Prev: Voltage Reflection Coefficient Question
Next: 36v power tool motor to replace scooter motor?
From: panfilero on 6 Aug 2010 14:57
Hello,

I have a question about where do frequencies end and pulses start?

I'm trying to make an impact detection circuit, I have some sensors on
a mat, and when something strikes the mat I get a small voltage (like
100mV).... so I'd like to amplify this 100mV to something around
1V....

I've been reading up online on how to do this, and the app notes and
other sources I've found say that it's important to take the frequency
of my system into account and make sure my amp can handle that
frequency range...

I don't know what the frequency of my system could be... if I get one
impact a minute... is that even a frequency? or just detecting a
pulse... do I even need to worry about the frequency range if I'm just
detecting random events? maybe I start seeing 1 impact per second...
now I'm at 1Hz.... does this require a special kind of op-amp?

Does a single voltage spike have a frequency associated with it that I
need to take into consideration when creating my amplifier?

as you can see I'm pretty confused about this low frequency/impulse
stuff.... any help is greatly appreciated,
thanks!
From: Jan Panteltje on 6 Aug 2010 15:07
On a sunny day (Fri, 6 Aug 2010 11:57:55 -0700 (PDT)) it happened panfilero
<panfilero(a)gmail.com> wrote in
<8eed662c-5e3a-45d4-89f0-83cc44dd1101(a)s9g2000yqd.googlegroups.com>:

>Hello,
>
>I have a question about where do frequencies end and pulses start?
>
>I'm trying to make an impact detection circuit, I have some sensors on
>a mat, and when something strikes the mat I get a small voltage (like
>100mV).... so I'd like to amplify this 100mV to something around
>1V....
>
>I've been reading up online on how to do this, and the app notes and
>other sources I've found say that it's important to take the frequency
>of my system into account and make sure my amp can handle that
>frequency range...
>
>I don't know what the frequency of my system could be... if I get one
>impact a minute... is that even a frequency? or just detecting a
>pulse... do I even need to worry about the frequency range if I'm just
>detecting random events? maybe I start seeing 1 impact per second...
>now I'm at 1Hz.... does this require a special kind of op-amp?
>
>Does a single voltage spike have a frequency associated with it that I
>need to take into consideration when creating my amplifier?
>
>as you can see I'm pretty confused about this low frequency/impulse
>stuff.... any help is greatly appreciated,
>thanks!

Any pulse has a rise time and fall time, or whatever waveform.
If you mathematically look at it, look up Fourrier transform,
it contains a lot of frequency components (harmonics).
In your case the rise time and fall time will not be important,
as you have plenty of time between pulses to react.
So anything with a bandwidth above a kHz should work perfectly,
maybe soften the edges of the input pulse, but that is good,
takes out RF interference.

From: Tim Wescott on 6 Aug 2010 15:31
On 08/06/2010 11:57 AM, panfilero wrote:
> Hello,
>
> I have a question about where do frequencies end and pulses start?
>
> I'm trying to make an impact detection circuit, I have some sensors on
> a mat, and when something strikes the mat I get a small voltage (like
> 100mV).... so I'd like to amplify this 100mV to something around
> 1V....
>
> I've been reading up online on how to do this, and the app notes and
> other sources I've found say that it's important to take the frequency
> of my system into account and make sure my amp can handle that
> frequency range...
>
> I don't know what the frequency of my system could be... if I get one
> impact a minute... is that even a frequency? or just detecting a
> pulse... do I even need to worry about the frequency range if I'm just
> detecting random events? maybe I start seeing 1 impact per second...
> now I'm at 1Hz.... does this require a special kind of op-amp?
>
> Does a single voltage spike have a frequency associated with it that I
> need to take into consideration when creating my amplifier?
>
> as you can see I'm pretty confused about this low frequency/impulse
> stuff.... any help is greatly appreciated,
> thanks!

If your amplifier has a frequency response that goes down to DC, and
goes up high enough to resolve the pulse, then that's all you need. If
you really wanted to figure this out from the standpoint of what the
frequency range of the amplifier does to the pulse, you'd figure out the
frequency content of a single pulse (not a train), and you'd run that by
the frequency response of your amplifier.

As a rough rule of thumb, an amplifier whose range goes from DC to about
10 / (pulse width) will be way more than enough. You could probably get
by with 2 / (pulse width) or even 1 / (pulse width), but then you'd have
to actually worry about the details -- if you can avoid that worry,
that's always nice.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html
From: Tim Williams on 6 Aug 2010 15:36
"Jan Panteltje" <pNaonStpealmtje(a)yahoo.com> wrote in message
news:i3hmi5$hhb$1(a)news.albasani.net...
> If you mathematically look at it, look up Fourrier transform,

Four rear transform?

Rule 34, it lives

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms


From: panfilero on 6 Aug 2010 15:48
Thanks, it's making more sense to me now, appreciate it
 |  Next  |  Last
Pages: 1 2
Prev: Voltage Reflection Coefficient Question
Next: 36v power tool motor to replace scooter motor?