memsic 2125 accelerometer from parallax
Archiver
Posts: 46,084
First I just want to thank parallax for their memsic accelerometer unit.
I'm very impressed with it so far.
Now on to the question.
Using the code straight from parallax it seems the longer the wires are to
the accelerometer the less G's are registered at 90 degrees off level on
either axis. The three feet of wire I have leading to the accelerometer
unit cause it to register .960. When I reduce this to 6 inches it reads
1.000. Is this due to the wire itself, the length of it, or do I need some
kind of buffer/driver closer to the accelerometer for a 3 foot cable run?
BTW I'm using 1 pair from a length of CAT5 for power and a second pair for
X/Y signals. Maybe I should try a separate pair for X and Y with one lead
tied to ground and the signal on the other?
-Rob
I'm very impressed with it so far.
Now on to the question.
Using the code straight from parallax it seems the longer the wires are to
the accelerometer the less G's are registered at 90 degrees off level on
either axis. The three feet of wire I have leading to the accelerometer
unit cause it to register .960. When I reduce this to 6 inches it reads
1.000. Is this due to the wire itself, the length of it, or do I need some
kind of buffer/driver closer to the accelerometer for a 3 foot cable run?
BTW I'm using 1 pair from a length of CAT5 for power and a second pair for
X/Y signals. Maybe I should try a separate pair for X and Y with one lead
tied to ground and the signal on the other?
-Rob
Comments
for data, one for data ground, and tie the shield to ground on one end only.
Does it work OK if you only hook up / monitor one axis?
Got me a couple of them things sitting on the bench also -- waiting to
figure out what to build. Are they sensitive enough to make an electronic
level?
Original Message
> First I just want to thank parallax for their memsic accelerometer unit.
> I'm very impressed with it so far.
>
> Now on to the question.
>
> Using the code straight from parallax it seems the longer the wires are to
> the accelerometer the less G's are registered at 90 degrees off level on
> either axis. The three feet of wire I have leading to the accelerometer
> unit cause it to register .960. When I reduce this to 6 inches it reads
> 1.000. Is this due to the wire itself, the length of it, or do I need
some
> kind of buffer/driver closer to the accelerometer for a 3 foot cable run?
> BTW I'm using 1 pair from a length of CAT5 for power and a second pair for
> X/Y signals. Maybe I should try a separate pair for X and Y with one lead
> tied to ground and the signal on the other?
about it (it will be installed in my auto-cross track car).
They are fairly sensative. The parallax demo code produces a result that is
"chunky", so level is .000 it moves to .008, then .016. The demo code also
shows the degrees of tilt, but I can't quite remember the resolution because
I commented out that section of code shortly after hooking it up for a test.
I just had to see how fast I could get it to update on a LCD so I commented
out everything I could. I do think it could be used as an electronic level,
but it wouldn't be much more accurate than eyeing up the bubble on a normal
level.
I was thinking about putting all my spare parts together and maybe fit
together a robot and use the memsic unit to keep the robot from tipping
itself while attempting an incline. Even make a device that logged
time/date of peaks in G's, then use that to keep an eye on movers, UPS, etc
[noparse];)[/noparse] So many ideas for such a small thing.
Rob
Original Message
From: "Rodent" <daweasel@s...>
To: <basicstamps@yahoogroups.com>
Sent: Monday, February 10, 2003 12:27 AM
Subject: Re: [noparse][[/noparse]basicstamps] memsic 2125 accelerometer from parallax
> Might try running two separate two-conductor shielded cables. One
conductor
> for data, one for data ground, and tie the shield to ground on one end
only.
>
> Does it work OK if you only hook up / monitor one axis?
>
> Got me a couple of them things sitting on the bench also -- waiting to
> figure out what to build. Are they sensitive enough to make an electronic
> level?
>
>
Original Message
>
> > First I just want to thank parallax for their memsic accelerometer unit.
> > I'm very impressed with it so far.
> >
> > Now on to the question.
> >
> > Using the code straight from parallax it seems the longer the wires are
to
> > the accelerometer the less G's are registered at 90 degrees off level on
> > either axis. The three feet of wire I have leading to the accelerometer
> > unit cause it to register .960. When I reduce this to 6 inches it reads
> > 1.000. Is this due to the wire itself, the length of it, or do I need
> some
> > kind of buffer/driver closer to the accelerometer for a 3 foot cable
run?
> > BTW I'm using 1 pair from a length of CAT5 for power and a second pair
for
> > X/Y signals. Maybe I should try a separate pair for X and Y with one
lead
> > tied to ground and the signal on the other?
>
>
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>
shielded cable solution worked. I may be running into this problem myself in
the future.
James
Original Message
From: Robert Staph <rstaph@a...>
Sent: 02/10/03 07:59 AM
To: basicstamps@yahoogroups.com
Subject: Re: [noparse][[/noparse]basicstamps] memsic 2125 accelerometer from parallax
>
> I'll try shielded cables, I should be using that anyway the more I think
about it (it will be installed in my auto-cross track car).
They are fairly sensative. The parallax demo code produces a result that
is
"chunky", so level is .000 it moves to .008, then .016. The demo code
also
shows the degrees of tilt, but I can't quite remember the resolution
because
I commented out that section of code shortly after hooking it up for a
test.
I just had to see how fast I could get it to update on a LCD so I
commented
out everything I could. I do think it could be used as an electronic
level,
but it wouldn't be much more accurate than eyeing up the bubble on a
normal
level.
I was thinking about putting all my spare parts together and maybe fit
together a robot and use the memsic unit to keep the robot from tipping
itself while attempting an incline. Even make a device that logged
time/date of peaks in G's, then use that to keep an eye on movers, UPS,
etc
[noparse];)[/noparse] So many ideas for such a small thing.
Rob
Original Message
From: "Rodent" <daweasel@s...>
To: <basicstamps@yahoogroups.com>
Sent: Monday, February 10, 2003 12:27 AM
Subject: Re: [noparse][[/noparse]basicstamps] memsic 2125 accelerometer from parallax
> Might try running two separate two-conductor shielded cables. One
conductor
> for data, one for data ground, and tie the shield to ground on one end
only.
>
> Does it work OK if you only hook up / monitor one axis?
>
> Got me a couple of them things sitting on the bench also -- waiting to
> figure out what to build. Are they sensitive enough to make an
electronic
> level?
>
>
Original Message
>
> > First I just want to thank parallax for their memsic accelerometer
unit.
> > I'm very impressed with it so far.
> >
> > Now on to the question.
> >
> > Using the code straight from parallax it seems the longer the wires
are
to
> > the accelerometer the less G's are registered at 90 degrees off level
on
> > either axis. The three feet of wire I have leading to the
accelerometer
> > unit cause it to register .960. When I reduce this to 6 inches it
reads
> > 1.000. Is this due to the wire itself, the length of it, or do I need
> some
> > kind of buffer/driver closer to the accelerometer for a 3 foot cable
run?
> > BTW I'm using 1 pair from a length of CAT5 for power and a second pair
for
> > X/Y signals. Maybe I should try a separate pair for X and Y with one
lead
> > tied to ground and the signal on the other?
>
>
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject
and
Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
>
>
>
To UNSUBSCRIBE, just send mail to:
basicstamps-unsubscribe@yahoogroups.com
from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
for a wide range of users. That said, the specs are available from Memsic
and through different methods you can get finer resolution in the results.
My December Nuts & Volts article uses a different method for angular
calculation and returns much finer movements than the demo code that ships
with the product.
There's always more than one way to skin a cat....
BTW, we have put the Memsic to use in our Toddler robot with quite a bit of
success. I will be putting it into our SumoBot as well.
-- Jon Williams
-- Parallax
In a message dated 2/10/2003 8:10:22 AM Central Standard Time,
rstaph@a... writes:
> I'll try shielded cables, I should be using that anyway the more I think
> about it (it will be installed in my auto-cross track car).
>
> They are fairly sensative. The parallax demo code produces a result that
> is
> "chunky", so level is .000 it moves to .008, then .016. The demo code also
> shows the degrees of tilt, but I can't quite remember the resolution
> because
> I commented out that section of code shortly after hooking it up for a
> test.
> I just had to see how fast I could get it to update on a LCD so I commented
> out everything I could. I do think it could be used as an electronic
> level,
> but it wouldn't be much more accurate than eyeing up the bubble on a normal
> level.
>
> I was thinking about putting all my spare parts together and maybe fit
> together a robot and use the memsic unit to keep the robot from tipping
> itself while attempting an incline. Even make a device that logged
> time/date of peaks in G's, then use that to keep an eye on movers, UPS, etc
> [noparse];)[/noparse] So many ideas for such a small thing.
>
> Rob
[noparse][[/noparse]Non-text portions of this message have been removed]
easier to see what's going on, which I'm sure was the intended effect.
Someone of a higher skill level like yourself, Jon, would most likely be
able to extract a higher degree of accuracy.
The memsic unit is well worth what parallax is charging, especially because
it seems to have an over all survivability edge on certain other
accelerometers.
Rob
Original Message
From: <jonwms@a...>
To: <basicstamps@yahoogroups.com>
Sent: Monday, February 10, 2003 11:23 AM
Subject: Re: [noparse][[/noparse]basicstamps] memsic 2125 accelerometer from parallax
> The Parallax code is "chunky" because we have to keep examples fairly
simple
> for a wide range of users. That said, the specs are available from Memsic
> and through different methods you can get finer resolution in the results.
> My December Nuts & Volts article uses a different method for angular
> calculation and returns much finer movements than the demo code that ships
> with the product.
>
> There's always more than one way to skin a cat....
>
> BTW, we have put the Memsic to use in our Toddler robot with quite a bit
of
> success. I will be putting it into our SumoBot as well.
>
> -- Jon Williams
> -- Parallax
>
> In a message dated 2/10/2003 8:10:22 AM Central Standard Time,
> rstaph@a... writes:
>
> > I'll try shielded cables, I should be using that anyway the more I think
> > about it (it will be installed in my auto-cross track car).
> >
> > They are fairly sensative. The parallax demo code produces a result
that
> > is
> > "chunky", so level is .000 it moves to .008, then .016. The demo code
also
> > shows the degrees of tilt, but I can't quite remember the resolution
> > because
> > I commented out that section of code shortly after hooking it up for a
> > test.
> > I just had to see how fast I could get it to update on a LCD so I
commented
> > out everything I could. I do think it could be used as an electronic
> > level,
> > but it wouldn't be much more accurate than eyeing up the bubble on a
normal
> > level.
> >
> > I was thinking about putting all my spare parts together and maybe fit
> > together a robot and use the memsic unit to keep the robot from tipping
> > itself while attempting an incline. Even make a device that logged
> > time/date of peaks in G's, then use that to keep an eye on movers, UPS,
etc
> > [noparse];)[/noparse] So many ideas for such a small thing.
> >
> > Rob
>
>
>
> [noparse][[/noparse]Non-text portions of this message have been removed]
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>
We try to sell things at a fair price; enough to pay our employees, provide
support materials for our customers, and develop new products.
The Memsic 2125 is a neat device and due to its unique design has a much
better shock tolerance than similar products.
-- Jon Williams
-- Parallax
In a message dated 2/10/2003 12:44:20 PM Central Standard Time,
rstaph@a... writes:
> The memsic unit is well worth what parallax is charging, especially because
> it seems to have an over all survivability edge on certain other
> accelerometers.
[noparse][[/noparse]Non-text portions of this message have been removed]