Searching \ for 'tilt sensor, or motor load indicator...' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/sensors.htm?key=sensor
Search entire site for: 'tilt sensor, or motor load indicator...'.

Truncated match.
PICList Thread
'tilt sensor, or motor load indicator...'
1999\04\15@194925 by William Chops Westfield

face picon face
I want to build a digital (ie yes or no) tilt sensor that will operate
correctly in free fall and other inertial reference frames (ie I can't
use gravity.)  This is nominally an application for a gyroscope, except
that most available gyros are complex analog-output things.  I figured
I'd use a small DC motor to spin up a "massive" disk, and sense the change
in load that the motor has to provide to build up angular momentum in the
new plane of rotation (for 90 degrees, that ought to be able equal to startup
loads, right?)

This seemed easy enough till I remembered that a DC motor doesn't have a
nice smooth load current that you can detect changes in, it has some sort
of complex waveform due to the brushes and magnets and coils and all
that stuff.  Any suggestions?  Target platform is a PIC16C71...

Thanks
Bill W

1999\04\15@201003 by Sean Breheny

face picon face
Hi Bill,

There may be a problem with your idea. The angular momentum of the whole
physically connected system is conserved,not just that of the disk. When
your free-falling object tilts,the motor/disk will exert a torque on the
object which contains them.

So, I'm saying that I don't think the load on your motor will change.
(Example: take a wheel and spin it. Then, change the direction of its axis.
YOU will feel a torque on your hands,but the wheel won't slow down).

You may be able to do what you want by having a spinning disk and measuring
how much torque it takes to hold it on some fixed axis relative to the
object. This would be some type of rate gyro,since the amount of torque
would represent the rate at which the spin axis is being changed. As you
probably know, rate gyros for model aircraft are available.

Sean


At 04:48 PM 4/15/99 PDT, you wrote:
{Quote hidden}

| Sean Breheny
| Amateur Radio Callsign: KA3YXM
| Electrical Engineering Student
\--------------=----------------
Save lives, please look at http://www.all.org
Personal page: http://www.people.cornell.edu/pages/shb7
spam_OUTshb7TakeThisOuTspamcornell.edu ICQ #: 3329174

1999\04\15@203800 by William Chops Westfield

face picon face
   There may be a problem with your idea. The angular momentum of the whole
   physically connected system is conserved,not just that of the disk. When
   your free-falling object tilts,the motor/disk will exert a torque on the
   object which contains them.

   So, I'm saying that I don't think the load on your motor will change.
   (Example: take a wheel and spin it. Then, change the direction of its axis.
   YOU will feel a torque on your hands,but the wheel won't slow down).

Good point.  I left out the flexible shaft coupling the motor to the disk,
and the housing.  I guess I didn't really expect the disk to change the load
by itself - rather, I expect it to (attempt to) maintain it's attitude while
the housing tilts, resulting in it hitting the housing, thus resulting in
increase motor load by friction between the disk and housing.

I'm attempting to achieve smaller, simpler, and MUCH cheaper that the model
aircraft gyros by settling for a binary output...

BillW

1999\04\15@223541 by Wagner Lipnharski

picon face
> (Example: take a wheel and spin it. Then, change the direction
> of its axis. YOU will feel a torque on your hands,but the wheel
> won't slow down).

I wonder where comes from the energy torque you feel in your hand when
tilting the spindle... I suppose it is exactly the angular inertia
trying to compensate the 3 vectors changing, so there are some energy
transfered.

The best way to read the current motor is just converting it to voltage,
filter it in someway to allow you to get a stable reading and feel that
tilt energy transfer.

But, I believe the actual missile guidance systems doesn't use anymore
the perfect gyros with "swiss precision" bearings and sensors... just
plain accelerometer chips, cheap, easy to read and very small.  

Take a look at Analog Devices ADXL50, it measures ±5 g range, outpus
analog signal from 0.25 to 4.75V with a power supply from +5 to +24V.  A
couple of them should solve your problem.

try: http://products.analog.com/products/info.asp?product=ADXL50

Wagner.

1999\04\15@233203 by Gerhard Fiedler

picon face
At 22:34 04/15/99 -0400, Wagner Lipnharski wrote:
>> (Example: take a wheel and spin it. Then, change the direction
>> of its axis. YOU will feel a torque on your hands,but the wheel
>> won't slow down).
>
>I wonder where comes from the energy torque you feel in your hand when
>tilting the spindle... I suppose it is exactly the angular inertia
>trying to compensate the 3 vectors changing, so there are some energy
>transfered.

not necessarily. force becomes only energy with movement; as long as you
don't move in the direction of the force, there is no energy transferred.

ge

1999\04\15@235104 by Sean Breheny

face picon face
Hi again Bill,

Ah, I see. Well, that's an interesting idea. Of course, it is still a rate
gyro,and since its only a binary rate gyro, you won't be able to get useful
tilt information out of it (you can't integrate the rotation rate,since you
only know if it is above or below a threshold, and you don't know its
sign). Then again, you may only want rate info.

I don't have enought experience in this area (or enough time to keep
thinking about it now, I have homework to finish!) to really say what would
work best, but I'm sure someone here will have a suggestion.

Sean


At 05:37 PM 4/15/99 PDT, you wrote:
{Quote hidden}

| Sean Breheny
| Amateur Radio Callsign: KA3YXM
| Electrical Engineering Student
\--------------=----------------
Save lives, please look at http://www.all.org
Personal page: http://www.people.cornell.edu/pages/shb7
.....shb7KILLspamspam@spam@cornell.edu ICQ #: 3329174

1999\04\16@002230 by Corey Drechsler

picon face
Bill,

How about mounting the motor/gyroscope so that it is free to move as the
device tilts - maybe make it so the the motor always faces up...  Then
you could attach switches to the motor mount that would tell you if the
motor had tilted relative to the actual device.

Just my two cents...

Corey Drechsler

1999\04\16@004419 by William Chops Westfield

face picon face
   Take a look at Analog Devices ADXL50, it measures 15 g range, outpus
   analog signal from 0.25 to 4.75V with a power supply from +5 to +24V.  A
   couple of them should solve your problem.

I understand that there are other solutions to the problem I am solving.  I
am specifically after an "under $10" complete circuit, which ought to be a
piece of cake with a $0.35 motor and a $3 PIC (plus hobbyist labor
"machining" the rotor and such.)  In particular getting a binary output from
a micromachined sensor doesn't change the cost of the sensor :-( (what I'm
trying to do might be described as a macromachined sensor with one bit of
resolution...)

BillW

1999\04\16@005319 by Sean Breheny

face picon face
Bill,

What about this, a variation on what was suggested before by someone else:

Have your motor spin a disk and suspend the motor from a low-friction
bearing so it is free to tilt. Now, use an object-interrupted optical pair
(IR led and phototrans) which get blocked by the motor when it is near the
zero deg. angle,but is unblocked when the motor is tilted by some amount.

Even this will drift some,because the zero position is arbitrary and if it
gets reset by various forces to something not alligned with your sensors,
it won't work right. However, all inertial reference systems have this
problem,in varying degrees, AFAIK.

Sean


At 09:44 PM 4/15/99 PDT, you wrote:
{Quote hidden}

| Sean Breheny
| Amateur Radio Callsign: KA3YXM
| Electrical Engineering Student
\--------------=----------------
Save lives, please look at http://www.all.org
Personal page: http://www.people.cornell.edu/pages/shb7
shb7spamKILLspamcornell.edu ICQ #: 3329174

1999\04\16@175643 by Eric Oliver

flavicon
face
If I follow this correctly, why would you even need a motor ?  Picture a
fork with only two tines.  Now a shaft horizontally between the two tines
on bearings. A small lever attached to the shaft hanging down vertically
between the two tines.  As you tilt the fork, the lever will remain
vertical.  Using the optical setup below, you could set it up so that the
lever blocked the beam when the fork was upright, but as it tilts, the
lever would move out of the way and allow the beam to trigger the receiver
on the other side.  It could be very small and it gives you what you need
digital ouput. The only kicker is the bearings. I don't know if you could
build it for the under $10 range.


Eric

{Original Message removed}

1999\04\16@181625 by William Chops Westfield

face picon face
   If I follow this correctly, why would you even need a motor ?  Picture a
   fork with only two tines.  Now a shaft horizontally between the two tines
   on bearings. A small lever attached to the shaft HANGING down vertically
   between the two tines.

"hanging" requires Gravity, or a mercury switch would be sufficient.
In essence, I'm looking for the simplest, cheapest equivilent of a mercury
switch that will detect an attitude change in free fall.

BillW

1999\04\16@211441 by Eric Oliver

flavicon
face
Ahhh. I think I missed the free fall part in the original post.

Eric

-----Original Message-----
From:   William Chops Westfield [SMTP:.....billwKILLspamspam.....CISCO.COM]
Sent:   Friday, April 16, 1999 5:15 PM
To:     EraseMEPICLISTspam_OUTspamTakeThisOuTMITVMA.MIT.EDU
Subject:        Re: tilt sensor, or motor load indicator...

   If I follow this correctly, why would you even need a motor ?  Picture a
   fork with only two tines.  Now a shaft horizontally between the two tines
   on bearings. A small lever attached to the shaft HANGING down vertically
   between the two tines.

"hanging" requires Gravity, or a mercury switch would be sufficient.
In essence, I'm looking for the simplest, cheapest equivilent of a mercury
switch that will detect an attitude change in free fall.

BillW

1999\04\16@213346 by Wagner Lipnharski

picon face
You could search for air bag sensors in junkcars depot. They should be
inexpensive after removed from damaged cars, since noboby install "old"
sensors in new cars, and after a crash, if the car would be possible to
be rebuild, they need to replace all the system, the air bags AND the
sensors.  They are accelerometers... probably for a couple of $.

William Chops Westfield wrote:
> I understand that there are other solutions to the problem I am solving.  I
> am specifically after an "under $10" complete circuit, which ought to be a
> piece of cake with a $0.35 motor and a $3 PIC (plus hobbyist labor
> "machining" the rotor and such.)  In particular getting a binary output from
> a micromachined sensor doesn't change the cost of the sensor :-( (what I'm
> trying to do might be described as a macromachined sensor with one bit of
> resolution...

1999\04\17@030239 by Craig R. Graf

flavicon
face
The problem with airbag sensors is the 50g rating.  Gravity is such a small
part of this.
Humphrey makes a terrific tilt/angle sensor that can be coupled with an SSI
ASIC to provide excellent thermal stability
http://www.remec.com/hmphry/hmphryf.htm  click on new products.

VTI Hamlin makes some great low g accelerometers.

I have also included a link to a very good lecture series on sensors
http://cdr.stanford.edu/DD/Courses/me220/list.html


{Original Message removed}

1999\04\17@104656 by miked

flavicon
face
> Bill,
>
> What about this, a variation on what was suggested before by someone else:
>
> Have your motor spin a disk and suspend the motor from a low-friction
> bearing so it is free to tilt. Now, use an object-interrupted optical pair
> (IR led and phototrans) which get blocked by the motor when it is near the
> zero deg. angle,but is unblocked when the motor is tilted by some amount.
>

This is how mechanical gyros work that are used in r/c helicopters.
Pretty much replaced by piezo units now.
 Mike DeMetz sect. Michiana R/C Choppers
 mikedspamspam_OUTtechnologist.com
 http://www.elkhart.net/~miked/

1999\04\27@182648 by John Payson

flavicon
face
|    If I follow this correctly, why would you even need a motor ?  Picture a
|    fork with only two tines.  Now a shaft horizontally between the two tines
|    on bearings. A small lever attached to the shaft HANGING down vertically
|    between the two tines.

|"hanging" requires Gravity, or a mercury switch would be sufficient.
|In essence, I'm looking for the simplest, cheapest equivilent of a mercury
|switch that will detect an attitude change in free fall.

What about this: place a balanced horizontal rod so that it can pivot
(like a see-saw).  Use a small, weak, spring to bias it toward staying
horizontal.  Place sensors above/below one end of the rod to detect if
it moves.  Because the rod is balanced, it will be unaffected by grav-
ity or linear acceleration.

Note, btw, that you may tilt the whole setup so the rod is vertical (or
any other angle you want).  Note also that if you wish to detect more
than one axis of rotation you'll either need to add another pivoting
rod or else design the pivot so the rod can move in more than one axis
(e.g. using a gimballed mount).

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...