k.

S&T
 
sticksandtriangles

If you are just doing this to inspire the youngsters, you can probably leave out some of the subtleties. If you really want to simulate El Centro there is a lot of high frequency stuff in the spectrum which would in my mind make sense with the coil based solution. If you could strip back your spectrum to 5Hz and slower, perhaps a geared system could work.

I am imagining a big servo motor, a belt, a slider and an arduino. Surely this would work at least well enough for the stated purposes...

In respect to backlash, insofar as its actually a problem, could you not precompress the system with a spring if it was a problem? You need a bigger actuator but its just an actuator, a controller and a slider.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor