Cracks and pops on audio Op-Amp
Adrian Black
channel not working. It uses a decent design with a switching power
supply, two NEC uPC4574C quad bipolar op amps, a uPC4570C dual op amp
and 4 LM391's to run the output transistors. After looking at the PDF
of the quad op amp, it seem to be similar to the LM837N. Low noise,
high slew, etc.
Testing with my scope show three channels work well but one channels
is filled with noise. Not humming, but crackling and poping.
After poking around with my scope sending a 1khz signal through it, I
find that the noise appears at the output of the first op-amp the
channel hits. The signal then goes to "Line-Out" connectors on the
unit and back into the same IC to a second op-amp.
But I'm at a loss to what could be causing the noise. I desoldered the
connection to the "line-out" thinking it was somethng downstream
causing the noise, but that had no effect.
Could the chip itself be bad? Or would it more likely be one of the
resitors, diodes or capacitors? All of those visually seem ok. I
resoldered all of their connections to make sure no cold joints were
causing trouble.
I thought the power supply might be noisy (+ and - 15v) but all three
other channels including the other "half" of this particular IC work
perfectly. I desoldered the +15v feed into the chip and when looking
at the power, it was totally clean and steady.
When I removed power from the chip, the noise went away. Somehow audio
was going all the way through the amplifier and conncting a speaker
confirmed a nice steady tone. Coneection the power again brought the
gain up and all the noise right back. The noise happens wether a
signal is present or not.
Looking at the input into the inverting input on the op-amps shows an
oscillator...? (it's what is looks like on thte scope and all the
other stages look the same.) The invering input is connected to the
output through a resistor as well as the non inverting input through a
cap and ground through a resistor and cap. It also looks like the
other channel going through the amp has the same capacitors, diode and
resistor as the bad channel.
Anyway, if you guys have any ideas for what I should try, I'de love to
hear them. I was thinking of ordering a LM837N and installing it as I
can't seem to find the NEC part. I just wonder if it's possible the
noise could be coming from the chip itself ...?
Thanks.
--Adrian
Larry Brasfield
Adrian Black (adr...@soliton.net) says...
> My friend recently gave me a 4-channel Kenwood car amplifier with one
> channel not working. It uses a decent design with a switching power
> supply, two NEC uPC4574C quad bipolar op amps, a uPC4570C dual op amp
> and 4 LM391's to run the output transistors. After looking at the PDF
> of the quad op amp, it seem to be similar to the LM837N. Low noise,
> high slew, etc.
>
> Testing with my scope show three channels work well but one channels
> is filled with noise. Not humming, but crackling and poping.
>
> After poking around with my scope sending a 1khz signal through it, I
> find that the noise appears at the output of the first op-amp the
> channel hits. The signal then goes to "Line-Out" connectors on the
> unit and back into the same IC to a second op-amp.
>
> But I'm at a loss to what could be causing the noise. I desoldered the
> connection to the "line-out" thinking it was somethng downstream
> causing the noise, but that had no effect.
>
> Could the chip itself be bad?
Yes. Excess noise of that nature is one of the
ways that static damage can be manifested. Also,
manufacturing flaws can degrade a transistor in
the IC causing that sort of noise.
> Or would it more likely be one of the
> resitors, diodes or capacitors?
Less likely. If it was one of those, applying
some cold spray (gently!) will make it worse or
better, revealing the culprit. If cold spray
has no effect on the noise, I would absolve
those parts.
> All of those visually seem ok. I
> resoldered all of their connections to make sure no cold joints were
> causing trouble.
>
> I thought the power supply might be noisy (+ and - 15v) but all three
> other channels including the other "half" of this particular IC work
> perfectly.
I agree with your logic for absolving the supply.
> I desoldered the +15v feed into the chip and when looking
> at the power, it was totally clean and steady.
>
> When I removed power from the chip, the noise went away.
That is at least consistent with the opamp being
responsible, but it is redundant with your other
data.
> Somehow audio
> was going all the way through the amplifier and conncting a speaker
> confirmed a nice steady tone.
The signal was probably going thru the feedback
elements to get around that opamp.
> Coneection the power again brought the
> gain up and all the noise right back. The noise happens wether a
> signal is present or not.
Yes, generated noise is usually additive.
> Looking at the input into the inverting input on the op-amps shows an
> oscillator...? (it's what is looks like on thte scope and all the
> other stages look the same.)
Stage oscillation could also lead to noise, but
as you've described it, (being present whether
signal is or not), it does not appear to be an
oscillation problem. The oscillation (if that's
what it is) may be another manifestation of a
bad opamp. I would say don't worry about it
unless it survives replacement of the chip.
> The invering input is connected to the
> output through a resistor as well as the non inverting input through a
> cap and ground through a resistor and cap. It also looks like the
> other channel going through the amp has the same capacitors, diode and
> resistor as the bad channel.
It would be very odd for the two channels to
be implemented with different circuit designs.
To me, this mainly confirms that you are not
confused about which stage is which.
> Anyway, if you guys have any ideas for what I should try, I'de love to
> hear them. I was thinking of ordering a LM837N and installing it as I
> can't seem to find the NEC part. I just wonder if it's possible the
> noise could be coming from the chip itself ...?
Yes. I think you've pretty much nailed it.
Replace it after the cold spray (or ice).
--
-Larry Brasfield
(address munged, s/sn/h/ to reply)
Geoff Wood
"Adrian Black" <adr...@soliton.net> wrote in message
news:fc045810.0212...@posting.google.com...
>
> Could the chip itself be bad?
Yes
> Or would it more likely be one of the
> resitors, diodes or capacitors? All of those visually seem ok. I
> resoldered all of their connections to make sure no cold joints were
> causing trouble.
>
Easiest thing is to change the chip. Then if it's still there, look deeper.
geoff
John Woodgate
wrote (in <fc045810.0212...@posting.google.com>) about
'Cracks and pops on audio Op-Amp', on Sat, 7 Dec 2002:
>Anyway, if you guys have any ideas for what I should try, I'de love to
>hear them. I was thinking of ordering a LM837N and installing it as I
>can't seem to find the NEC part. I just wonder if it's possible the
>noise could be coming from the chip itself ...?
It sound like it from your description. Just check that tapping the
resistors round that section of the op-amp doesn't produce any effect.
You might have one that is micro-cracked.
--
Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk
Interested in professional sound reinforcement and distribution? Then go to
http://www.isce.org.uk
PLEASE do NOT copy news posts to me by E-MAIL!
Kevin McMurtrie
adr...@soliton.net (Adrian Black) wrote:
It sounds like moisture creeped into the op-amp. Sometimes such a chip
will magically work again for a few days if you hold a soldering iron to
it for a few seconds. Then it goes all random again.
carltons
McMurtrie <mcmu...@sonic.net> wrote:
Adrian,
Kevin may be on to something here. The noise does sound as if it is "pop
corn" noise which is a surface contamination problem in op amps and other
ic's. His solution is easy to try.
Steve WB4CZR
Adrian Black
Hello everyone. Thank you for your replies on this topic. The
concensus seems to be that the op-amp IC itself is the culplit to my
problems, so I'm going to order a new IC and install it.
I'll post a follow up when I do that. Thanks guys.
--Adrian