On 8/10/2012 6:53 AM, tony sayer wrote:
In article , Trevor Wilson trevor@SPAM
BLOCKrageaudio.com.au scribeth thus
On 8/7/2012 4:12 AM, Dave Plowman (News) wrote:
Have a circuit from the '70s I wish to try and it uses BC 109/179
complimentary pairs. Is there a better choice these days that is a direct
substitute?


As PA suggested, the BC109 came in a hermetically sealed metal can. That
made it superior to almost any plastic pack device.

Any reason why that should be Trevor?..

Or did you mean at the time the BC 109 came onto the market?..

**Yes and, no. IME, low power, metal can devices demonstrate superior
long term reliability to pre-1980-ish plastic devices. That said, modern
plastic devices appear to provide excellent long term reliability as well.



--
Trevor Wilson www.rageaudio.com.au

"tony sayer"

Or did you mean at the time the BC 109 came onto the market?


** The original BC108 series in TO18 pack preceded plastic equivalents.

The first plastic versions were the BC148 series and BC208 series in SOT25
and TO106 respectively.

The now ubiquitous TO 92 pack came later again.

See:

http://home.hccnet.nl/piet.blaas/gro...bc148b-500.jpg

and

http://www.buy-transistors.com/media...c/bc208b_1.jpg

The TO18 versions, being hermetically sealed, have proved to have longer
service and storage lives.


.... Phil

In article , Trevor Wilson
wrote:


**Yes and, no. IME, low power, metal can devices demonstrate superior
long term reliability to pre-1980-ish plastic devices. That said, modern
plastic devices appear to provide excellent long term reliability as
well.

The key word, though, is 'can' sorry for the pun!. I've known more than
one 'metal can' transistor that were unreliable for various reasons. So as
with types using other packaging it all comes down to how well the
individual devices were made.

In one case Oops, another pun WRT higher power devices I recall a TO3
device that exploded when powered up at low levels and the top hit part of
the wood around the bench so hard that it embossed the maker's name there.
Turned out to be a batch with water vapour inside the can. More common was,
it seems, simply poor contacts so that the device came free of the
metalwork.

Slainte,

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 8/10/2012 6:02 PM, Jim Lesurf wrote:
In article , Trevor Wilson
wrote:


**Yes and, no. IME, low power, metal can devices demonstrate superior
long term reliability to pre-1980-ish plastic devices. That said, modern
plastic devices appear to provide excellent long term reliability as
well.

The key word, though, is 'can' sorry for the pun!. I've known more than
one 'metal can' transistor that were unreliable for various reasons. So as
with types using other packaging it all comes down to how well the
individual devices were made.

In one case Oops, another pun WRT higher power devices I recall a TO3
device that exploded when powered up at low levels and the top hit part of
the wood around the bench so hard that it embossed the maker's name there.
Turned out to be a batch with water vapour inside the can. More common was,
it seems, simply poor contacts so that the device came free of the
metalwork.


**Back in the 1970s, I worked on a lot of Marantz products (as the
Australian service manager). Back then, there were a couple of Japanese
suppliers and one US supplier of output devices for Marantz products.
The US supplier was Motorola and their devices were in the, now
infamous, aluminium TO3 cases. The Japs were supplied in (usually)
copper cases, though some may have been steel. The Motorola devices were
far less reliable than the Japanese ones. In fact, often when the
Motorola devices failed, a tiny hole was punctured in the aluminum case,
as part of the innards was expelled outwards. About that time, I read a
white paper published by RCA where it was claimed that their steel TO3
cases could provide more than 1 million hot/cold cycles without failure,
whereas the aluminium cases used by Motorola could only sustain 100,000
hot/cold cycles. Early plastic packs could only manage around 10,000
cycles. Anecdotal evidence and chatting to a couple of people in the
industry seemed to bear this out. I was informed by one person that RCA
used superior wire bonding techniques which assisted with long term
reliability for their devices (they second-sourced Motorola chips in
some cases). As for plastic packs, the only reciever in the Marantz
line-up that used plastic pack output devices enjoyed a 67% failure rate
within the warranty period. That is contrasted by the fabulous 1070
amplifier (that used copper TO3 outputs) which exhibited an impressive
0.5% failure rate within the warranty period. We won't talk about the
mighty Model 500 power amp (I still own mine). Every single one failed,
several times within the warrnty period. Legend has it that the company
spent US$3 million on the design, development, production and warranty
support on the 300 amplifiers produced. Aluminium cased output devices. Yuk.


--
Trevor Wilson www.rageaudio.com.au

"Trevor Wilson"

**Back in the 1970s, I worked on a lot of Marantz products (as the
Australian service manager). Back then, there were a couple of Japanese
suppliers and one US supplier of output devices for Marantz products. The
US supplier was Motorola and their devices were in the, now infamous,
aluminium TO3 cases.

** Infamous ??

That IS putting it a bit strongly !!

FYI:

1. Motorola chips were never bonded direct to Aluminium ( or steel) - but
instead to a small copper alloy slug that sat in a depression in the base of
the Al pack. Solder like material held the chip to the slug and the slug to
the base metal.

( Steel devices have a rather large copper slug that sits on top of the
thin base. )

2. I must have cut a hundred or more TO3 devices open ( mostly Motorola ) to
see the internal construction and identify fakes.


The Japs were supplied in (usually) copper cases, though some may have
been steel. The Motorola devices were far less reliable than the Japanese
ones.

** For some reason other than the one you are crapping on about.


About that time, I read a white paper published by RCA where it was
claimed that their steel TO3 cases could provide more than 1 million
hot/cold cycles without failure, whereas the aluminium cases used by
Motorola could only sustain 100,000 hot/cold cycles.

** I think you have moved the decimal point one place to the right.

HOWEVER a "hot / cold cycle " is one where the chip goes from room temp to
rated max - so would be a rare event in a domestic hi-fi amp.

( Different story for a linear regulated PSU or pro audio amplifiers )

Early plastic packs could only manage around 10,000 cycles. Anecdotal
evidence and chatting to a couple of people in the industry seemed to bear
this out.

** Gossip is very cheap.

And every plastic power device I have ever seen uses a copper heatsinking
base.


..... Phil

"Don Pearce" wrote in message
...

You can safely ignore the stuff from downunder. Switching transistors
are just as linear as any other kind. What makes them good for
switching is that when you turn them on fully they drop less voltage
across the collector and emitter.

My undergraduate engineering circuits class challenge project was for teams
to build as good of a RIAA preamp as could be built with 2N404A transistors.
The prof forgot to limit how many transistors could be used, so I used two
Darlington pairs per gain block, full loop feedback. There was a RIAA preamp
followed by a line level amp with a gain control in-between. It easily
outperformed just about anything commercial in the day for both noise and
distortion. Note that the 2N404A was a dirt cheap switching transistor.

"Trevor Wilson" wrote in message
...

About that time, I read a
white paper published by RCA where it was claimed that their steel TO3
cases could provide more than 1 million hot/cold cycles without failure,
whereas the aluminium cases used by Motorola could only sustain 100,000
hot/cold cycles.

I remember that, too.

RCA made a ton of money making power transistors for early automotive
electronic ignition systems.