Spot the Design Error

[Phil Allison]

Hi to all my pals,

a certain "design genius" here called John thinks any clever engineer can quickly decide if a design is good or otherwise by assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_502576.jpg

Nothing very unusual about it, dif pair input with level shifters, cascode class A stage with bootstrapped load and complementary Darlington output using paralleled BJTs. All the semis are well rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp misbehave terribly. Another bad flaw involves the thermal switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.




..... Phil

 

[Phil Allison]

Robert Baer wrote:

-------------------

Been ages..could be very wrong, but methinks that problem lies with
the TR110, TR115 circuitry.

** That pair act as Vbe multiplier and buffer to bias the Darlington stage.

TR115 is mounted on the heatsink.

No issues there.

And if so, then the "published specs" did not mention crossover
distortion.

** THD is low ( between 0.01% and 0.002% )and vanishes into noise at the sub 1 watt level.



...... Phil

 

[Robert Baer]

Phil Allison wrote:

Hi to all my pals,

a certain "design genius" here called John thinks any clever engineer can quickly decide if a design is good or otherwise by assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_502576.jpg

Nothing very unusual about it, dif pair input with level shifters, cascode class A stage with bootstrapped load and complementary Darlington output using paralleled BJTs. All the semis are well rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp misbehave terribly. Another bad flaw involves the thermal switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.




.... Phil

Been ages..could be very wrong, but methinks that problem lies with
the TR110, TR115 circuitry.
And if so, then the "published specs" did not mention crossover
distortion.

 

[Bill Sloman]

On Tuesday, October 22, 2019 at 12:00:52 PM UTC+11, Phil Allison wrote:

Hi to all my pals,

a certain "design genius" here called John thinks any clever engineer can quickly decide if a design is good or otherwise by assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_502576.jpg

Nothing very unusual about it, dif pair input with level shifters, cascode class A stage with bootstrapped load and complementary Darlington output using paralleled BJTs. All the semis are well rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp misbehave terribly. Another bad flaw involves the thermal switches.

Who here can spot one or both first ?

Use the schematic, don't go searching web forums.

R155 and R156 seem to be there to monitor the residual current at low voltage swings for class B operation.

There are a whole lot of other resistors tied together with them, feeding the bases of TR113 and 114, which would presumably be sensing the voltage developed.

It looks like an incoherent mess, as if somebody wanted to use the same resistors to drive a current limiting system as well.

--
Bill Sloman, Sydney

 

[Phil Allison]

jurb...@gmail.com wrote:


-------------------------

I guess this is not your cup of tea. Or moose piss.

** ROTFL !!

A poster here used the handle "Moosefet".

Dunno if he is still about, with a new name.




..... Phil

 

I guess this is not your cup of tea. Or moose piss.

No, those resistors provide what is call dual slope current limiting. The current is monitored but the point at which it activates changes with the instantaneous voltage level of the output.

What this does is effectively detect a short at lower voltage yet still allows larger currents to flow at higher output levels.

I am not going on to someone who is my superior in the field. Just run through a few keywords, dual slope, current, audio, limiter. It's out there.

 

[Phil Allison]

DecadentLinux...@decadence.org wrote:

------------------------------------

Hi to all my pals,

a certain "design genius" here called John thinks any clever
engineer can quickly decide if a design is good or otherwise by
assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in
large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_5025
76.jpg

Nothing very unusual about it, dif pair input with level shifters,
cascode class A stage with bootstrapped load and complementary
Darlington output using paralleled BJTs. All the semis are well
rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.



Maybe the heat sinks are supposed to be electrically isolated from
each other and they are not on the chassis?

** The TO3s area all on a large finned heatsink.

The TO220s have individual ones on the PCB.

The design flaws are discoverable from examining the schematic.

" However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches. "


Obviously, the misbehavior is not evident all the time or Yamaha would never have sold any.



..... Phil

 

[bitrex]

On 10/21/19 9:00 PM, Phil Allison wrote:

Hi to all my pals,

a certain "design genius" here called John thinks any clever engineer can quickly decide if a design is good or otherwise by assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_502576.jpg

Nothing very unusual about it, dif pair input with level shifters, cascode class A stage with bootstrapped load and complementary Darlington output using paralleled BJTs. All the semis are well rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp misbehave terribly. Another bad flaw involves the thermal switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.




.... Phil

C125 is too small

 

Phil Allison <pallison49@gmail.com> wrote in
news:84ab086e-2457-4458-88a9-a4acaae54efc@googlegroups.com:

Hi to all my pals,

a certain "design genius" here called John thinks any clever
engineer can quickly decide if a design is good or otherwise by
assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in
large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_5025
76.jpg

Nothing very unusual about it, dif pair input with level shifters,
cascode class A stage with bootstrapped load and complementary
Darlington output using paralleled BJTs. All the semis are well
rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.




.... Phil

Maybe the heat sinks are supposed to be electrically isolated from
each other and they are not on the chassis?

That is a guess.

 

[bitrex]

On 10/22/19 1:11 AM, Phil Allison wrote:

bitrex wrote:

------------

C125 is too small


** No issue there, the output stage is quite stable.

You need to know about good power amp design in some detail.

The thermal switch issue is pretty glaring though.


.... Phil

"misbehave terribly" is a bit broad and subjective but too much detail
on the particular symptoms gives the game away I suppose

 

[Phil Allison]

bitrex wrote:

------------

C125 is too small

** No issue there, the output stage is quite stable.

You need to know about good power amp design in some detail.

The thermal switch issue is pretty glaring though.


..... Phil

 

[Phil Allison]

bitrex wrote:

---------------

Second guess is I think something's maybe not right with TR111, TR112,
D110, D111 usually the base of the driver transistor is connected to the
junction of the bias transistor network and temperature compensation diode.

** TR110 is the bias transistor with TR115 the temp sensor.

FYI:

TR115 is a TO126 device, so easy to bolt down.

TR110 is TO92 type.


...... Phil

 

[John Robertson]

On 2019/10/21 10:11 p.m., Phil Allison wrote:

bitrex wrote:

------------

C125 is too small


** No issue there, the output stage is quite stable.

You need to know about good power amp design in some detail.

The thermal switch issue is pretty glaring though.


.... Phil

I'd think that C117 (and possibly C116) should be more like 25V instead
of 16V.

I'll assume that the "E" on the input ground is physically tied to the
"E" (common) on the other side of the amp...

John :-#)#

 

[Phil Allison]

bitrex wrote:

-------------

Hi to all my pals,

a certain "design genius" here called John thinks any clever
engineer can quickly decide if a design is good or otherwise by
assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in
large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_5025
76.jpg

Nothing very unusual about it, dif pair input with level shifters,
cascode class A stage with bootstrapped load and complementary
Darlington output using paralleled BJTs. All the semis are well
rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.



Maybe the heat sinks are supposed to be electrically isolated from
each other and they are not on the chassis?



** The TO3s area all on a large finned heatsink.

The TO220s have individual ones on the PCB.

The design flaws are discoverable from examining the schematic.

" However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches. "


Obviously, the misbehavior is not evident all the time or Yamaha would never have sold any.

------------------------------------------------------

It's something that they could have not had happen without spending any
(or very little more money), yeah?

** Yep.

That is to say it's surely a design flaw vs a cost-cutting measure

** Cost a few cents in parts fix.

that they figured wouldn't bother with because it didn't happen too often.

** Bet Yamaha's designer had no idea it could happen at all.



..... Phil

 

[Phil Allison]

bitrex wrote:

"misbehave terribly" is a bit broad and subjective but too much detail
on the particular symptoms gives the game away I suppose

** It would.

If nobody gets close, in a couple of days I will post the symptoms.

But you can nail the thermal switch one, right now.




...... Phil

 

[bitrex]

On 10/22/19 1:08 AM, Phil Allison wrote:

DecadentLinux...@decadence.org wrote:

------------------------------------

Hi to all my pals,

a certain "design genius" here called John thinks any clever
engineer can quickly decide if a design is good or otherwise by
assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in
large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_5025
76.jpg

Nothing very unusual about it, dif pair input with level shifters,
cascode class A stage with bootstrapped load and complementary
Darlington output using paralleled BJTs. All the semis are well
rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.



Maybe the heat sinks are supposed to be electrically isolated from
each other and they are not on the chassis?



** The TO3s area all on a large finned heatsink.

The TO220s have individual ones on the PCB.

The design flaws are discoverable from examining the schematic.

" However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches. "


Obviously, the misbehavior is not evident all the time or Yamaha would never have sold any.



.... Phil

It's something that they could have not had happen without spending any
(or very little more money), yeah? That is to say it's surely a design
flaw vs a cost-cutting measure that they figured wouldn't bother with
because it didn't happen too often.

Equipment is sold with no protection from plugging an AC adapter "wall
wart" into a DC jack all the time but that can be a "design flaw" or
cost-cutting measure depending on how you look at it, the mfgr probably
does save 50 cents and also fuck you dummy learn to read the labeling.

 

[bitrex]

On 10/22/19 1:11 AM, Phil Allison wrote:

bitrex wrote:

------------

C125 is too small


** No issue there, the output stage is quite stable.

You need to know about good power amp design in some detail.

The thermal switch issue is pretty glaring though.


.... Phil

Second guess is I think something's maybe not right with TR111, TR112,
D110, D111 usually the base of the driver transistor is connected to the
junction of the bias transistor network and temperature compensation diode.

 

[bitrex]

On 10/22/19 2:24 AM, Phil Allison wrote:

bitrex wrote:

-------------


Hi to all my pals,

a certain "design genius" here called John thinks any clever
engineer can quickly decide if a design is good or otherwise by
assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in
large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_5025
76.jpg

Nothing very unusual about it, dif pair input with level shifters,
cascode class A stage with bootstrapped load and complementary
Darlington output using paralleled BJTs. All the semis are well
rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.



Maybe the heat sinks are supposed to be electrically isolated from
each other and they are not on the chassis?



** The TO3s area all on a large finned heatsink.

The TO220s have individual ones on the PCB.

The design flaws are discoverable from examining the schematic.

" However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches. "


Obviously, the misbehavior is not evident all the time or Yamaha would never have sold any.



------------------------------------------------------


It's something that they could have not had happen without spending any
(or very little more money), yeah?


** Yep.


That is to say it's surely a design flaw vs a cost-cutting measure


** Cost a few cents in parts fix.


that they figured wouldn't bother with because it didn't happen too often.


** Bet Yamaha's designer had no idea it could happen at all.



.... Phil

Was this amp designed to pair with the NS-10s? The joke about the NS-10s
is that every studio had a pair because they sounded so unremarkable
that if a mix sounded good on them it would sound good on anything.

 

[bitrex]

On 10/22/19 2:24 AM, Phil Allison wrote:

bitrex wrote:

-------------


Hi to all my pals,

a certain "design genius" here called John thinks any clever
engineer can quickly decide if a design is good or otherwise by
assessing only a schematic.

I say that is bollocks, maybe a simple test will make my point.

--------------------------------------------------------------

This schem is for the famous Yamaha P2200 power amplifier sold in
large numbers throughout the 1980s.

http://bilder.hifi-forum.de/max/588660/yamaha-p-2200-schematic_5025
76.jpg

Nothing very unusual about it, dif pair input with level shifters,
cascode class A stage with bootstrapped load and complementary
Darlington output using paralleled BJTs. All the semis are well
rated for their respective jobs.

Plus the amp meets its published specs easily.

However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches.

Who here can spot one or both first ?

Use the schem, don't go searching web forums.



Maybe the heat sinks are supposed to be electrically isolated from
each other and they are not on the chassis?



** The TO3s area all on a large finned heatsink.

The TO220s have individual ones on the PCB.

The design flaws are discoverable from examining the schematic.

" However, within the power stage is a flaw that makes the amp
misbehave terribly. Another bad flaw involves the thermal
switches. "


Obviously, the misbehavior is not evident all the time or Yamaha would never have sold any.



------------------------------------------------------


It's something that they could have not had happen without spending any
(or very little more money), yeah?


** Yep.


That is to say it's surely a design flaw vs a cost-cutting measure


** Cost a few cents in parts fix.


that they figured wouldn't bother with because it didn't happen too often.


** Bet Yamaha's designer had no idea it could happen at all.



.... Phil

 

[Phil Allison]

John Robertson wrote:

---------------------

I'd think that C117 (and possibly C116) should be more like 25V instead
of 16V.

** Well spotted - but that is a typo in the schematic.

The boost rail is 21V DC so I'm sure the caps fitted to real amps were 25V types.



..... Phil