low frequency amplification

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!

Reply to
panfilero
Loading thread data ...

On a sunny day (Fri, 6 Aug 2010 11:57:55 -0700 (PDT)) it happened panfilero wrote in :

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.

Reply to
Jan Panteltje

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
 Click to see the full signature
Reply to
Tim Wescott

Four rear transform?

Rule 34, it lives

Tim

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

Thanks, it's making more sense to me now, appreciate it

Reply to
panfilero

No, its 'furrier' transform. Because the math is pretty hairy.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
 Click to see the full signature
Reply to
Paul Hovnanian P.E.

Or 'farrier' transform, because it'll kick you in the head if you aren't careful?

--
Tim Wescott
Wescott Design Services
 Click to see the full signature
Reply to
Tim Wescott

On a sunny day (Fri, 06 Aug 2010 14:04:58 -0700) it happened "Paul Hovnanian P.E." wrote in :

Fourier is actually simple. I even did it in ASM on a PIC with sine lookup table:

formatting link

Reply to
Jan Panteltje

Has anyone done a squarewave transform? It would be almost trivial to compute, no multiplication (aside from signs) and no sine tables. Just convolve with squares instead.

I was going to do it but hadn't gotten around to proving the Fourier equivalence. Thing is, the fundamental has odd harmonics, which contribute to those terms in the FT, whereas the highest order sample has none (they alias out), so the terms are not quite one-to-one and the transformation isn't quite as simple. Math has a way of biting you in the ass, sooner or later; the conversion may require sines anyway. (This is referred to as the Conservation of Homework law.)

Tim

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

You want to look at the frequency _response_ of your system. The higher the frequency response, the faster it'll be able to respond to the edge of a pulse or a contact like yours.

As far as frequency itself, that would be more like how often are people stepping on the pad, which I kinda intuit isn't what you're looking for here. :-)

Have Fun! Rich

Reply to
Rich Grise

Google for "Hadamard transform" and/or "Walsh functions".

Best regards,

Bob Masta DAQARTA v5.10 Data AcQuisition And Real-Time Analysis

formatting link
Scope, Spectrum, Spectrogram, Sound Level Meter Frequency Counter, FREE Signal Generator Pitch Track, Pitch-to-MIDI DaqMusic - FREE MUSIC, Forever! (Some assembly required) Science (and fun!) with your sound card!

Reply to
Bob Masta

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.