I have some troubles using ISD, I'm using an amplifier to get 3W or so,
but when the message ends, a "pop" is heard, I'm tryng to eliminate it but I
couldn't up to now...
I also used a LM386 (suggested by manufacturer) but nothing, if somebody
had this problem before and, fixed it, please help...
> ISD's manuals discuss this in detail for several of their parts, I can't
> remember where right now but I suggest you search.
>
> Bruce Cannon
> Style Management Systems
> http://siliconcrucible.com
> (510) 787-6870
> 1228 Ceres ST Crockett CA 94525
>
> Remember: electronics is changing your world...for good!
>
Gonzalo wrote:
> Hi all,
>
> I have some troubles using ISD, I'm using an amplifier to get 3W or so,
>but when the message ends, a "pop" is heard, I'm tryng to eliminate it but I
>couldn't up to now...
>
> I also used a LM386 (suggested by manufacturer) but nothing, if somebody
>had this problem before and, fixed it, please help...
>
Just a thought - if you use a PIC to time the message, then
maybe you can use it to turn off the amplifier output just
before the pop occurs.
>> I have some troubles using ISD, I'm using an amplifier to get 3W or so,
>>but when the message ends, a "pop" is heard, I'm tryng to eliminate it but I
>>couldn't up to now...
Does this ISD have a balanced output? If you are amplifying it, you could
use your amp as a differential amp hooked to both outputs and you'll cancel
the pop. Note that will not take care of any pop that the amp has by itself.
Alternately if your amp has a bypass pin, use a very large capacitor on it
so it biases slower than all the other caps in the system.
Cheers,
Bob
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
Gonzalo Jiménez Yamasaki wrote:
>
> Hi all,
>
> I have some troubles using ISD, I'm using an amplifier to get 3W or so,
> but when the message ends, a "pop" is heard, I'm tryng to eliminate it but I
> couldn't up to now...
>
> I also used a LM386 (suggested by manufacturer) but nothing, if somebody
> had this problem before and, fixed it, please help...
>
> Thanks in advance,
>
> Gonzalo
I have had "pop" problems many times with digitally sampled
sound, I'm fairly sure that your digital samples are not being "faded"
back to zero (to mid waveform) when they end. The further the sample
end point is from the zero level (ie too high or too low) the
louder it will pop when the sample ends.
If this is the problem you have two options, either fade the sample
to zero in software or with hardware. You only need a very quick
fade, ie about 1/10th second or even less. Hope this is of some
help??
-Roman
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:"
I think that the ISD1420 drops the output to 0 when the playback stops.
This happens very abruptly. Use a steep low pass filter between the ISD
and the amp, and also a mute circuit using a transistor to sense the DC
level at the ISD output. This can be used to turn off the amplifier power
(after the filter capacitor) using a high side switch.
Peter
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
At 07:20 PM 10/13/00 -0500, you wrote:
> I've done, but nothing...
Hi,
Go to their web page and search on 'pop', first result gets you
to an on-line FAQ.
There are a couple of things, first the possibility that there is a pop
*recorded* with the sound (due to an on-board switch click, say), and
second that you are not using the differential outputs to get rid of the
1/2 Vdd bias shift.
There may be something else that is chip dependent, but those two are
fairly generic.
Best regards,
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..." "The Journey is the reward" EraseMEspeffspam_OUTTakeThisOuTinterlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
The "pop" is almost certainly due to the manner in which the recording is
terminated.
To prove this try the following - record to the ISD continually until it
reaches the end of the memory. Do NOT stop the recording. Now listen to the
result, If there is no pop but a sudden clean end to the recorded message it
shows that what you are doing is the cause of the noise.
If so, look at how you are ending the recording - are you removing the
record signal?
Is the microphone supply still connected at this stage? - perhaps you remove
the microphone supply first and the chip is 'hearing" this.
If you remove them both together then a small delay may be useful.
> Hi all,
>
> I have some troubles using ISD, I'm using an amplifier to get 3W or so,
>but when the message ends, a "pop" is heard, I'm tryng to eliminate it but I
>couldn't up to now...
>
> I also used a LM386 (suggested by manufacturer) but nothing, if somebody
>had this problem before and, fixed it, please help...
>
> Thanks in advance,
>
> Gonzalo
I have used the ISD chips a lot. You MUST use the differential outputs
differentially, as they drop sharply to 0V after the message. I
usually use a stereo power amp chip in bridge mode, which also
eliminates the need for an output capacitor. Some single-ended amps
may be useable in differential mode (e.g. LM380N8), otherwise you need
to add a differential amp stage - an LM358 + 4 resistors costs next to
nothing and works fine.
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:"
> The "pop" is almost certainly due to the manner in which the recording is
> terminated.
> To prove this try the following - record to the ISD continually until it
> reaches the end of the memory. Do NOT stop the recording. Now listen to
the
> result, If there is no pop but a sudden clean end to the recorded message
it
> shows that what you are doing is the cause of the noise.
I did it, but the pop still there...
I'm also using a diferential amplifier, using LM386...
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
> There are a couple of things, first the possibility that there is a pop
> *recorded* with the sound (due to an on-board switch click, say), and
This is a problem that had really bothered me and I was surprised to finally
find that I was actually recording the sound of the switch! And of course
solved it with less-stupid layout...
Bruce Cannon
Style Management Systems http://siliconcrucible.com
(510) 787-6870
1228 Ceres ST Crockett CA 94525
Remember: electronics is changing your world...for good!
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
First of all, thanks again to all of you, I've finally solve my problem.
Solution:
At the output of the amplifier, I put a diode to rectify the AC signal,
then use a cap. (between cathode and ground) in parallel with a resistor,
with that I obtain a DC level while a message is playing, when the messages
ends this DC level goes to zero, I use this falling edge to interrupt a PIC
and shut the amplifier off.
Well, now I have another problem, how to reduce "buzz"...
I guess it's little enough but, I'd like to improve it... does anybody
have some secrets to fight buzz?
Thanks in advance, again...
Gonzalo
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
wrote:
> Hi, and thanks to all for your help...
>
>
>> The "pop" is almost certainly due to the manner in which the recording is
>> terminated.
>> To prove this try the following - record to the ISD continually until it
>> reaches the end of the memory. Do NOT stop the recording. Now listen to
>the
>> result, If there is no pop but a sudden clean end to the recorded message
>it
>> shows that what you are doing is the cause of the noise.
>
> I did it, but the pop still there...
>
> I'm also using a diferential amplifier, using LM386...
>
I missed some of this thread. Is the pop inserted at the end of the
msg during recording, or an artifact produced during playback?
-- http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other "[BUY]:","[AD]:" =Ads
On 14 Oct 00 at 20:29, Gonzalo Jiménez Yamasaki wrote:
>
> Well, now I have another problem, how to reduce "buzz"...
>
> I guess it's little enough but, I'd like to improve it... does anybody
> have some secrets to fight buzz?
>
Is the 'buzz' 50/60Hz ?
The most generic answer would be to shorten or lower the resistance
of the common/ return tracks...i.e 0volts at one part of the circuit
should pretty close to the 0volts at another.
If the psu is from a mains source then try running the unit on
batteries, if the noise goes away then more filtering is needed on
the psu.
Without more details I cannot offer more solutions (there could be
thousands of possible causes).
Lance Allen
Technical Officer
Embedded Systems Lab
Computer Systems Engineering
Department of Electrical
and Electronic Engineering
University of Auckland
New Zealand
> First of all, thanks again to all of you, I've finally solve my
problem.
> Solution:
>
> At the output of the amplifier, I put a diode to rectify the AC
signal,
> then use a cap. (between cathode and ground) in parallel with a resistor,
> with that I obtain a DC level while a message is playing, when the
messages
> ends this DC level goes to zero, I use this falling edge to interrupt a
PIC {Quote hidden}
> and shut the amplifier off.
>
> Well, now I have another problem, how to reduce "buzz"...
>
> I guess it's little enough but, I'd like to improve it... does anybody
> have some secrets to fight buzz?
>
> Thanks in advance, again...
>
> Gonzalo
>
> --
> http://www.piclist.com hint: PICList Posts must start with ONE topic:
> "[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other
> wrote:
> > Hi, and thanks to all for your help...
> >
> >
> >> The "pop" is almost certainly due to the manner in which the recording
is
> >> terminated.
> >> To prove this try the following - record to the ISD continually until
it
> >> reaches the end of the memory. Do NOT stop the recording. Now listen to
> >the
> >> result, If there is no pop but a sudden clean end to the recorded
message {Quote hidden}
> >it
> >> shows that what you are doing is the cause of the noise.
> >
> > I did it, but the pop still there...
> >
> > I'm also using a diferential amplifier, using LM386...
> >
>
>
> I missed some of this thread. Is the pop inserted at the end of the
> msg during recording, or an artifact produced during playback?
>
> --
> http://www.piclist.com hint: PICList Posts must start with ONE topic:
> "[PIC]:","[SX]:","[AVR]:" =uP ONLY! "[EE]:","[OT]:" =Other
>The most generic answer would be to shorten or lower the resistance
>of the common/ return tracks...i.e 0volts at one part of the circuit
>should pretty close to the 0volts at another.
I'll try this...
>If the psu is from a mains source then try running the unit on
>batteries, if the noise goes away then more filtering is needed on
>the psu.
No problem...
Well, I guess ISD chip records some noise, is it possible to eliminate
it? maybe while recording... any tricks?
> Well, I guess ISD chip records some noise, is it possible to eliminate
> it? maybe while recording... any tricks?
Once the board layout is good (nice separation between ana and dig, plenty
of bypassing, fat short traces where needed, etc per datasheets), using the
sound recorded with no mic is a good basis for comparison. Then for a
second comparison make one no-mic recording with a battery and the other
with a line supply. Changing AGC and playing with other values can also be
instructive. But if you mean below the noise threshold of the system's
design...?
Bruce Cannon
Style Management Systems http://siliconcrucible.com
(510) 787-6870
1228 Ceres ST Crockett CA 94525
Remember: electronics is changing your world...for good!
mitvma.mit.edu
