Good Vibrations

QNX Newsgroups (Archives) qdn.cafe
1

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?
  2. What can we do to reduce or eliminate vibrations inside a computer
    rack?
2

Bill Caroselli wrote:

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?

You could use an accelerometer.

  1. What can we do to reduce or eliminate vibrations inside a computer
    rack?

Well, assuming you have 3 accelerometers (one for each of X,Y,Z
axis), and leveraged piezo elements isolating the frame of the
computer from its mount, you could implement an active stabilization
system (basically measure the vectors from the accellerometers, and
issue equal but opposite movements using the piezo elements
(depending on the amplitude of the vibrations peizos may not have
sufficient travel, so another means of imparting motion may be
required).

Implementing the active stabilization would probably be more fun
than the actual application, and could very well be more expensive
:slight_smile:

The other option, is you could simply use flash instead of rotating
media, or use a passive dampening system such as rubber mounts.

3

I would probably also look at laptop hard drives. They’re a lot more
robust.

cheers,

Kris

“Rennie Allen” <rallen@csical.com> wrote in message
news:bnudfc$bki$1@inn.qnx.com…

Bill Caroselli wrote:
NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?

You could use an accelerometer.

  1. What can we do to reduce or eliminate vibrations inside a computer
    rack?

Well, assuming you have 3 accelerometers (one for each of X,Y,Z
axis), and leveraged piezo elements isolating the frame of the
computer from its mount, you could implement an active stabilization
system (basically measure the vectors from the accellerometers, and
issue equal but opposite movements using the piezo elements
(depending on the amplitude of the vibrations peizos may not have
sufficient travel, so another means of imparting motion may be
required).

Implementing the active stabilization would probably be more fun
than the actual application, and could very well be more expensive
:slight_smile:

The other option, is you could simply use flash instead of rotating
media, or use a passive dampening system such as rubber mounts.

4

Rennie Allen <rallen@csical.com> wrote:
RA > Bill Caroselli wrote:

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?

RA > You could use an accelerometer.

  1. What can we do to reduce or eliminate vibrations inside a computer
    rack?

RA > Well, assuming you have 3 accelerometers (one for each of X,Y,Z
RA > axis), and leveraged piezo elements isolating the frame of the
RA > computer from its mount, you could implement an active stabilization
RA > system (basically measure the vectors from the accellerometers, and
RA > issue equal but opposite movements using the piezo elements
RA > (depending on the amplitude of the vibrations peizos may not have
RA > sufficient travel, so another means of imparting motion may be
RA > required).

RA > Implementing the active stabilization would probably be more fun
RA > than the actual application, and could very well be more expensive
RA > :slight_smile:

Curiously enough, that’s exactly what these trucks have. They use
accellerometers to determine how straight, level, etc. the track is.

We also do active control by applying a constant pressure to the rails
to determine hwo much the rails move. I.E. if the raisl move apart
more than a certain amount, then the rail support is weak. Since the
rails may also move slightly apart or closer together (gage) we need
to adjust the pressure against the rails 300 times a second.

HOWEVER, . . . .

The accellerometers on the truck body are telling us that this truck is
vibrating no more than the other trucks.

RA > The other option, is you could simply use flash instead of rotating
RA > media, or use a passive dampening system such as rubber mounts.

5

Ask the NASA, they have build an active compensation system
w/ 3D sensors into the STS to control the engines and
reducing vibration :wink:.
BTW: HiFi enthusiast use this sensors to actively drive speaker
membranes in a closed loop.
What are approx. the range of amplitude of the vibration ?
Do we talk about millimeters and frequencies below 100Hz ?
cheers, peter

Bill Caroselli wrote:

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?
  2. What can we do to reduce or eliminate vibrations inside a computer
    rack?
6

Implementing the active stabilization would probably be more fun
than the actual application, and could very well be more expensive
:slight_smile:

The other option, is you could simply use flash instead of rotating
media, or use a passive dampening system such as rubber mounts.

I have seen HD design for vibrating environment, but they were very
expensive (army stuff)

7

Bill Caroselli wrote:

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

A flash IDE drive doesn’t have bearings :slight_smile:

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?

Seimograph :slight_smile:

  1. What can we do to reduce or eliminate vibrations inside a computer
    rack?

mount the rack on top of rubber cushion and increase its mass.

Armin

8

“Bill Caroselli” <qtps@earthlink.net> wrote in message
news:bnuamc$9el$1@inn.qnx.com…

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?
  2. What can we do to reduce or eliminate vibrations inside a computer
    rack?

Bill
It’s interesting that only one computer, on one vehicle, has the
problem. Maybe the computer mounting in that vehicle transmits vibration
that is at the rotational speed of the drive (or a harmonic).
You could alter the mounting to try and change it’s natural frequency.
If the drive and/or computer are on flexible mounts try securely fastening a
weight to the suspended device. My guess is that you should use a weight
that is 15 to 25% of the device weight.
Small air bags make really good isolation mounts, and you can change the
air pressure to alter the characteristics of the assembly.
If you want I can dig up some info on air bags.

Pat

9

Check out the www.sandisk.com for flash-ide compatible drives, if you don’t
need more than 2GB.

“Bill Caroselli” <qtps@earthlink.net> wrote in message
news:bnuamc$9el$1@inn.qnx.com…

NOT!

We have vehicles that test railroad track. They run down the track
testing various attrbutes for safty, etc. There are four on board
computers. This is know to be a harsh environment for computers, both
because of tempratures and virbations.

But in one computer on one of the vehicles (out of a fleet of siz such
vehicles) it seems to go through a hard drive about every four months.
We sent one of these hard drives back to the manufacturer. They said
that the bearings were worn out from excessive vibration.

So, two questinos:

  1. What can we use to measure vibration inside a computer rack?
  2. What can we do to reduce or eliminate vibrations inside a computer
    rack?