Searching \ for '[EE]: Analog peak capture' 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/atod.htm?key=analog
Search entire site for: 'Analog peak capture'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Analog peak capture'
2004\03\30@112850 by Joe Jansen/TECH/HQ/KEMET/US

flavicon
face
Hey everyone!

I have been unsuccessful googling, and was hoping someone here would
either have an answer, a web link, or even recommended reading for me.  I
am trying to build an electronic drum set similar to the overpriced models
I see at the music store.  My thought is to use a piezo element embedded
in a rubber pad, through an op-amp, and feed the signal to an A/D
converter to do the velocity detection.  This would then go to a PIC,
which would convert the velocity info into a MIDI signal and feed it to
the keyboard for wave generation and playback.  Not too terribly fancy,
but should be functional.

My question is:  How do I insure that I am running my A/D conversion on
the highest point of the piezo elements output?  the spike that is
generated will be proportional to the strength of the impact, but I want
to make sure that I run my conversion on the highest point of the output.
Is there a way to capture the peak output and then run the A/D conversion
on that?  Any other suggestions?  I definitely want the velocity info to
control the volume of the playback, or the whole thing becomes a gimmick.

Thanks for any suggestions!!!!!

--Joe Jansen

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spam_OUTlistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body

2004\03\30@114507 by Alessandro Queri

flavicon
face
...increasing enough sampling rate? ;-)
Consider you will not achieve a great linearity with piezos: I happened to
see on a drumset they use single chip accelerometers (made by AD, but I'm
not sure...)

Ale

On Tue, 30 Mar 2004, Joe Jansen/TECH/HQ/KEMET/US wrote:

{Quote hidden}

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservspamKILLspammitvma.mit.edu with SET PICList DIGEST in the body

2004\03\30@162403 by Joe Jansen/TECH/HQ/KEMET/US

flavicon
face
<G>

Sampling rate is going to be another issue for me to resolve, anyway.  I
did get a couple good suggestions for the Peak hold amp.

Also, I have a pair of ADXL150JQC accelerometer samples coming to try out,
per your suggestion.

Thanks all.  I know there are a few people here who play with MIDI, so I
may be back once I get to dabbling with that portion.  First task is
getting the ADXL's in and tied into Brian's Pk hold circuit.

--Joe Jansen




pic microcontroller discussion list <.....PICLISTKILLspamspam.....MITVMA.MIT.EDU> wrote on
03/30/2004 11:43:42 AM:

> ...increasing enough sampling rate? ;-)
> Consider you will not achieve a great linearity with piezos: I happened
to
> see on a drumset they use single chip accelerometers (made by AD, but
I'm
{Quote hidden}

models
> > I see at the music store.  My thought is to use a piezo element
embedded
> > in a rubber pad, through an op-amp, and feed the signal to an A/D
> > converter to do the velocity detection.  This would then go to a PIC,
> > which would convert the velocity info into a MIDI signal and feed it
to
> > the keyboard for wave generation and playback.  Not too terribly
fancy,
> > but should be functional.
> >
> > My question is:  How do I insure that I am running my A/D conversion
on
> > the highest point of the piezo elements output?  the spike that is
> > generated will be proportional to the strength of the impact, but I
want
> > to make sure that I run my conversion on the highest point of the
output.
> > Is there a way to capture the peak output and then run the A/D
conversion
> > on that?  Any other suggestions?  I definitely want the velocity info
to
> > control the volume of the playback, or the whole thing becomes a
gimmick.
{Quote hidden}

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email @spam@listservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body

2004\03\30@222140 by Ken Pergola

flavicon
face
Joe Jansen wrote:

> Thanks for any suggestions!!!!!


Hi Joe,

I still have my E-mu Systems Drumulator drum machine (I love the '80s!). I
never bought the pad programmer for it, but here is some information that
might be useful to you regarding their piezo interface:



From E-mu Systems Drumulator Pad Programmer manual:
---------------------------------------------------

{
Pad Programmer - Theory of Operation - Revision of 2/27/84
The Pad Programmer consists of four independent trigger sensors with
appropriate
circuitry, with a common housing and power supply.
The power supply is a simple three terminal regulator +5V supply. The input
voltage is
between 8 and 25 VDC and the current requirements are small enough (around
40 mA
when an LED is firing, less quiescently) that heat sinking is not required.
The plug in
wall unit supplies nominally 10V DC, which is filtered and regulated.
The sensor is a piezo-electric crystal mounted in a housing designed to
transmit
pressure from a vertical strike by a drumstick, and attenuate vibrations
conducted by the
housing. This assembly is glued together and tested at the factory. Service
on these
assemblies will generally be by replacement, as repairs are difficult to
accomplish and
will frequently be unreliable.
The sensor produces a substantial voltage spike (10 to 100+ volts), which is
loaded by a
parallel capacitor and resistor, and rectified by a full wave bridge. The
resulting positive
voltage is attenuated and applied to one terminal of a comparator. The other
terminal of
the comparator is adjusted for the desired sensitivity threshold. The trim
range is set to
stay within the common mode range of the comparator. Positive feedback
around the
comparator is provided to minimize marginal switching. The comparator output
triggers
a one-shot which will both swallow any comparator oscillation or multiple
triggering, and
provide sufficient pulse width for the LED amplifier and for the output
trigger pulse. Both
of these signals are buffered from the one-shot output.
For information on adjusting the sensitivity trims, refer to the earlier
section.
}


Best regards,

Ken Pergola

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email KILLspamlistservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body

2004\03\31@091539 by Joe Jansen/TECH/HQ/KEMET/US

flavicon
face
Thx Ken.  Correct me if I am wrong, but it sounds like it only detects
on/off, rather than impact strength.  I wasn't aware that the piezo's put
out that high of a voltage.  hmmmmm.........

It does give me some evidence that the piezo will do the job, however.  I
still want to try out the accelerometers, also.

Good stuff!

--Joe Jansen

pic microcontroller discussion list <RemoveMEPICLISTTakeThisOuTspamMITVMA.MIT.EDU> wrote on
03/30/2004 10:20:36 PM:

> Joe Jansen wrote:
>
> > Thanks for any suggestions!!!!!
>
>
> Hi Joe,
>
> I still have my E-mu Systems Drumulator drum machine (I love the '80s!).
I
> never bought the pad programmer for it, but here is some information
that
{Quote hidden}

input
> voltage is
> between 8 and 25 VDC and the current requirements are small enough
(around
> 40 mA
> when an LED is firing, less quiescently) that heat sinking is not
required.
> The plug in
> wall unit supplies nominally 10V DC, which is filtered and regulated.
> The sensor is a piezo-electric crystal mounted in a housing designed to
> transmit
> pressure from a vertical strike by a drumstick, and attenuate vibrations
> conducted by the
> housing. This assembly is glued together and tested at the factory.
Service
> on these
> assemblies will generally be by replacement, as repairs are difficult to
> accomplish and
> will frequently be unreliable.
> The sensor produces a substantial voltage spike (10 to 100+ volts),
which is
> loaded by a
> parallel capacitor and resistor, and rectified by a full wave bridge.
The
> resulting positive
> voltage is attenuated and applied to one terminal of a comparator. The
other
> terminal of
> the comparator is adjusted for the desired sensitivity threshold. The
trim
> range is set to
> stay within the common mode range of the comparator. Positive feedback
> around the
> comparator is provided to minimize marginal switching. The comparator
output
> triggers
> a one-shot which will both swallow any comparator oscillation or
multiple
{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
TakeThisOuTpiclist-unsubscribe-requestEraseMEspamspam_OUTmitvma.mit.edu

2004\03\31@110628 by Ken Pergola

flavicon
face
Joe Jansen wrote:

> Correct me if I am wrong, but it sounds like it only detects
> on/off, rather than impact strength.  I wasn't aware that the piezo's put
> out that high of a voltage.  hmmmmm.........

Hi Joe,

I've never used the E-mu Systems Pad Programmer (only the Drumulator
itself), but I suspect that you are right. The Drumulator, had level
controls for all of the drum sounds, but getting dynamics out the drum
sounds via the pad programmer probably is not possible given the manual
excerpt below.

Have you done any research on Roland's V-Drums? That is much newer
technology.
http://www.rolandus.com


Best regards,

Ken Pergola




From the Drumulator Pad Programmer manual:

Adjusting Pad Sensitivity

The trigger threshold of each pad (i.e. how hard you have to hit the pad in
order to have
it cause the Drumulator to make a Sound) is set with the sensitivity
trimmers accessible
through the rear of the Pad Programmer. Turning a trimmer wheel to the right
(clockwise) decreases sensitivity (you'll have to hit the pad harder) while
turning it to the
left (counterclockwise) increases sensitivity.
The object in adjusting each pad is to get the maximum sensitivity that
doesn't result in
"crosstalk". Crosstalk occurs when you hit a pad so hard that the impact is
transmitted
to an adjacent pad, causing it to trigger. By properly adjusting each pad,
crosstalk can
be eliminated in all but the most extreme cases. The actual adjustment
depends on your
playing style. If you play with a light touch, the sensitivity can be set
rather high. If, on
the other hand, you typically pound your drums into submission, a lower
sensitivity level
is called for.

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestspamTakeThisOuTmitvma.mit.edu

2004\03\31@133212 by Hazelwood Lyle

flavicon
face
>-----Original Message-----
>From: Joe Jansen/TECH/HQ/KEMET/US [JoeJansenEraseMEspam.....KEMET.COM]

An intersting look at one solution, complete with schematics and theory of operation:

http://paia.com/drumsens.htm


Lyle

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspammitvma.mit.edu

2004\03\31@160507 by John N. Power

flavicon
face
> From:         Joe Jansen/TECH/HQ/KEMET/US[SMTP:RemoveMEJoeJansenEraseMEspamEraseMEKEMET.COM]
> Sent:         Tuesday, March 30, 2004 11:27 AM
> To:   RemoveMEPICLISTspam_OUTspamKILLspamMITVMA.MIT.EDU
> Subject:      [EE]: Analog peak capture

> Hey everyone!

. . . .

{Quote hidden}

A simple peak detector can be made from a diode and a capacitor.The signal
is applied to the anode of the diode. The cathode (banded end) of the diode
goes to one end of the capacitor and to the A/D input. The other end of the
capacitor goes to ground. The capacitor will charge to the highest value of
the input; when the input falls below the capcitor voltage, the diode turns off
preventing the capacitor from discharging. After digitizing the voltage, you
can discharge the capacitor by setting the PIC port pin to digital output and
taking it to zero.

The following ideas will be useful:

Remember that the capacitor voltage will be one diode drop below the peak
input voltage.
To protect the port pin during discharge, put a series resistor between the
top of the capacitor and the pin. Keep the resistance low enough to satisfy
the maximum source inpedance allowed by the A/D (typically 10k).
The larger the capacitor, the more current it will need in order to charge
up in a short time. The driving amplifier must be able to provide this
current.
For more precise work, there are op amp circuits available which put
the capacitor inside of a feedback loop. Some of the other replies to your
post undoubtably will refer you to these.

John Power

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestTakeThisOuTspamspammitvma.mit.edu


'[EE]: Analog peak capture'
2004\04\02@233715 by PicDude
flavicon
face
How about a microphone to actually detect the sound level?  You could use simple bandpass filters to prevent picking up stray signals from the other drums.

Cheers,
-Neil.



On Wednesday 31 March 2004 08:14 am, Joe Jansen/TECH/HQ/KEMET/US scribbled:
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\04\05@083553 by Koen van Leeuwen

flavicon
face
On Saturday 03 April 2004 06:38, PicDude wrote:
> How about a microphone to actually detect the sound level?  You could use
> simple bandpass filters to prevent picking up stray signals from the other
> drums.
>
> Cheers,
> -Neil.

Perhaps glueing a small speaker to the drum is better - most speakers give
very clean signals when tapped on, but the mechanical contact will assure it
ignores tapping on the other drums.

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestSTOPspamspamspam_OUTmitvma.mit.edu

More... (looser matching)
- Last day of these posts
- In 2004 , 2005 only
- Today
- New search...