On 27/10/2016 22:49, Trevor Wilson wrote:
On 28/10/2016 1:18 AM, N_Cook wrote:
With the built in temp comp diodes, 1 silicon in the N type and 5
Schotky in the P and mirror-reversed pinning .
Owner put a short across one speaker line, magic smoke but other channel
is normal. Internal emitter resistor .22R of the p type is now not .22R.
I intend removing both , of the failed channel , to check out of
circuit, but anyone jumpered in an external .2R in these circumstances
and no further problems from cracked die etc as this R is part of the
die?
The datasheet I think , in fractured English, says the thermal safe
operating over-current zone, for the built in .22R is deliberately
lower than the Darlington itself. Anyone ever disentangled the "English"
referring to deliberately? destroying the internal R so as to use an
external emitter R,or is it just referring to doubling-up ?
datasheet
http://pdf1.alldatasheet.com/datashe.../datasheet.pdf



**I did once in a client's amp, to see what would happen. It worked
fine, but I decided to fit new devices to be certain. These Sanken
things are horrible. Sanken has a nasty track record of releasing
devices, then ceasing production after a few years. After which,
substitutes can be either difficult or impossible to obtain. Give me a
standard 3 terminal transistor every day. I'm presently working on a
1975 vintage Accuphase, which will accept modern replacement devices
quite nicely. In 10 years, those Sanken things will be unavailable and
the units that they're fitted to will be boat anchors. I advise anyone
who will listen to avoid products that use Sanken devices for that reason.


After removing , it became obvious low ohmic C-E, so bang goes that
idea. Perhaps 120V,12A standard pDarlington, with external .2R and 5 SMD
Shottky diodes RTV'd on the top, rather than expensive old-stock SAP15.
Cambridge A5 amp, made in DRC = The Congo ?