In article ,
Phil Allison wrote:
Dave Plowman (Nutcase) wrote:

Pretty basic MOSFET power amp.

I'm used to it thumping at switch on. No speaker relay in this basic
design.


** You posted the same here in December last year.

About switch on thump. This is at switch off, and a fault.

But one of them as taken to thumping at switch off. Not always, which
makes fault finding more tricky.

Most likely cause?


** Old and bad electros are the cause of almost everything.

They are all relatively new, and good quality. Not to say one can't have
failed. There are only 4 electros in this pretty basic design. Two are
across the supply rails, one on the input, and the most likely one a low
ESR type which is part of a RC network across the speaker output.

But the fault hasn't occurred again.


BTW; anyone familiar with the famous Crown DC300A ?

Both channels would emit a long squeal about 10 seconds after switch off
- it sounded much like a party balloon being let down while squeezing
the neck.

There was a note inserted in later service manuals that *no attempt*
should be made to fix the problem, as any possible fix would render the
amp unstable.




.... Phil

--
*HOW DO THEY GET DEER TO CROSS THE ROAD ONLY AT THOSE YELLOW ROAD SIGNS?

Dave Plowman London SW
To e-mail, change noise into sound.

On Fri, 29 Apr 2016 19:17:55 -0700 (PDT), Phil Allison
wrote:

Don Pearce wrote:


BTW; anyone familiar with the famous Crown DC300A ?

Both channels would emit a long squeal about 10 seconds after switch off
- it sounded much like a party balloon being let down while squeezing
the neck.

There was a note inserted in later service manuals that *no attempt*
should be made to fix the problem, as any possible fix would render the
amp unstable.




So they had an instability they couldn't fix, but sales took it off
the engineers and put it in the market anyway? Why am I not
surprised?


** The DC300A is perfectly stable with any load and all frequencies.

When the AC supply is removed, an internal +10V boost rail above the main positive rail quickly dies removing current from the class A driver stage - which pretty much disables the output stage.

When the DC rails to the input op-amp ( a uA739 ) drop below a working level, the squealing sound arrives. Its not particularly loud and does not depends on load conditions. The frequency is a few kHz, falling in pitch until there is almost no charge left in the main filter caps.

I know about it since I was asked to fix the noise by a customer who owned a DC300A. The solution was to fit a speaker relay with switch on delay and fast off.

I only mentioned this as an example of how trying to fix a small problem can lead to creating a much worse one. I suspect trying to fix inbuilt turn on/off transients in power amps are mostly like this.

IOW - " leave good enough alone ".



... Phil

I certainly agree that there is very little point for the end user to
try and fix this. But it is a design flaw and should have been fixed
by the original engineers.

Speaker relay? Yes, that will stop you hearing it, but it doesn't fix
the problem.

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

In article , Don Pearce
wrote:
I certainly agree that there is very little point for the end user to
try and fix this. But it is a design flaw and should have been fixed by
the original engineers.

Speaker relay? Yes, that will stop you hearing it, but it doesn't fix
the problem.

FWIW When designing the 730/732 amps the development *preamp* design at
one point tended to give bursts of about 1MHz about ten secs after being
turned off. It was coming from ICs being used to regulate and stabilise the
rails.

I solved it by not using ICs in the PSU.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

On Sat, 30 Apr 2016 12:44:07 +0100, Jim Lesurf
wrote:

In article , Don Pearce
wrote:
I certainly agree that there is very little point for the end user to
try and fix this. But it is a design flaw and should have been fixed by
the original engineers.

Speaker relay? Yes, that will stop you hearing it, but it doesn't fix
the problem.

FWIW When designing the 730/732 amps the development *preamp* design at
one point tended to give bursts of about 1MHz about ten secs after being
turned off. It was coming from ICs being used to regulate and stabilise the
rails.

I solved it by not using ICs in the PSU.

Jim

Fair enough. Why were you regulating the rails - couldn't get the kind
of CMRR you were after from the preamp?

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

In article , Don Pearce
wrote:

Fair enough. Why were you regulating the rails - couldn't get the kind
of CMRR you were after from the preamp?

So far as I recall, it was purely because I started off assuming I might as
well. At the time the 'conventional wisdom' was to use ICs to regulate
rails, and discrete devices for the amp stages.

But I eventually concluded that making a 'smooth' PSU was simpler and
worked nicely. FWIW The stages had decent rejection, and I added per-stage
rail decoupling anyway. So I ended up with some transistors in the PSU
(essentially capacitance multipliers) and ICs for the gain stages. Simply
found it worked nicely.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

Don Pearce wrote:

Phil Allison


So they had an instability they couldn't fix, but sales took it off
the engineers and put it in the market anyway? Why am I not
surprised?


** The DC300A is perfectly stable with any load and all frequencies.

When the AC supply is removed, an internal +10V boost rail above the
main positive rail quickly dies removing current from the class A
driver stage - which pretty much disables the output stage.

When the DC rails to the input op-amp ( a uA739 ) drop below a working
level, the squealing sound arrives. Its not particularly loud and does
not depends on load conditions. The frequency is a few kHz, falling
in pitch until there is almost no charge left in the main filter caps.

I know about it since I was asked to fix the noise by a customer who
owned a DC300A. The solution was to fit a speaker relay with switch on
delay and fast off.

I only mentioned this as an example of how trying to fix a small problem
can lead to creating a much worse one. I suspect trying to fix inbuilt
turn on/off transients in power amps are mostly like this.

IOW - " leave good enough alone ".



I certainly agree that there is very little point for the end user to
try and fix this. But it is a design flaw and should have been fixed
by the original engineers.


** It was not a problem that needed fixing.

The DC300A was marketed as a laboratory & professional use amplifier - not a home audio one.


Speaker relay? Yes, that will stop you hearing it, but it doesn't fix
the problem.


** Speaker muting relays have been standard features in countless home and professional audio amplifiers for decades. Their main purpose is to stop transients at switch on/off from being heard.

The last Onkyo 7.2 channel receiver I serviced had 9 chunky relays, one for each power amp, one in the AC supply and one in the main AC secondary for standby.

They are very much the solution to the problem.



..... Phil

On Sat, 30 Apr 2016 18:01:30 +0100, Jim Lesurf
wrote:

In article , Don Pearce
wrote:

Fair enough. Why were you regulating the rails - couldn't get the kind
of CMRR you were after from the preamp?

So far as I recall, it was purely because I started off assuming I might as
well. At the time the 'conventional wisdom' was to use ICs to regulate
rails, and discrete devices for the amp stages.

But I eventually concluded that making a 'smooth' PSU was simpler and
worked nicely. FWIW The stages had decent rejection, and I added per-stage
rail decoupling anyway. So I ended up with some transistors in the PSU
(essentially capacitance multipliers) and ICs for the gain stages. Simply
found it worked nicely.

Jim

And back then regulators, as I remember, were incredibly noisy at
their outputs. You could kill that with a series resistor plus big
electrolytic, but that kind of destroyed the regulation. As you say,
for the kind of current drawn by a preamp, smoothing the PSU is really
no problem.

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

In article ,
Phil Allison wrote:
Speaker relay? Yes, that will stop you hearing it, but it doesn't fix
the problem.


** Speaker muting relays have been standard features in countless home
and professional audio amplifiers for decades. Their main purpose is to
stop transients at switch on/off from being heard.

They can also include a circuit to make them drop out if they detect a
large DC offset. Very good way of protecting speakers in event of a fault.

--
*Cleaned by Stevie Wonder, checked by David Blunkett*

Dave Plowman London SW
To e-mail, change noise into sound.

Dave Plowman (News) wrote:



** Speaker muting relays have been standard features in countless home
and professional audio amplifiers for decades. Their main purpose is to
stop transients at switch on/off from being heard.


They can also include a circuit to make them drop out if they detect a
large DC offset. Very good way of protecting speakers in event of a fault.



** Few speaker relay systems provide reliable protection if that happens - nor do they claim to.

The simple fact being that the kind of relays used are only capable of switching *off* DC voltages to a maximum of 30V at rated current. IOW, their current breaking capacity is very limited for DC voltages.

What happens instead is a continuous arc forms between the contacts as they open and persists until the relay is destroyed along with the speaker voice coil as well.

Amplifiers with DC supply rails under 50V may allow the arc to break, but any higher and there is practically no chance.



.... Phil

In article ,
Phil Allison wrote:
They can also include a circuit to make them drop out if they detect a
large DC offset. Very good way of protecting speakers in event of a
fault.



** Few speaker relay systems provide reliable protection if that happens
- nor do they claim to.

I'm talking about those which do.

The simple fact being that the kind of relays used are only capable of
switching *off* DC voltages to a maximum of 30V at rated current. IOW,
their current breaking capacity is very limited for DC voltages.

Then use a relay fit for purpose. Fraction of the cost of a speaker repair.

--
*Why don't sheep shrink when it rains?

Dave Plowman London SW
To e-mail, change noise into sound.