AVR-Chat

I have a circuit with an Accelerometer connected to an ATmega32.  The
Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
The motors I am using on this bot cause vibrations when running and I
can't get a clean signal fromt he Accel.
Is there a way to filter the noise from the motors out of the Accel
signal through CodevisionAVR?
Thanks!
~Mike

I have kind of had this problem for a little while, I am assuming your 
accelerometer is on from Analog Devices and also outputs a PWM wave just 
like the analog signal. I haven't figured out a way to measure like 4 
PWM waves, out of phase (possibly) in my application. I am afraid that 
when I do the counting, an interrupt might occur and I will have obscure 
results. I cannot disable the interrupts temporarily because the 
interrupts are needed, all the time (one of them generates PWM signals 
for servos, the other counts encoder clicks, etc) I am implementing this 
on an ATmega16. Does anyone have a clue as to how I would go about it?


Is there a way to filter the noise from the motors out of the Accel
signal through CodevisionAVR?

I think you could average your results, do you need to have a high 
sample rate? It also might be the AVR's internal ADC, the motor migth be 
generating a lot of noise on the power supply


michaelsteinbach wrote:

> I have a circuit with an Accelerometer connected to an ATmega32.  The
> Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
> The motors I am using on this bot cause vibrations when running and I
> can't get a clean signal fromt he Accel.
> Is there a way to filter the noise from the motors out of the Accel
> signal through CodevisionAVR?
> Thanks!
> ~Mike
>
>
>
> ------------------------------------------------------------------------
> YAHOO! GROUPS LINKS
>
>     *  Visit your group "AVR-Chat
>       <http://groups.yahoo.com/group/AVR-Chat>" on the web.
>        
>     *  To unsubscribe from this group, send an email to:
>        AVR-Chat-unsubscribe@yahoogroups.com
>       <mailto:AVR-Chat-unsubscribe@yahoogroups.com?subject=Unsubscribe>
>        
>     *  Your use of Yahoo! Groups is subject to the Yahoo! Terms of
>       Service <http://docs.yahoo.com/info/terms/>.
>
>
> ------------------------------------------------------------------------
>

Gustavo Goretkin wrote:
> I have kind of had this problem for a little while, I am assuming your 
> accelerometer is on from Analog Devices and also outputs a PWM wave just 
> like the analog signal. I haven't figured out a way to measure like 4 
> PWM waves, out of phase (possibly) in my application. I am afraid that 
> when I do the counting, an interrupt might occur and I will have obscure 
> results. I cannot disable the interrupts temporarily because the 
> interrupts are needed, all the time (one of them generates PWM signals 
> for servos, the other counts encoder clicks, etc) I am implementing this 
> on an ATmega16. Does anyone have a clue as to how I would go about it?

You could make your other interrupts reentrant by the pwm interrupt.
Do the minimum stuff in interrupts to minimize jitter problems.
With multiple pulse sources, you really need external hardwired
logic or a cpld/fpga.

> Is there a way to filter the noise from the motors out of the Accel
> signal through CodevisionAVR?

Feed the pulses thru an RC lowpass filter with the lowest cutoff
sensible, then feed thru a comparator with hysteresis.

> I think you could average your results, do you need to have a high 
> sample rate? It also might be the AVR's internal ADC, the motor migth be 
> generating a lot of noise on the power supply
> 
> michaelsteinbach wrote:
> 
>>I have a circuit with an Accelerometer connected to an ATmega32.  The
>>Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
>>The motors I am using on this bot cause vibrations when running and I
>>can't get a clean signal fromt he Accel.

Mount the sensor assembly on a lead weight, then fasten that
with a soft rubber block with adequate damping coefficient.

>>Is there a way to filter the noise from the motors out of the Accel
>>signal through CodevisionAVR?

You can do box or exponential averaging.

On Sun, 11 Sep 2005 05:13:09 -0000 you wrote:

>I have a circuit with an Accelerometer connected to an ATmega32.  The
>Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
>The motors I am using on this bot cause vibrations when running and I
>can't get a clean signal fromt he Accel.
>Is there a way to filter the noise from the motors out of the Accel
>signal through CodevisionAVR?
>Thanks!
>~Mike

A moving average is quite easy to implement. Just put the values into an array and
maintain a sum, adding the latest value and subtracting the oldest. I got better results
using a hardware filter - a dual op amp connected as a low pass filter.
--
Ralph Hilton
http://www.ralphhilton.org
C-Meter: http://www.cmeter.org
FZAOINT http://www.fzaoint.net

Gustavo Goretkin wrote:
> I have kind of had this problem for a little while, I am assuming your 
> accelerometer is on from Analog Devices and also outputs a PWM wave just 
> like the analog signal. I haven't figured out a way to measure like 4 
> PWM waves, out of phase (possibly) in my application. I am afraid that 
> when I do the counting, an interrupt might occur and I will have obscure 
> results. I cannot disable the interrupts temporarily because the 
> interrupts are needed, all the time (one of them generates PWM signals 
> for servos, the other counts encoder clicks, etc) I am implementing this 
> on an ATmega16. Does anyone have a clue as to how I would go about it?

You could make your other interrupts reentrant by the pwm interrupt.
Do the minimum stuff in interrupts to minimize jitter problems.
With multiple pulse sources, you really need external hardwired
logic or a cpld/fpga.

> Is there a way to filter the noise from the motors out of the Accel
> signal through CodevisionAVR?

To filter noise glitches from the square-wave edges, feed the pulses
thru an RC lowpass filter with the lowest cutoff sensible, then feed
thru a comparator with hysteresis.

> I think you could average your results, do you need to have a high 
> sample rate? It also might be the AVR's internal ADC, the motor migth be 
> generating a lot of noise on the power supply
> 
> michaelsteinbach wrote:
> 
>>I have a circuit with an Accelerometer connected to an ATmega32.  The
>>Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
>>The motors I am using on this bot cause vibrations when running and I
>>can't get a clean signal fromt he Accel.

Mount the sensor assembly on a lead weight, then fasten that
with a soft rubber block with adequate damping coefficient.

>>Is there a way to filter the noise from the motors out of the Accel
>>signal through CodevisionAVR?

You can do box or exponential averaging.

http://www.geocities.com/leon_heller

----- Original Message ----- 
From: "michaelsteinbach" <michaelsteinbach@yahoo.com>
To: <AVR-Chat@yahoogroups.com>
Sent: Sunday, September 11, 2005 6:13 AM
Subject: [AVR-Chat] Sensor filtering


>I have a circuit with an Accelerometer connected to an ATmega32.  The
> Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
> The motors I am using on this bot cause vibrations when running and I
> can't get a clean signal fromt he Accel.
> Is there a way to filter the noise from the motors out of the Accel
> signal through CodevisionAVR?

DSP might be applicable, if you have enough spare processing capability.

Leon


-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.344 / Virus Database: 267.10.21/96 - Release Date: 10/09/2005

---
[This E-mail has been scanned for viruses but it is your responsibility 
to maintain up to date anti virus software on the device that you are
currently using to read this email. ]

On Sep 11, 2005, at 8:49 AM, Leon Heller wrote:

>> I have a circuit with an Accelerometer connected to an ATmega32.  The
>> Accel measures 2.5V at rest and goes to .5v or up to 4.5v when  
>> tilted.
>> The motors I am using on this bot cause vibrations when running and I
>> can't get a clean signal fromt he Accel.
>> Is there a way to filter the noise from the motors out of the Accel
>> signal through CodevisionAVR?
>>
>
> DSP might be applicable, if you have enough spare processing  
> capability.

Yes. Unless I read michaelsteinbach's question incorrectly he was  
asking how to do DSP functions using CodevisionAVR.

Maybe after he has tried it he will find the AVR is sufficient. Maybe  
not.

I would suggest a simple averaging of the inputs from the  
accelerometer. If its being shaken then I'd guess half the time its  
going one direction and half the time its going the other direction  
due to vibration. And that the DC offset is the value he is really  
looking for.

Keep a running average. Store a circular buffer of each and every  
sample. Sum the samples as they go in, subtract the old samples as  
they fall off. The average is this value divided by the number of  
samples.

uint16_t samples[COUNT];
uint8_t pos, cnt;
uint32_t sum;

void init_samples(void)
{
     sum = 0;
     pos = cnt = 0;
}

void newsample( uint16_t new )
{
     sum += new;                  //  add to running total
     if( cnt < COUNT ) {
         samples[cnt++] = new;    //  only until samples is full
     } else {
         sum -= samples[pos];     //  remove oldest from running total
         samples[pos++] = new;    //  save the new in oldest space
         if( pos >= COUNT )
             pos = 0;
     }
}

And any time you want to know the average its sum/cnt. If you don't  
mind goofy values the first COUNT samples then remove if( cnt <  
COUNT ) and use sum/COUNT.

void newsample( uint16_t new )
{
     sum += new;
     sum -= samples[pos];
     samples[pos++] = new;
     if( pos >= COUNT )
         pos = 0;
}

--
David Kelly N4HHE, dkelly@HiWAAY.net
========================================================================
Whom computers would destroy, they must first drive mad.

At 03:26 PM 9/11/05 +0200, Ralph Hilton wrote:
>On Sun, 11 Sep 2005 05:13:09 -0000 you wrote:
>
> >I have a circuit with an Accelerometer connected to an ATmega32.  The
> >Accel measures 2.5V at rest and goes to .5v or up to 4.5v when tilted.
> >The motors I am using on this bot cause vibrations when running and I
> >can't get a clean signal fromt he Accel.
> >Is there a way to filter the noise from the motors out of the Accel
> >signal through CodevisionAVR?
> >Thanks!
> >~Mike
>
>A moving average is quite easy to implement. Just put the values into an 
>array and
>maintain a sum, adding the latest value and subtracting the oldest. I got 
>better results
>using a hardware filter - a dual op amp connected as a low pass filter.


I'd second the use of a HW filter.  A simple RC will likely help a lot.  It 
will be useful even if you add software filtering.  It is necessary to get 
the input frequency down so that you avoid 'folding' higher noise 
frequencies into your input where averaging may actually amplify them if 
you are unlucky.  Just as bad I've seen noise spike do funny things to A/Ds 
if there aren't filtered out.  I ran into that when I used a system where I 
knew the sensor bandwidth was low and left out the Nyquist filter, the 
input line picked up spikes from the power switching elements and fed them 
in the A/D lines which led to occasional bad readings.

On a second note it is entirely possible you are getting electrical rather 
than mechanical noise coupled into your system.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,   be 
they legal, genetic, or physical.  If you don't believe me, try to chew a 
radio signal. "  -- Kelvin Throop, III
http://www.aeolusdevelopment.com/

A little more detail on sensor used, how it is read and the nature of the
noise that is troublesome and what the application is would help in
determining what, if any, filtering is needed.  

You can get only so much resolution using RC filter.  In the case of the AVR
the ADC is only 10 bits.  However, if you use an IIR filter (i.e. weighted
average, or exponential filter) you can not only reduce bandwidth and filter
out noise, but you can also increase your resolution one or two bits. The
weighted average filter is the digital version of a single stage RC filter.
A weighted average filter looks something like this:

Output = (prev_output - input)* K1 + input

Where K1 is equal to e^-(dt/tau).  Dt = the sampling period and Tau is the
rise time desired (in seconds).  The frequency response is 1/2*PI*tau.  Or,
tau = 1/2*pi*frequency.  I recently implemented this filter in the following
way:

Int32_t results;

Void InitFilter(int16_t val)
{
	Results = val * 256;
}

void DoFilter(int16_t val)
{
	int32_t t = (int32_t)val * 256;	// Turn into 24.8 fraction

	results = ((((int32_t)K1 * (results - t) + 128L) / 256L) + t);
}

uint16_t GetFilter(void)
{
	return results / 256;	// Divide by 128 or 64 for more 
}					// bits resolution...

K1 is a 8.8 binary fraction e.g.  K1 = 256 * e^-(dt/tau) 

The intermediate result is a 24.8 bit number.  I picked an 8 bit fraction
since it is so easy to convert back & forth if the compiler is clever: just
shuffle bytes.

Weighted average filters (IIR) are compact, computationally efficient and
don't require storing lots of previous values, but they have limited
frequency response & phase shift characteristics.  I recommend Smith's "The
Scientist and Engineers Guide to Digital Signal Processing" for an easy to
read tutorial on practical filter design.

Cheers!

I second Larry's suggestion for using IIR filtering.  I use a sequence 
of them for
my noisy accelerometer applications, (as well as other sense inputs), 
and gain
flexible low-pass filtering and increased resolution.  They are very easy to
implement in AVR.

Ken


Larry Barello wrote:

>A little more detail on sensor used, how it is read and the nature of the
>noise that is troublesome and what the application is would help in
>determining what, if any, filtering is needed.  
>
>You can get only so much resolution using RC filter.  In the case of the AVR
>the ADC is only 10 bits.  However, if you use an IIR filter (i.e. weighted
>average, or exponential filter) you can not only reduce bandwidth and filter
>out noise, but you can also increase your resolution one or two bits. The
>weighted average filter is the digital version of a single stage RC filter.
>A weighted average filter looks something like this:
>
>Output = (prev_output - input)* K1 + input
>
>Where K1 is equal to e^-(dt/tau).  Dt = the sampling period and Tau is the
>rise time desired (in seconds).  The frequency response is 1/2*PI*tau.  Or,
>tau = 1/2*pi*frequency.  I recently implemented this filter in the following
>way:
>
>Int32_t results;
>
>Void InitFilter(int16_t val)
>{
>	Results = val * 256;
>}
>
>void DoFilter(int16_t val)
>{
>	int32_t t = (int32_t)val * 256;	// Turn into 24.8 fraction
>
>	results = ((((int32_t)K1 * (results - t) + 128L) / 256L) + t);
>}
>
>uint16_t GetFilter(void)
>{
>	return results / 256;	// Divide by 128 or 64 for more 
>}					// bits resolution...
>
>K1 is a 8.8 binary fraction e.g.  K1 = 256 * e^-(dt/tau) 
>
>The intermediate result is a 24.8 bit number.  I picked an 8 bit fraction
>since it is so easy to convert back & forth if the compiler is clever: just
>shuffle bytes.
>
>Weighted average filters (IIR) are compact, computationally efficient and
>don't require storing lots of previous values, but they have limited
>frequency response & phase shift characteristics.  I recommend Smith's "The
>Scientist and Engineers Guide to Digital Signal Processing" for an easy to
>read tutorial on practical filter design.
>
>Cheers!
>
>
>
>
> 
>Yahoo! Groups Links
>
>
>
> 
>
>
>
>  
>

At 04:54 PM 9/11/05 -0700, Ken Holt wrote:
>I second Larry's suggestion for using IIR filtering.  I use a sequence
>of them for
>my noisy accelerometer applications, (as well as other sense inputs),
>and gain
>flexible low-pass filtering and increased resolution.  They are very easy to
>implement in AVR.

They may be easy to implement but you still need an external hardware 
filter.  Particularly if you have a 'noisy' signal and or impulse noise.  A 
simple RC can give you a low pass sufficient to allow SW filtering to 
handle the rest but you must remove or at least reduce the frequencies 
above the Nyquist frequency.

Robert


" 'Freedom' has no meaning of itself.  There are always restrictions,   be 
they legal, genetic, or physical.  If you don't believe me, try to chew a 
radio signal. "  -- Kelvin Throop, III
http://www.aeolusdevelopment.com/

Yes, I always put in an external RC between any sensor and the
A-D input. It's a bargain in space and cost, taking advantage
of the high input impedance of the A-D (i.e.no opamp required).
Ken

> At 04:54 PM 9/11/05 -0700, Ken Holt wrote:
>>I second Larry's suggestion for using IIR filtering.  I use a sequence
>>of them for
>>my noisy accelerometer applications, (as well as other sense inputs),
>>and gain
>>flexible low-pass filtering and increased resolution.  They are very easy
>> to
>>implement in AVR.
>
> They may be easy to implement but you still need an external hardware
> filter.  Particularly if you have a 'noisy' signal and or impulse noise.
> A
> simple RC can give you a low pass sufficient to allow SW filtering to
> handle the rest but you must remove or at least reduce the frequencies
> above the Nyquist frequency.
>
> Robert
>
>
> " 'Freedom' has no meaning of itself.  There are always restrictions,   be
> they legal, genetic, or physical.  If you don't believe me, try to chew a
> radio signal. "  -- Kelvin Throop, III
> http://www.aeolusdevelopment.com/
>
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
>

On 12 Sep 2005 at 14:04, kholt@sonic.net wrote:

> 
> Yes, I always put in an external RC between any sensor and the
> A-D input. It's a bargain in space and cost, taking advantage
> of the high input impedance of the A-D (i.e.no opamp required).
> Ken

	Hummm, what do you call "high input impedance" ?

100 Megohm on the mega88. I call that high as well!

Dave Pallant

AVR-Chat@yahoogroups.com wrote on 12/09/2005 19:15:56:

> On 12 Sep 2005 at 14:04, kholt@sonic.net wrote:
>
> >
> > Yes, I always put in an external RC between any sensor and the
> > A-D input. It's a bargain in space and cost, taking advantage
> > of the high input impedance of the A-D (i.e.no opamp required).
> > Ken
>
> � �Hummm, what do you call "high input impedance" ?
>
>
>
>
> ------------------------ Yahoo! Groups Sponsor --------------------~-->
> Get Bzzzy! (real tools to help you find a job). Welcome to the Sweet Life.
> http://us.click.yahoo.com/A77XvD/vlQLAA/TtwFAA/dN_tlB/TM
> --------------------------------------------------------------------~->
>
> �
> Yahoo! Groups Links
>
> <*> To visit your group on the web, go to:
> � � http://groups.yahoo.com/group/AVR-Chat/
>
> <*> To unsubscribe from this group, send an email to:
> � � AVR-Chat-unsubscribe@yahoogroups.com
>
> <*> Your use of Yahoo! Groups is subject to:
> � � http://docs.yahoo.com/info/terms/
> �
>
>
>

�

Promethean Technologies Group Ltd and or associated and or subsidiary companies�:

The views expressed in this communication may not necessarily be the views held by Promethean Technologies Group Ltd and or associated and or subsidiary companies.

This e-mail is for the exclusive use of the addressee(s). Unauthorised disclosure, copying or distribution is prohibited.

This e-mail message has been swept for the presence of computer viruses.

Promethean Technologies Group Ltd and or associated and or subsidiary companies�accepts no liability for any loss resulting from this email transmission.

If you've got a decent sized filter capacitor (e.g. 10n) between the input pin and ground I imagine that would constitute a 'very' low impedance to charge anything inside the Mega88 even if the capacitor was fed from a series 100k.

Dave Pallant

AVR-Chat@yahoogroups.com wrote on 13/09/2005 10:14:40:

> � � From the Mega88 datasheet:
>
> � � "The ADC is optimized for analog signals with an output
> impedance of approximately 10 kOhms or less. If such a source is
> used, the sampling time will be negligible. If a source with higher
> impedance is used, the sampling time will depend on how long time
> the source needs to charge the S/H capacitor, with can vary widely.
> The user is recommended to only use low impedant sources with slowly
> varying signals, since this minimizes the required charge transfer
> to the S/H capacitor."
>
> � � You have to use a low impedance source.
> � �
>
> � � "Signal components higher than the Nyquist frequency (fADC/2)
> should not be present for either kind of channels, to avoid
> distortion from unpredictable signal convolution. The user is
> advised to remove high frequency components with a low-pass filter
> before applying the signals as inputs to the ADC."
>
> � � Care must be taken here. Better use an OpAmp filter.
>
> � � Mark Jordan
>
> On 13 Sep 2005 at 9:44, david.pallant@prometheanworld wrote:
>
> > 100 Megohm on the mega88. I call that high as well!
> > Dave Pallant
> >
> >
> > AVR-Chat@yahoogroups.com wrote on 12/09/2005 19:15:56:
> >
> > > On 12 Sep 2005 at 14:04, kholt@sonic.net wrote:
> > >
> > > >
> > > > Yes, I always put in an external RC between any sensor and the
> > > > A-D input. It's a bargain in space and cost, taking advantage
> > > > of the high input impedance of the A-D (i.e.no opamp required).
> > > > Ken
> > >
> > > �Hummm, what do you call "high input impedance"?
> > >
> > >
> > >
> > >
> > > ------------------------ Yahoo! Groups Sponsor --------------------~--
> > >
> > > Get Bzzzy! (real tools to help you find a job). Welcome to the Sweet
> > Life.
> > > http://us.click.yahoo.com/A77XvD/vlQLAA/TtwFAA/dN_tlB/TM
> > > --------------------------------------------------------------------~-
> > >
> > >
> > >
> > > Yahoo! Groups Links
> > >
> > > <*> To visit your group on the web, go to:
> > > � http://groups.yahoo.com/group/AVR-Chat/
> > >
> > > <*> To unsubscribe from this group, send an email to:
> > > � AVR-Chat-unsubscribe@yahoogroups.com
> > >
> > > <*> Your use of Yahoo! Groups is subject to:
> > > � http://docs.yahoo.com/info/terms/
> > >
> > >
> > >
> > >
> >
> > Promethean Technologies Group Ltd and or associated and or
> subsidiary companies:
> > The views expressed in this communication may not necessarily be
> the views held by
> > Promethean Technologies Group Ltd and or associated and or
> subsidiary companies.
> > This e-mail is for the exclusive use of the addressee(s).
> Unauthorised disclosure,
> > copying or distribution is prohibited.
> > This e-mail message has been swept for the presence of computer viruses.
> > Promethean Technologies Group Ltd and or associated and or subsidiary
> > companiesaccepts no liability for any loss resulting from this
> email transmission.
> >
> >
> > *****************************************************************
> > This email has been checked by the e-Sweeper Service
> > *****************************************************************
> >
> >
> > YAHOO! GROUPS LINKS
> >
> > * � � � Visit your group "AVR-Chat" on the web.
> >
> > * � � � To unsubscribe from this group, send an email to:
> > � � � � AVR-Chat-unsubscribe@yahoogroups.com
> >
> > * � � � Your use of Yahoo! Groups is subject to the Yahoo! Terms
> of Service.
> >
> >
> >
>
> YAHOO! GROUPS LINKS
>
> �Visit your group "AVR-Chat" on the web.
> �
> �To unsubscribe from this group, send an email to:
> �AVR-Chat-unsubscribe@yahoogroups.com
> �
> �Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

�

Promethean Technologies Group Ltd and or associated and or subsidiary companies�:

The views expressed in this communication may not necessarily be the views held by Promethean Technologies Group Ltd and or associated and or subsidiary companies.

This e-mail is for the exclusive use of the addressee(s). Unauthorised disclosure, copying or distribution is prohibited.

This e-mail message has been swept for the presence of computer viruses.

Promethean Technologies Group Ltd and or associated and or subsidiary companies�accepts no liability for any loss resulting from this email transmission.

On Tuesday 13 September 2005 21:57, david.pallant@prometheanworld.com wrote:
> <html><body>
>
>
>
> <p><font size=2 face="sans-serif">If you've got a decent sized filter
> capacitor (e.g. 10n) between the input pin and ground I imagine that
> would constitute a 'very' low impedance to charge anything inside the
> Mega88 even if the capacitor was fed from a series 100k.</font>
> <p><font size=2 face="sans-serif">Dave Pallant</font>
> <br>
> <br><font size=2><tt>AVR-Chat@yahoogroups.com wrote on 13/09/2005
> 10:14:40:<br> <br>
> >     From the Mega88 datasheet:</tt></font>
> <br><font size=2><tt>> <br>
> >     "The ADC is optimized for analog signals with an
> output <br>
> > impedance of approximately 10 kOhms or less. If such a source is
> <br> > used, the sampling time will be negligible. If a source with

Hmmm... not so nice to read, is it?

Please post emails in plain text.

-- 
/"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia
\ /  ASCII ribbon campaign | I'm a .signature virus!
 X   against HTML mail     | Copy me into your ~/.signature
/ \  and postings          | to help me spread!

> Hmmm... not so nice to read, is it?

In digest mode I only get:

-Message: 2         
-   Date: Tue, 13 Sep 2005 14:57:59 +0100
-   From: david.pallant@prometheanworld.com
-Subject: Re: Sensor filtering
-
-
-[This message is not in displayable format]
-
-

So if you want to reach all members of the list: Post in ASCII!

On 11 Sep 2005 at 14:15, Larry Barello wrote:

> The weighted average filter is the digital version of a single
> stage RC filter. A weighted average filter looks something like
> this: 
> 
> Output = (prev_output - input)* K1 + input
> 
> Where K1 is equal to e^-(dt/tau).  Dt = the sampling period and Tau is the
> rise time desired (in seconds).  The frequency response is 1/2*PI*tau.
> Or, tau = 1/2*pi*frequency.
> 

	Is there an 'easy to implement' second-order IIR filter formula?

	Thanks.
	Mark Jordan

Do a Google search on Infinite Impulse Response filters.  

I first learned about them in the book "The scientist and Engineers guide to
digital signal processing", Steven Smith.  Chapter 19, 20, ...

You can make pretty fancy filters by feeding forward and backwards several
discrete time values (inputs and outputs).

I have never used anything more than a simple single order filter.


-----------
Larry Barello
www.barello.net


| -----Original Message-----
| From: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] On Behalf
| Of Mark Jordan
| Sent: Monday, September 19, 2005 8:30 PM
| To: AVR-Chat@yahoogroups.com
| Subject: RE: [AVR-Chat] Sensor filtering
| 
| On 11 Sep 2005 at 14:15, Larry Barello wrote:
| 
| > The weighted average filter is the digital version of a single
| > stage RC filter. A weighted average filter looks something like
| > this:
| >
| > Output = (prev_output - input)* K1 + input
| >
| > Where K1 is equal to e^-(dt/tau).  Dt = the sampling period and Tau is
| the
| > rise time desired (in seconds).  The frequency response is 1/2*PI*tau.
| > Or, tau = 1/2*pi*frequency.
| >
| 
| 	Is there an 'easy to implement' second-order IIR filter formula?
| 
| 	Thanks.
| 	Mark Jordan
| 
| 
| 
| 
| 
| 
| Yahoo! Groups Links
| 
| 
| 
| 
| 
|

Thanks, I have found that book online at http://www.dspguide.com

	Mark Jordan

On 19 Sep 2005 at 21:38, Larry Barello wrote:

> Do a Google search on Infinite Impulse Response filters.  
> 
> I first learned about them in the book "The scientist and Engineers guide to
> digital signal processing", Steven Smith.  Chapter 19, 20, ...
> 
> You can make pretty fancy filters by feeding forward and backwards several
> discrete time values (inputs and outputs).
> 
> I have never used anything more than a simple single order filter.
> 
> 
> -----------
> Larry Barello
> www.barello.net
> 
> 
> | -----Original Message-----
> | From: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] On Behalf |
> Of Mark Jordan | Sent: Monday, September 19, 2005 8:30 PM | To:
> AVR-Chat@yahoogroups.com | Subject: RE: [AVR-Chat] Sensor filtering | | On 11
> Sep 2005 at 14:15, Larry Barello wrote: | | > The weighted average filter is the
> digital version of a single | > stage RC filter. A weighted average filter looks
> something like | > this: | > | > Output = (prev_output - input)* K1 + input | >
> | > Where K1 is equal to e^-(dt/tau).  Dt = the sampling period and Tau is | the
> | > rise time desired (in seconds).  The frequency response is 1/2*PI*tau. | >
> Or, tau = 1/2*pi*frequency. | > | | 	Is there an 'easy to implement'
> second-order IIR filter formula? | | 	Thanks. | 	Mark Jordan | | | | | | |
> Yahoo! Groups Links | | | | | | 
> 
> 
> 
> 
> 
> 
> 
> Yahoo! Groups Links
> 
> 
> 
> 
> 
> 
>