question about 16ohm to 8ohm speaker

[Joe]

I have a small (digital, I guess) recorder/player that I bought from the rat
shack last week. It runs on a 9 volt battery and you can record 20seconds
worth of sound and then play it back. It has a 16 ohm speaker on the output
(a really fragile piezo type (?), with a clear plastic film over it). My
problem is that it does not play back loud enough. I am measuring about 2V
peak to peak signal on my oscope with the speaker disconnected (unloaded). I
would like to change the 16 ohm speaker to a better 8ohm speaker and maybe
be able to boost the signal along the way, maybe have a pot to control the
volume. When I connect the 8ohm speaker in place of the 16ohm speaker, the
volume drops substantially (no surprise there, I guess).

I first tried connecting the output to a LM386 audio amp, and then
outputting that to an 8ohm speaker (LM386 was wired for minimum parts count,
gain of 20). There was a lot of distortion, enough so that I could not
really hear the sound that had been recorded originally. All kinds of weird
noises were coming out of it, but it was really loud!

Then I thought of replacing the LM386 with a common collector type amp to
act as a buffer for the signal. I used a 2N2222 transistor as a common
collector, but volume was still low. It did improve a little with a 2N4401,
but is still not loud enough.

Any ideas on how I could boost the signal enough to tweak the volume with a
pot, and drive an 8ohm speaker ? I don't know the wattage of the 8ohm
speaker, it was salvaged from something, but it is about 2 inches in
diameter.


TIA,
Joe

 

[tempus fugit]

When I connect the 8ohm speaker in place of the 16ohm speaker, the

volume drops substantially (no surprise there, I guess).

I would think it would be louder - lower Z = more current coming out of the
device.

I first tried connecting the output to a LM386 audio amp, and then
outputting that to an 8ohm speaker (LM386 was wired for minimum parts count,
gain of 20). There was a lot of distortion, enough so that I could not
really hear the sound that had been recorded originally. All kinds of weird
noises were coming out of it, but it was really loud!

Then I thought of replacing the LM386 with a common collector type amp to
act as a buffer for the signal. I used a 2N2222 transistor as a common
collector, but volume was still low. It did improve a little with a 2N4401,
but is still not loud enough.

A small signal transistor like these is not really capable of delivering
enough power to drive a speaker to any appreciable level.

Why not go back to the 386 and drop the gain to, say, 5 or 10 and see what
the sound is like then? You can easily connect a volume control to the 386
as well. The 386 was developed specifically for applications like yours, I
think.

 

[Goran Tomaš]

volume. When I connect the 8ohm speaker in place of the 16ohm speaker, the
volume drops substantially (no surprise there, I guess).

Actually, the volume should have got louder (less resistance, more
current) but I guess the 8 Ohm speaker is much less sensitive...

I first tried connecting the output to a LM386 audio amp, and then
outputting that to an 8ohm speaker (LM386 was wired for minimum parts count,
gain of 20). There was a lot of distortion, enough so that I could not
really hear the sound that had been recorded originally. All kinds of weird
noises were coming out of it, but it was really loud!

Go back there and either reduce the gain or (better) put a pot between
output and LM386 input to act as a volume control.


Regards,
Goran Tomas

 

[Joe]

When I connect the 8ohm speaker in place of the 16ohm speaker, the


I would think it would be louder - lower Z = more current coming out of the

A small signal transistor like these is not really capable of delivering
enough power to drive a speaker to any appreciable level.

Why not go back to the 386 and drop the gain to, say, 5 or 10 and see what
the sound is like then? You can easily connect a volume control to the 386
as well. The 386 was developed specifically for applications like yours, I
think.

Thanks tempus, I can try that. I am not sure how to connect the LM386 for a
gain of 5 tho. I followed the data sheet and it shows how to connect for a
gain of 20, but I dont see a feedback and source resistor like when using a
regular opamp. Which components would I change for a gain of 5?

Joe

 

[Joe]

Actually, the volume should have got louder (less resistance, more
current) but I guess the 8 Ohm speaker is much less sensitive...


Go back there and either reduce the gain or (better) put a pot between
output and LM386 input to act as a volume control.


Regards,
Goran Tomas

Hi Goran,

thank you for the advice, when I do that, I get a squealing noise coming
from the speaker at certain ranges of the 10K pot. I still have it at a gain
of 20 tho, because that's the circuit that is shown in the data sheet. If I
knew how to reduce the gain, I think it would work. I am not sure which
components to change to bring the gain to about 5.

Joe

 

[Ian Bell]

I have a small (digital, I guess) recorder/player that I bought from the
rat shack last week. It runs on a 9 volt battery and you can record
20seconds worth of sound and then play it back. It has a 16 ohm speaker on
the output (a really fragile piezo type (?), with a clear plastic film
over it). My problem is that it does not play back loud enough. I am
measuring about 2V peak to peak signal on my oscope with the speaker
disconnected (unloaded). I would like to change the 16 ohm speaker to a
better 8ohm speaker and maybe be able to boost the signal along the way,
maybe have a pot to control the volume. When I connect the 8ohm speaker in
place of the 16ohm speaker, the volume drops substantially (no surprise
there, I guess).

I first tried connecting the output to a LM386 audio amp, and then
outputting that to an 8ohm speaker (LM386 was wired for minimum parts
count, gain of 20). There was a lot of distortion, enough so that I could
not really hear the sound that had been recorded originally. All kinds of
weird noises were coming out of it, but it was really loud!

Then I thought of replacing the LM386 with a common collector type amp to
act as a buffer for the signal. I used a 2N2222 transistor as a common
collector, but volume was still low. It did improve a little with a
2N4401, but is still not loud enough.

Any ideas on how I could boost the signal enough to tweak the volume with
a pot, and drive an 8ohm speaker ? I don't know the wattage of the 8ohm
speaker, it was salvaged from something, but it is about 2 inches in
diameter.

I think you were on the right lines with the LM386. However you really do
not need the gain of 20. I suspect the output impedance of the RS device
is relatively high which is why you get lees when you use an 8ohm speaker.
Try again with unity gain.

Ian

 

[Joe]

I think you were on the right lines with the LM386. However you really do
not need the gain of 20. I suspect the output impedance of the RS device
is relatively high which is why you get lees when you use an 8ohm speaker.
Try again with unity gain.

Ian

Hi Ian,

Which components do I change to achieve a lower gain?
Is there a formula for the LM386? The data sheet I have says "the gain is
internally set to 20 to keep external part count low, but the addition of an
external resistor and capacitor between pins 1 and 8 will increase the gain
to any value between 20 and 200."

So if I read that right, 20 is the lowest gain I can achieve with the LM386.
Is that right?

Thanks,
Joe

 

[John Popelish]

Hi Ian,

Which components do I change to achieve a lower gain?
Is there a formula for the LM386? The data sheet I have says "the gain is
internally set to 20 to keep external part count low, but the addition of an
external resistor and capacitor between pins 1 and 8 will increase the gain
to any value between 20 and 200."

So if I read that right, 20 is the lowest gain I can achieve with the LM386.
Is that right?

The LM386 is not guaranteed to be stable for closed loop gains less
than 9 (page 3 of the data sheet talks about how to go that low), but
you can easily divide the input signal by any amount including 20
before applying it to the amplifier. Just connect your signal to one
end of a pot with the other end grounded, and feed the amplifier from
the wiper. See the first figure under Typical Applications, page 5.
http://www.nari.ee.ethz.ch/teaching/PPS/PPS02/doc/LM386.pdf

 

[Goran Tomaš]

thank you for the advice, when I do that, I get a squealing noise coming
from the speaker at certain ranges of the 10K pot. I still have it at a gain
of 20 tho, because that's the circuit that is shown in the data sheet. If I
knew how to reduce the gain, I think it would work. I am not sure which
components to change to bring the gain to about 5.

Joe,

the total gain in the system is product of pot attenuation and
amplifier gain. So, if the gain of the amplifier is 20 and you adjust
a potentiometer so that it creates a voltage drop of 1/20, you will
have 20 * 1/20 = 1, unity gain.

A pot before the amp will give you a range from complete attenuation
(silence) when wiper is turned all the way to the one side (and that
side is grounded) to max amplification of 20 when the wiper is turned
all the way to the other side (and that side is connected to the
output of a device).


Regards,
Goran Tomas

 

[Joe]

The LM386 is not guaranteed to be stable for closed loop gains less
than 9 (page 3 of the data sheet talks about how to go that low), but
you can easily divide the input signal by any amount including 20
before applying it to the amplifier. Just connect your signal to one
end of a pot with the other end grounded, and feed the amplifier from
the wiper. See the first figure under Typical Applications, page 5.
http://www.nari.ee.ethz.ch/teaching/PPS/PPS02/doc/LM386.pdf

Hi John,

Thanks for the link. I think that is the same data sheet I was reading from.
I just didn't read down far enough. The pot at the input works as long as I
don't turn it up too far. If I do, the speaker starts squealing and I see
very noisy waveforms on the scope. I am using a single turn pot tho. Later
on today, I was going to use a 10 turn pc mount pot and see if I can tweak
it a little better.

Thank you,
Joe

 

[Joe]

Joe,

the total gain in the system is product of pot attenuation and
amplifier gain. So, if the gain of the amplifier is 20 and you adjust
a potentiometer so that it creates a voltage drop of 1/20, you will
have 20 * 1/20 = 1, unity gain.

A pot before the amp will give you a range from complete attenuation
(silence) when wiper is turned all the way to the one side (and that
side is grounded) to max amplification of 20 when the wiper is turned
all the way to the other side (and that side is connected to the
output of a device).


Regards,
Goran Tomas

Thanks Goran,
I am working on it, I think I need a multiturn pot, because there's a spot
on the single turn pot where the output is very low and then suddenly jumps
loud with lots of noise. I was going to try a 10 turn 10k pot later on
today.

Joe

 

[John Fields]

Thanks Goran,
I am working on it, I think I need a multiturn pot, because there's a spot
on the single turn pot where the output is very low and then suddenly jumps
loud with lots of noise. I was going to try a 10 turn 10k pot later on
today.

---
Sounds like you've got a bad pot.

In any case, since your R/S thing is designed to drive a low impedance
(16 ohms, you say) you may just be buying yourself trouble with that 10k
pot. Try 1000 ohms or even 100 ohms to see what happens. I'll bet if
you do your problem will go away.

 

[Michael A. Terrell]

Sounds like you've got a bad pot.

In any case, since your R/S thing is designed to drive a low impedance
(16 ohms, you say) you may just be buying yourself trouble with that 10k
pot. Try 1000 ohms or even 100 ohms to see what happens. I'll bet if
you do your problem will go away.

Or even a 20 ohm speaker fader control.

--
We now return you to our normally scheduled programming.

Take a look at this little cutie! ;-)
http://home.earthlink.net/~mike.terrell/photos.html

Michael A. Terrell
Central Florida

 

[Joe]

---
Sounds like you've got a bad pot.

In any case, since your R/S thing is designed to drive a low impedance
(16 ohms, you say) you may just be buying yourself trouble with that 10k
pot. Try 1000 ohms or even 100 ohms to see what happens. I'll bet if
you do your problem will go away.

Thanks John, I'll give that a try.

Joe

 

[John Popelish]

Thanks for the link. I think that is the same data sheet I was reading from.
I just didn't read down far enough. The pot at the input works as long as I
don't turn it up too far. If I do, the speaker starts squealing and I see
very noisy waveforms on the scope. I am using a single turn pot tho. Later
on today, I was going to use a 10 turn pc mount pot and see if I can tweak
it a little better.

Just add a fixed resistor in series with the input end of the pot,
till the desired maximum volume corresponds to the pot turned all the
way up.

 

[Goran Tomaš]

Thanks Goran,
I am working on it, I think I need a multiturn pot, because there's a spot
on the single turn pot where the output is very low and then suddenly jumps
loud with lots of noise. I was going to try a 10 turn 10k pot later on
today.

A multiturn pot or trimmer won't do it. Linear pot won't either. You
need logarithmic potentiometer!

With linear pot you'll get exactly what you're getting - nothing,
nothing and then suddenly very loud. Ear responds to volume in
logarithmic manner. Therefore you need logarithmic potentiometer. A
10k log pot should do it...


Regards,
Goran Tomas

 

[Joe]

A multiturn pot or trimmer won't do it. Linear pot won't either. You
need logarithmic potentiometer!

With linear pot you'll get exactly what you're getting - nothing,
nothing and then suddenly very loud. Ear responds to volume in
logarithmic manner. Therefore you need logarithmic potentiometer. A
10k log pot should do it...


Regards,
Goran Tomas

Hi Goran,

I guess an audio taper is close to what you are calling logarithmic. I have
one of those. With all these suggestions, my results are definitely
improving. Thanks for the advice.

Joe

 

[Joe]

Just add a fixed resistor in series with the input end of the pot,
till the desired maximum volume corresponds to the pot turned all the
way up.

Hi John,

Thanks I will try that. The problem is working itself out as I get these
suggestions. I am going to try an audio taper pot also. I think maybe I also
need a better quality speaker.

Joe