[ANSS-netops] Memphis shake table

[John R Evans]

Hi All,

On behalf of Mitch Withers, Greg Steiner, and myself, here is an  
exchange of the shake table available at CERI in Memphis:

------------------------------------------------------------------------ 
----------------------------

John,

Thanks for the posting on accelerometer testing.

FYI, CERI has an MTS hydraulic shaker that operates from DC to about  
100 Hz (accurate to about 50 Hz).  This shaker has no seals around  
the piston to reduce hysteresis in the table motion (scavenger pumps  
suck the oil out of the end caps).  The table rides on high pressure  
oil bearings and is pretty smooth. Vertical and horizontal motions  
are possible, but rotation around the horizontal axis is uncontrolled  
[when vertical?].

While I can't speak for Mitch anymore, I'm sure he would make it  
available to you, if you need some high amplitude shaking.

It has a peak acceleration of 7G at about 10 Hz and a maximum stroke  
of 2" p-p.  The main problem is the background noise of all the HVAC  
units in the building.  We  routinely did our cals in the middle of  
the night  to get about a 12-18 db improvement in background noise.   
[Table] calibration was checked occasionally using a micrometer and a  
recently calibrated LVDT.  Accuracies of 1-2 % were possible using an  
FFT to single out the motion of interest and averaging over 10 to 100  
cycles (pretty damn slow at 10 seconds period!).

Greg

------------------------------------------------------------------------ 
----------------------------

Hi Greg,

Thanks for the info.  Sounds like a very interesting instrument,  
particularly at high frequencies.

On the Anorads we can push 5 kg to 4 g at 32 Hz, but I smoked one of  
the linear motors by running it at 64 Hz for an extended period --  
one can get only 5-10 s safely at that frequency.

I think air or oil bearings will be essential to creating any shaker  
that is quiet enough for definitive cross-axis work and that long  
baselines between bearings will be needed to constrain rotations  
tightly enough, so I was gratified to hear of your oil bearings --  
how much trouble are they to maintain?

Cheers,
John

------------------------------------------------------------------------ 
----------------------------

John,

The table actuator can be mounted in either vertical or horizontal  
position.  It can shake up to about 150 lbs.  They have a an overhead  
crane with a chain winch that picks up the weight of the actuator,  
then you can tilt it and reinstall the mounting bolts in the other  
plane. The table sits on a concrete reaction mass and there is a  
gussetted right angle plate with 2 sets of mounting holes mounted to  
the reaction mass.

The oil bearings do not require any maintenance other than every 10  
years or so replacing the hydraulic fluid.  They are a standard  
design feature in all hydraulic shakers -- the custom part was the  
removal of the o-ring dydnamic seals.  The only problem i had was  
once forgetting to turn the scavenger pumps on, in about a minute you  
have a big puddle of oil around the base of the table!

I hesitated to post it to the blog, since I retired from CERI and  
didn't want to volunteer Mitch or the CERI folks for something . I  
just thought it might come in handy if you neede it for some testing.  
feel free to pass it on to interested parties.

Greg

------------------------------------------------------------------------ 
----------------------------

The shake table is certainly open to anyone at ASL who wants to use  
it but as Greg said, you have to do it in the middle of the night.   
We're located in the middle of a 1M+ population metro area on top of  
a kilometer of sediments that shake like jello during the day.  And  
then there's the railway within about 300 meters or so with about 40  
trains per day.

It hasn't been tested let alone NIST traceable calibrated in several  
years.  I guess we "trust" the notion that today's sensors have fewer  
moving parts and are more predictable (when they're working) than in  
years past.  Besides, it would take ten years to swap out all the  
sensors in the network for recal.

Mitch

------------------------------------------------------------------------ 
----------------------------

Cheers,
John

John R Evans
jrevans at usgs.gov

Usually in Menlo Park:
U.S. Geological Survey
345 Middlefield Rd, MS-977
Menlo Park  CA  94025-3591
650-329-4753

Intermittently in Albuquerque:
Albuquerque Seismological Laboratory
U.S. Geological Survey
P.O. Box 82010
Albuquerque  NM  87198-2010
505-846-1793








-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geohazards.usgs.gov/pipermail/anss-netops/attachments/20090626/43992906/attachment.html