[This followup was posted to sci.electronics.repair and a copy was sent
to the cited author.]

In article ,
says...

Excellent pictures, thanks. I believe the actual marking code is DF;
the R indicates the gain and the G is a prouction lot or date code.

My favorite source for arcane marking info: http://www.s-manuals.com

Which leads to: http://www.s-manuals.com/smd/df

And the datasheet: (watych out for line breaks)
http://www.s-manuals.com/pdf/datashe...C_2sd1768s%2C_
2sd1863%2C_2sd1898_rohm.pdf
IF you are in the USA, they are available from Digi-Key
(ww.digikey.com) for $.53 each. (Choose USPS First Class Mail for
lowest shipping cost.)

As a general rule, I dissaprove of replacing parts because they 'look
bad'. A few minutes making measurements with a DVM can give a wealth
of information.

PlainBill

THanks for the info. I had also done some research on some other sites
and find that there are multiple possibilities for this device marked
with a DF.

Sanyo NPN 2SD1623
Philips PNP BF621
Siemens PNP BNF21
Rohm NPN 2SD1898 **

** The part that you located.

I agree that this part should not be replaced if it is working. I had
done some preliminary checking with the meter and it did not seem to
respond with any forward biased junction readings and so I needed to get
information as to whether this was some type of FET or a blown BJT
transistor.

When I get a power cord setup to be able fire up the board I will do
some checking. I may also need to rig up a temoprary heatsink for a
large diode bridge that gets screwed to part of the case that had to be
removed to get at the board. Initial tests can probably avoid that if I
just put small loads on the output rails for testing,

--

Michael Karas
Carousel Design Solutions
http://www.carousel-design.com