Mounting Hard Drives on fast Accelerating Vehicles

[trezaei]

Hi all,

I have recently bought a DVR (Called the ARCHOS AV500) this device
contains a 100GB hard drive.
My intention was to mount this on a motorcycle which can accelerate
from 0-60 in just under 3 seconds.

My question is if the device is recording video at a bit-rate of around
2500k/s the the acceleration of the bike (I am calculating around 1G)
would have a major effect on the drive.

How should I mount the drive if I must? and please don't say don't
because I am just looking for the best way and I know this isn't the
recommended use of such devices.

Option 1 - Mount the drive horizontally with the spin axis pointing to
the sky?
Option 2 - Mount the drive horizontally with the spin axis pointing to
the ground?
Option 3 - Mount the drive vertically with the spin axis facing the
front of the bikes?
Option 4 - Mount the drive vertically with the spin axis facing the
back of the bikes?

When I say spin axis I am using the right-hand-rule of physics.

My inclination is to mount such that the spin axis points to the front
of the bike. but I am only saying that because the forces would at
least somewhat cancel each other.

Thanks

 

[Arno Wagner]

Hi all,

I have recently bought a DVR (Called the ARCHOS AV500) this device
contains a 100GB hard drive.
My intention was to mount this on a motorcycle which can accelerate
from 0-60 in just under 3 seconds.

My question is if the device is recording video at a bit-rate of around
2500k/s the the acceleration of the bike (I am calculating around 1G)
would have a major effect on the drive.

It might temporarily prevent the drive from doing anything.
A notebook HDD might be more tolerant. However I doubbt it will be
a problem.

How should I mount the drive if I must? and please don't say don't
because I am just looking for the best way and I know this isn't the
recommended use of such devices.

Option 1 - Mount the drive horizontally with the spin axis pointing to
the sky?
Option 2 - Mount the drive horizontally with the spin axis pointing to
the ground?
Option 3 - Mount the drive vertically with the spin axis facing the
front of the bikes?
Option 4 - Mount the drive vertically with the spin axis facing the
back of the bikes?

When I say spin axis I am using the right-hand-rule of physics.

It should not matter. Actually best might be to mount the disk
so that the platters and the wheels are in a plane. That way
you do not get precession.

My inclination is to mount such that the spin axis points to the front
of the bike. but I am only saying that because the forces would at
least somewhat cancel each other.

If there wera an effect it would be to pus some heads onto the platter
and others away from it. However I doubt it will make much difference.

Arno

 

[Rod Speed]

Hi all,

I have recently bought a DVR (Called the ARCHOS AV500) this device
contains a 100GB hard drive.
My intention was to mount this on a motorcycle which can accelerate
from 0-60 in just under 3 seconds.

My question is if the device is recording video at a bit-rate of
around 2500k/s the the acceleration of the bike (I am calculating
around 1G) would have a major effect on the drive.

How should I mount the drive if I must? and please don't say don't
because I am just looking for the best way and I know this isn't the
recommended use of such devices.

Option 1 - Mount the drive horizontally with the spin axis pointing to
the sky?
Option 2 - Mount the drive horizontally with the spin axis pointing to
the ground?
Option 3 - Mount the drive vertically with the spin axis facing the
front of the bikes?
Option 4 - Mount the drive vertically with the spin axis facing the
back of the bikes?

When I say spin axis I am using the right-hand-rule of physics.

My inclination is to mount such that the spin axis points to the front
of the bike. but I am only saying that because the forces would at
least somewhat cancel each other.

Yeah, that's the way I'd do it, have the axis in the direction
of the maximum Gs its likely to experience so you dont move
that axis in that particular situation.

Iffy tho, you can certainly make a case that it should be vertical
because it will see a lot less movements of that axis that way.

 

[trezaei]

Okay it seems I need to say why I even posted this message and why this
is an issue.

When the bike is sitting still, he archos records hours upon hours of
video flawlessly. But when its in motion the story is different. Any
file longer than 5 minutes eventually becomes corrupt in recording and
shows as "Corrupt or incomplete" I have not been able to determine why
this might be happening.

Although this might be the issue, it could also be the bullet cam which
is providing the video analog feed. Before I had it mounted on the oil
radiator I took it for a 10 minute ride filled with fast acceleration
and abrupt stops. No problems. The after mounting it, it seems that I
get corrupt files all the time. I have a hard time believing it the
Bullet cam but I guess it might well be. Maybe the heat from the
radiator is messing it up and it somehow corrupts the file. But that
makes almost no sense at all.

 

[Rod Speed]

Okay it seems I need to say why I even posted this message and why
this is an issue.

When the bike is sitting still, he archos records hours upon hours of
video flawlessly. But when its in motion the story is different. Any
file longer than 5 minutes eventually becomes corrupt in recording and
shows as "Corrupt or incomplete" I have not been able to determine why
this might be happening.

Although this might be the issue, it could also be the bullet cam
which is providing the video analog feed. Before I had it mounted on
the oil radiator I took it for a 10 minute ride filled with fast
acceleration and abrupt stops. No problems. The after mounting it, it
seems that I get corrupt files all the time. I have a hard time
believing it the Bullet cam but I guess it might well be. Maybe the
heat from the radiator is messing it up and it somehow corrupts the
file. But that makes almost no sense at all.

It shouldnt be hard to work out whether the video files are
getting corrupted because of the hard drive or the camera etc.
If its the hard drive, that should show up in the error logs.

 

[Folkert Rienstra]

(I am calculating around 1G)

A pityful 1G. Wow. What if you hit a pothole.

It might temporarily prevent the drive from doing anything.

So what, as long as it records it eventually (from it's buffers).

A notebook HDD might be more tolerant.

Which is what it is using.

However I doubbt it will be a problem.

Your 'doubbts' are your problem.

It should not matter.
Actually best might be to mount the disk so that the platters and
the wheels are in a plane. That way you do not get precession.

So much for your "It should not matter".

You should mount such that any shock is diminished to acceptable levels
for the drive, or rather for the Archos as a whole.

What forces.

 

[Timothy Daniels]

I have recently bought a DVR (Called the ARCHOS AV500) this device
contains a 100GB hard drive.
My intention was to mount this on a motorcycle which can accelerate
from 0-60 in just under 3 seconds.

My question is if the device is recording video at a bit-rate of around
2500k/s the the acceleration of the bike (I am calculating around 1G)
would have a major effect on the drive.

How should I mount the drive if I must? and please don't say don't
because I am just looking for the best way and I know this isn't the
recommended use of such devices.

Option 1 - Mount the drive horizontally with the spin axis pointing to
the sky?
Option 2 - Mount the drive horizontally with the spin axis pointing to
the ground?
Option 3 - Mount the drive vertically with the spin axis facing the
front of the bikes?
Option 4 - Mount the drive vertically with the spin axis facing the
back of the bikes?

When I say spin axis I am using the right-hand-rule of physics.

My inclination is to mount such that the spin axis points to the front
of the bike. but I am only saying that because the forces would at
least somewhat cancel each other.

Thanks

How can you determine direction of spin axis without knowing the
direction of rotation of the hard disk? Assuming that the bike will
be cornered, I'd align the spin axis of the disk longitudinally with the
bike so the lean of cornering wouldn't induce a torque on the spindle
of the disk.

The normal road vibration of a road-going motorcycle would be
what I'd worry about, though. The head flies about 10 *nanometers*
above the disk surface, and the shocks from road irregularities at
any decent speed would certainly exceed 1G and put a load on the
air film that separates the read/write head from the disk. Why not
look into telemetry of the video data?

*TimDaniels*

 

[J. Clarke]

Hi all,

I have recently bought a DVR (Called the ARCHOS AV500) this device
contains a 100GB hard drive.
My intention was to mount this on a motorcycle which can accelerate
from 0-60 in just under 3 seconds.

My question is if the device is recording video at a bit-rate of around
2500k/s the the acceleration of the bike (I am calculating around 1G)
would have a major effect on the drive.

How should I mount the drive if I must? and please don't say don't
because I am just looking for the best way and I know this isn't the
recommended use of such devices.

Option 1 - Mount the drive horizontally with the spin axis pointing to
the sky?
Option 2 - Mount the drive horizontally with the spin axis pointing to
the ground?
Option 3 - Mount the drive vertically with the spin axis facing the
front of the bikes?
Option 4 - Mount the drive vertically with the spin axis facing the
back of the bikes?

When I say spin axis I am using the right-hand-rule of physics.

My inclination is to mount such that the spin axis points to the front
of the bike. but I am only saying that because the forces would at
least somewhat cancel each other.

If you read the docs for the drive you'll find that 1g is not an issue for
it in any axis. However if it is hard-mounted it will likely be getting
vibration at much higher levels. Try helmet-mounting the camera and
storing the Archos in a pocket rather than attaching both to the bike and
see if there is an improvement. If so, then hard-mount one and see what
happens, if there's a problem you've found the culprit.

On the other hand, many modern disks have fall-sensing circuitry of one sort
or another that is intended to shut down the drive in the event that the
machine containing it is dropped--this works fast enough to have the drive
in a safe state between the time that a laptop slips off the edge of a
table and the time that it hits the floor. It could be that the drive in
your Archos is designed in this fashion and that some combination of
factors is generating a false trigger of that protective system.

Another possibility is that you're hitting a vibrational resonance frequency
of the drive at some point. Try running at constant RPM for a bit if you
can find traffic conditions that allow it and see if you get corruption, if
not, try a higher RPM until you've gone through the entire range.

 

[Steve Cousins]

How can you determine direction of spin axis without knowing the
direction of rotation of the hard disk?

Physics 101. Since he knows the right-hand rule, he should be able to
figure this out.

As others have pointed out, it is probably not the 0-60 acceleration
that is causing the problems. The key is probably to dampen both the
camera and the Archos to take care of the bumps.

Good luck. Sounds like a fun project.

Steve

 

[Timothy Daniels]

Physics 101. Since he knows the right-hand rule, he should be able to
figure this out.

Physics 101 - You have to know the direction of rotation to *apply*
the Right-Hand Rule. But knowing the direction of a spin vector -
which is what a hand rule tells - is irrelevant. What's relevant is
whether the HD's spin *axis* is parallel to the bike's longitudinal axis.
The fact that the OP even mentions a hand rule shows that he doesn't
appreciate its irrelevance.

*TimDaniels*

 

[Steve Cousins]

Physics 101 - You have to know the direction of rotation to *apply*
the Right-Hand Rule.

What happens if you:
Turn the drive on.
Pivot it 90 degrees.

Try pivoting it by different axes.

But knowing the direction of a spin vector -
which is what a hand rule tells - is irrelevant. What's relevant is
whether the HD's spin *axis* is parallel to the bike's longitudinal
axis.

The fact that the OP even mentions a hand rule shows that he doesn't
appreciate its irrelevance.

He was using it as a point of reference which was fine. He may not be
taking into consideration what happens when he turns though. Not only
is he leaning,(as you point out) and so is the hard drive unless it is
in a gyro (not the sandwich!), but conservation of angular momentum will
mean that the drive will try to swing. If the drive is fixed (mounted
to the bike), that momentum has to go somewhere and it will probably put
a significant force on the platters. Maybe enough for them to make
contact with the heads due to a flexing platter. I doubt this is the
cause of the problem though. I'm guessing the platters and spindle are
rigid enough to withstand this. I think it more likely the vibration and
jarring that you and others mentioned.

 

[Timothy Daniels]

What happens if you:
Turn the drive on. Pivot it 90 degrees.

Assuming
1) that the drive is up to speed so that angular momentum effects
are maximized, and
2) if one pivots the drive *axis* 90º, a torque will be exerted during
the pivoting that is a vector pointing 90º to the direction of pivoting
and proportional to the angular speed of pivoting. In the case of
the motorcycle, the direction of the torque is irrelevant since the
lean down into the turn will be approximately the same as the
straightening back to vertical while exiting the turn.

He may not be taking into consideration what happens when
he turns though. Not only is he leaning,(as you point out) and
so is the hard drive unless it is in a gyro (not the sandwich!),
but conservation of angular momentum will mean that the drive
will try to swing. If the drive is fixed (mounted to the bike), that
momentum has to go somewhere and it will probably put a significant force on the platters. Maybe enough for them to
make contact with the heads due to a flexing platter.

That's why I suggested that the axis of the spinning disk be
made parallel to the bike's longitudinal axis (i.e. along the
length of the bike) which turns only as fast as the bike follows
the turn in the road. That torque on the axis is thus less than
the torque that would be exerted on the axis if the axis were
lateral and pivoted by the faster lean into and out of the turn

I doubt this is the cause of the problem though.

Cause of what problem?

I think it more likely the vibration and jarring that you and others
mentioned.

Yup. The pros use telemetry for a reason.

*TimDaniels*

 

[CWatters]

Hi all,

I have recently bought a DVR (Called the ARCHOS AV500) this device
contains a 100GB hard drive.
My intention was to mount this on a motorcycle which can accelerate
from 0-60 in just under 3 seconds.

I suspect the g due to vibration is worse than the g due to acceleration.

 

[Quaoar]

Okay it seems I need to say why I even posted this message and why this
is an issue.

When the bike is sitting still, he archos records hours upon hours of
video flawlessly. But when its in motion the story is different. Any
file longer than 5 minutes eventually becomes corrupt in recording and
shows as "Corrupt or incomplete" I have not been able to determine why
this might be happening.

Although this might be the issue, it could also be the bullet cam which
is providing the video analog feed. Before I had it mounted on the oil
radiator I took it for a 10 minute ride filled with fast acceleration
and abrupt stops. No problems. The after mounting it, it seems that I
get corrupt files all the time. I have a hard time believing it the
Bullet cam but I guess it might well be. Maybe the heat from the
radiator is messing it up and it somehow corrupts the file. But that
makes almost no sense at all.

Ignition system electrical noise and/or stray frame currents?
Acceleration and braking are likely, IMO, to have little or no effect
since the acceleration in g's is very low. G-shock from bumps is a worry
for any device not shock-hardened or isolated (gel casing, etc.) since
g-shock has accelerations much greater than simple bike acceleration and
braking.

Q

 

[David Lesher]

http://en.wikipedia.org/wiki/Solid_state_disk