Audio Amplifier Distortion Tracing

C

Cursitor Doom

Guest
Hi all,

I'm using an old Advance AF generator (15hz-50khz) in an attempt to
discover what kind of distortion my Technics hifi amplifier is
producing so I can fix it. However, for some reason I'm getting
terrible self-oscillation from the otherwise stable (though
not-very-nice-sounding) amp that can't be stopped once it starts. I'm
just wondering if I should be using some sort of buffer between the
generator and the amp input? At the moment I'm just feeing it direct.
It's the only cause I can think of. I've checked the generator by
itself on a suitable scope and it's absolutely rock solid right across
the range and the attenuators are fine so it can't be the gennie
itself; must be something I'm doing wrong. What doesn't help much is
the gennie doesn't state what its output impedance is. Not sure if
that's relevant or not, though.
Any suggestions?
cheers, cd
 
On 30/08/2014 17:44, Cursitor Doom wrote:
Hi all,

I'm using an old Advance AF generator (15hz-50khz) in an attempt to
discover what kind of distortion my Technics hifi amplifier is
producing so I can fix it. However, for some reason I'm getting
terrible self-oscillation from the otherwise stable (though
not-very-nice-sounding) amp that can't be stopped once it starts. I'm
just wondering if I should be using some sort of buffer between the
generator and the amp input? At the moment I'm just feeing it direct.
It's the only cause I can think of. I've checked the generator by
itself on a suitable scope and it's absolutely rock solid right across
the range and the attenuators are fine so it can't be the gennie
itself; must be something I'm doing wrong. What doesn't help much is
the gennie doesn't state what its output impedance is. Not sure if
that's relevant or not, though.
Any suggestions?
cheers, cd
Click to expand...

Both channels affected?
Stable power rails?
Can you split at the preamp-poweramp stage?
 
On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom
<cd@spamfreezone.net> wrote:

Correction! It's the scope loading on the output that's making the
amplifier unstable. Apologies...
 
In article <dpc40a9j2citosjsiti6mtm48hsbkktj6s@4ax.com>,
cd@spamfreezone.net says...
On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom
cd@spamfreezone.net> wrote:

Correction! It's the scope loading on the output that's making the
amplifier unstable. Apologies...
Click to expand...

Sounds like your problem is due to the fact that
the amp is on the edge of oscillating. This can
cause distortion, obviously.. The mere fact that you
attach something to it is the trigger. Or maybe you don't
have a proper test load on it, a test load should be a
non inductive load the amp is designed for.

Maybe you need one of those resistor and cap in series
across the output to force the impedance for the high frequencies
down to a manageable impedance the amp likes?

Or, you could be having issues with the miller feed back compensation
network, which is to help keep the amp from Os-Kill-lating! Not
to confuse that with ovulating!


Jamie
 
In article <dpc40a9j2citosjsiti6mtm48hsbkktj6s@4ax.com>,
Cursitor Doom <cd@spamfreezone.net> wrote:

On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom
cd@spamfreezone.net> wrote:

Correction! It's the scope loading on the output that's making the
amplifier unstable. Apologies...
Click to expand...

Such is life. My amplifiers oscillate but my oscillators don't.

You can measure your generator's output impedance with a variable
resistor. Measure open circuit voltage. Then connect the variable
resistor and adjust for one half the voltage. Measure the resistor,
which should be the same as your output impedance.

Your distortion might be related to the instability. An oscilloscope
should have negligible effect on the extremely low output impedance of
an audio power amplifier. The amplifier might be rated for an 8 Ohm
speaker, but must have much lower output impedance to keep Voltage
constant for a varying load impedance over the audio frequency range.

The first things I would check are the electrolytic capacitors.
Assuming it is an older unit, several of the capacitors might have
developed high equivalent series resistance (ESR). That includes power
supply filters, audio bypass capacitors as well as a capacitor in the
output stage that blocks DC from getting to your speakers.

An ESR meter is handy to have if you do much repair work. I have the
AnaTek Blue ESR meter, built from a kit.

Fred
 
On Sunday, August 31, 2014 6:27:44 AM UTC+10, Cursitor Doom wrote:
On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom

cd@spamfreezone.net> wrote:



Correction! It's the scope loading on the output that's making the

amplifier unstable.
Click to expand...

** Really?

Until you connect the scope you are unable to observe the oscillations.

BTW what sort of frequency are you seeing?


.... Phil
 
On Sat, 30 Aug 2014 18:26:09 -0400, Fred McKenzie <fmmck@aol.com>
wrote:

In article <dpc40a9j2citosjsiti6mtm48hsbkktj6s@4ax.com>,
Cursitor Doom <cd@spamfreezone.net> wrote:

On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom
cd@spamfreezone.net> wrote:

Correction! It's the scope loading on the output that's making the
amplifier unstable. Apologies...

Such is life. My amplifiers oscillate but my oscillators don't.

You can measure your generator's output impedance with a variable
resistor. Measure open circuit voltage. Then connect the variable
resistor and adjust for one half the voltage. Measure the resistor,
which should be the same as your output impedance.

Your distortion might be related to the instability. An oscilloscope
should have negligible effect on the extremely low output impedance of
an audio power amplifier. The amplifier might be rated for an 8 Ohm
speaker, but must have much lower output impedance to keep Voltage
constant for a varying load impedance over the audio frequency range.

The first things I would check are the electrolytic capacitors.
Assuming it is an older unit, several of the capacitors might have
developed high equivalent series resistance (ESR). That includes power
supply filters, audio bypass capacitors as well as a capacitor in the
output stage that blocks DC from getting to your speakers.

An ESR meter is handy to have if you do much repair work. I have the
AnaTek Blue ESR meter, built from a kit.

Fred
Click to expand...

Thanks for the Thevenin reminder!
I'm not using a resistive probe to measure so additional loading is
scope's intrinsic 1Meg || 20pf.
Still can't believe that little across 8 ohm can cause such mayhem,
though. :-/
 
"Cursitor Doom" wrote in message
news:khl40at82qra28pj094ue6dmejclafbqv4@4ax.com...

I'm not using a resistive probe to measure so additional loading is
scope's intrinsic 1Meg || 20pf.
Still can't believe that little across 8 ohm can cause such mayhem,
though. :-/
Click to expand...

Ground problems?

You need to pull everything apart, and test items two at a time.
 
On 31/08/2014 2:44 AM, Cursitor Doom wrote:
Hi all,

I'm using an old Advance AF generator (15hz-50khz) in an attempt to
discover what kind of distortion my Technics hifi amplifier is
producing so I can fix it. However, for some reason I'm getting
terrible self-oscillation from the otherwise stable (though
not-very-nice-sounding) amp that can't be stopped once it starts. I'm
just wondering if I should be using some sort of buffer between the
generator and the amp input? At the moment I'm just feeing it direct.
It's the only cause I can think of. I've checked the generator by
itself on a suitable scope and it's absolutely rock solid right across
the range and the attenuators are fine so it can't be the gennie
itself; must be something I'm doing wrong. What doesn't help much is
the gennie doesn't state what its output impedance is. Not sure if
that's relevant or not, though.
Any suggestions?
cheers, cd
Click to expand...

**It's highly likely that you are doing something seriously wrong.
Domestic amplifiers, from major manufacturers are usually VERY stable.
Here's what you should do:

0) Check to see if the amp uses bridged outputs. If so, then only test
one channel at a time and ensure that your 'scope is ungrounded, or you
are using a differential input (if available).
1) Connect your amp to a pair of isolated dummy loads.
2) Switch on. Obsrve if there is an oscillation. If no oscillation, move
to step 3.
3) Connect your sig gen and test as normal.
4) Report.




--
Trevor Wilson www.rageaudio.com.au
 
On 31/08/2014 2:44 AM, Cursitor Doom wrote:
Hi all,

I'm using an old Advance AF generator (15hz-50khz) in an attempt to
discover what kind of distortion my Technics hifi amplifier is
producing so I can fix it. However, for some reason I'm getting
terrible self-oscillation from the otherwise stable (though
not-very-nice-sounding) amp that can't be stopped once it starts. I'm
just wondering if I should be using some sort of buffer between the
generator and the amp input? At the moment I'm just feeing it direct.
It's the only cause I can think of. I've checked the generator by
itself on a suitable scope and it's absolutely rock solid right across
the range and the attenuators are fine so it can't be the gennie
itself; must be something I'm doing wrong. What doesn't help much is
the gennie doesn't state what its output impedance is. Not sure if
that's relevant or not, though.
Any suggestions?
cheers, cd
Click to expand...

**Oh yeah, the model number of the amp might assist.

--
Trevor Wilson www.rageaudio.com.au
 
On Sunday, August 31, 2014 9:06:48 PM UTC+10, Cursitor Doom wrote:
On Sat, 30 Aug 2014 19:47:44 -0700 (PDT), Phil Allison

pallison49@gmail.com> wrote:



On Sunday, August 31, 2014 6:27:44 AM UTC+10, Cursitor Doom wrote:

On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom



cd@spamfreezone.net> wrote:





Correction! It's the scope loading on the output that's making the



amplifier unstable.





** Really?



Until you connect the scope you are unable to observe the oscillations.



Correct.


BTW what sort of frequency are you seeing?



1khz; the test frequency from the gen.
Click to expand...

** Ask a silly question ...



.... Phil
 
On Sun, 31 Aug 2014 17:01:45 +1000, Trevor Wilson
<trevor@SPAMBLOCKrageaudio.com.au> wrote:


>**Oh yeah, the model number of the amp might assist.

SA EH750 FWIW. Don't much like it anyway - too modern.
Having slept on it, I'll make up another 'scope test lead from
screened coax; the current bodge just uses speaker twin core which
could itself be causing the feedback.
 
On Sat, 30 Aug 2014 17:04:56 -0700, "William Sommerwerck"
<grizzledgeezer@comcast.net> wrote:

"Cursitor Doom" wrote in message
news:khl40at82qra28pj094ue6dmejclafbqv4@4ax.com...

I'm not using a resistive probe to measure so additional loading is
scope's intrinsic 1Meg || 20pf.
Still can't believe that little across 8 ohm can cause such mayhem,
though. :-/

Ground problems?
Click to expand...

Might be! See my reply to Trevor.
 
On Sat, 30 Aug 2014 19:47:44 -0700 (PDT), Phil Allison
<pallison49@gmail.com> wrote:

On Sunday, August 31, 2014 6:27:44 AM UTC+10, Cursitor Doom wrote:
On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom

cd@spamfreezone.net> wrote:



Correction! It's the scope loading on the output that's making the

amplifier unstable.


** Really?

Until you connect the scope you are unable to observe the oscillations.
Click to expand...

Correct.

BTW what sort of frequency are you seeing?
Click to expand...

1khz; the test frequency from the gen.


... Phil
Click to expand...
 
On Sat, 30 Aug 2014 21:27:00 +0100, N_Cook <diverse@tcp.co.uk> wrote:


>Both channels affected?

Cant tell yet...

>Stable power rails?

Good point.

>Can you split at the preamp-poweramp stage?

No. It's combined. But there's a graphic equaliser inline that could
also be to blame. The distortion I'm getting is identical to a poorly
set-up GE.
This dumb modern integration of everything makes it so tough to
seperate possible component causes....
 
In article <gc060a9hcf0mrdc5jb1emg9d18apj2hu7o@4ax.com>,
cd@spamfreezone.net says...
On Sat, 30 Aug 2014 19:47:44 -0700 (PDT), Phil Allison
pallison49@gmail.com> wrote:

On Sunday, August 31, 2014 6:27:44 AM UTC+10, Cursitor Doom wrote:
On Sat, 30 Aug 2014 18:44:50 +0200, Cursitor Doom

cd@spamfreezone.net> wrote:



Correction! It's the scope loading on the output that's making the

amplifier unstable.


** Really?

Until you connect the scope you are unable to observe the oscillations.

Correct.


BTW what sort of frequency are you seeing?

1khz; the test frequency from the gen.
Click to expand...

Nothing like a troll..


Jamie
 
On 31/08/2014 9:04 PM, Cursitor Doom wrote:
On Sun, 31 Aug 2014 17:01:45 +1000, Trevor Wilson
trevor@SPAMBLOCKrageaudio.com.au> wrote:


**Oh yeah, the model number of the amp might assist.

SA EH750 FWIW. Don't much like it anyway - too modern.
Having slept on it, I'll make up another 'scope test lead from
screened coax; the current bodge just uses speaker twin core which
could itself be causing the feedback.
Click to expand...

**OK. The amplifier is not bridged, but it is bi-amped. That said, it
would seem that you have no clue about such things. Take it to a
professional.

[shakes head] Buy a damned 'scope probe. You can buy them for less than
10 Bucks out of China and they'll work fine for your purposes.

--
Trevor Wilson www.rageaudio.com.au
 
>"..I don't really like how they design but there audio isn't all that bad."

Dammit.
 
>"SA EH750 "

That uses a RSN311W64 six channel IC with all inverting inputs. Grounds are pretty important.

Tell you one thing I noticed Technics does is to save copper by using the metal chassis as a ground. First thing I would do is to tighten all the screws that hold down the circuit board. Loosen them a bit first, and see how tight they are. some will not be critical but some will.

I don't really like how they design but there audio isn't all that bad. I'll pass though because those SVI series chips run up into the hundreds to buy.

At any rate, if this is affecting all channels it is "global" so grounds are it. That chip looks like it is getting about five supplies, one set called for at like +/-59 volts.

Another thing about amps designed like this, when you hook a scope to the speaker you also use the speaker ground as a ground. If you use the chassis you might incur oscillation with a 1X probe. not that it happens all the time.

This is all assuming you ran ESR on a bunch of the caps. Look for those little damn 4.7uFs.

If this problem is as stated and I haven't fcked up reading it, I would think about taing a cap from an input to a local ground, inputting the signal to another channel and see if the bypassed channel produces ditortion.

For example, connect a ten uF between pins 26 and 24. Run the same signal through it and see what you get at pin 2, which is where the amp fed from pin 25 feeds.There is only one more ground on that chip at pin 10. try bypassing to that. If you get any oscillation at pin 2, simply confirm clean supplies, and if it has clean supplies it is the chip itself. We took the ground local to make sure of that. The ONLY other thing can be unbypassed Vxxes. Looks like about five of them or so at first glance, but I just got the print five minutes ago.

Let me ask, are you running this into speakers or unloaded ? If it does it unloaded and you are checkiong it unoaded you can just short instead of use the capacitor. With a load no because if the chip has any DC offset it oculd at most be fried, or at best throw off your results. We need clean results.

These power rails also ned to be checked while the distortion is being produced. Don't just scope them with nothing going through there. Doesn't show shit.

Let us know what you find.
 
On Mon, 01 Sep 2014 08:43:35 +1000, Trevor Wilson
<trevor@SPAMBLOCKrageaudio.com.au> wrote:

On 31/08/2014 9:04 PM, Cursitor Doom wrote:
On Sun, 31 Aug 2014 17:01:45 +1000, Trevor Wilson
trevor@SPAMBLOCKrageaudio.com.au> wrote:


**Oh yeah, the model number of the amp might assist.

SA EH750 FWIW. Don't much like it anyway - too modern.
Having slept on it, I'll make up another 'scope test lead from
screened coax; the current bodge just uses speaker twin core which
could itself be causing the feedback.


**OK. The amplifier is not bridged, but it is bi-amped. That said, it
would seem that you have no clue about such things. Take it to a
professional.

[shakes head] Buy a damned 'scope probe. You can buy them for less than
10 Bucks out of China and they'll work fine for your purposes.
Click to expand...

Already have several high quality probes - and 5 scopes! - but none
with suitable tips for this purpose.


---
Diese E-Mail ist frei von Viren und Malware, denn der avast! Antivirus Schutz ist aktiv.
http://www.avast.com
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top