
Technics SE-A3
For
so long I have been trying to find the time to update this SE-A3 page
although my SE-A5 and
SE-A3MK2 pages are all done and
thank you to everyone who has contributed to and commented on those pages!
Mainly the reason I never get around to this SE-A3 page is because my
SE-A3 is stuck in the UK and I am not! Sometime I'll get it over to
Florida and I can finish up this page! Until I do, the following
paragraphs answer the most often asked question I get about the SE-A3, so
I hope that this helps some of you out. If anyone would like to
contribute any further comments or photos,
please let me know, it
would be good to get this page in shape! Also, if anyone has an
SE-A3 that needs fixing,
please let me know as I would
like to do another so I can photograph the process.
Where Can I Get Technics DLPT OD503A-P/Q Transistors?
OK, so before going any further, what exactly is a Technics OD503A-P Dual Linear Power Transistor?
Its it A: A funky six legged thing that the Technics
Marketers thought would Sex-Up the SE-A3 and blow every audiophiles
mind in desperation to have some?
Is it B: A monumental design failure that caused Technics
great embarrassment because so many SE-A3's were returned under warranty?
Is it C: Just two supposedly matched no name 230 Volt 15
Amp 180 Watt PNP and NPN bipolar power transistors stuck together in a
plastic cap?
Is it D: The single most common point of failure in the
Technics SE-A3 design?
Well,
the fact is that the Technics DLPT OD503A-P power transistor is All
of the above! I even heard a story once from a guy that was a
soldier in the Danish army who was posted to Greenland. They had a
radio station there that was setup in the early 80's and in true Western
European style, the radio station managers ordered the biggest and very
best, sexiest rack of floor to ceiling power amps they could! Yes, a
whole stack of Technics SE-A3's! Two weeks after install, they were
all busted! All of them, every single one... One can only
imaging the tears and heartbreak that must have rippled through the Technics world... I heard that Technics came to some arrangement
with the radio station due to the immense cost of shipping the things back
to Japan. Some of them are probably still stacked up there, who
could bring themselves to trash them? And the same story has been the
fate of just about every Technics SE-A3 I have ever seen. In short,
there are two kinds of Technics SE-A3's. First, the ones that have had
blown DLPT's, and second, the ones that will, the latter category being in
a very small minority. You see, the problem was that back in the
early 1980's, bi-polar power transistor production was in the stone age
compared with today. Those matched pairs were just not matched well
enough. A blow out in one, causes a ripple effect all through the
amp ending in terminal failure until the whole set can be replaced and the
little yellow protection lamp is the only one to get any glory.
So at the end of the day, you don't want any more Technics DLPT OD503A-P
power transistors, what you need is a more modern alternative that will
provide greater reliability and a smoother frequency response with which
to rebuild the entire amp and bring the performance up to the full
Electric-Sex level implied by the original intent of the design.
I
have also heard of some SE-A3's being fitted with OD503A-Q devices and
there may be even more revisions, but of course, the only way I hear about
them is because they are blown...
Modern Toshiba Replacement Transistors
OK,
so now we know all this, finding replacements is as easy as picking up any
Toshiba power transistor product
list, and looking for what we need. Ah, yes, right there on page
53, the Toshiba 2SA1987
and the Toshiba
2SC5359 will fit the bill because they are 230 volt, 15 Amp, 180
Watts, complementary to each other and intended for audio power amp
applications up to 100 Watts each! The Technics SE-A3 design uses
doubled up pairs to make the 200 Watt output of the design. In fact
it is that doubling up that is the cause of most of the problems where the
transistors working in parallel are not properly matched between different
OD503A-P devices. Most any pair
of these modern Toshiba transistors will be closely enough matched for the
SE-A3. Just make sure that they are real Toshiba products with the
proper markings. It's best to replace all sixteen of the transistors
and get them all from the same vendor at the same time. Do not mix &
match! Now just search for them on the net and abracadabra,
I found some right here in about 2 seconds! If that link is down,
just search for another. Just incase you don't know, the
Toshiba Semiconductor Company is one of the premier electronics
component vendors for Hi-Fi equipment and I defy anyone to look in their
stereo gear and declare that they have no Toshiba components in there!
These Toshiba transistors will serve well in the Technics SE-A3 and bring
all those old broken units back to new life for many more years to come!
So
the Toshiba 2SC5359
is our NPN transistor with the pins in the order, E, C, B and the
Toshiba 2SA1987 is
our PNP device with pins also in the order, E, C, B. The two
transistors in the OD503A-P
are mounted with the NPN on the left and the PNP on the right. This
can be seen by comparing the
Power Amp Layout to the left and the
Power Amp Circuit to the
right from the
service manual. You can see in the
Power Amp Circuit that pin
G of
the top connector goes to the base of an NPN transistor in both Q127
and Q128. Then referring to the
Power Amp Layout, it can be
seen that the track from pin G on the left connector leads to the base of the transistor on the
left hand side (very left pin) on the OD503A-P on both Q127 and Q128 when viewed from the top of
the board. So its
2SC5359 on the left and
2SA1987 on the
right when comparing to an OD503A-P. Now there is a slight problem
as you can see
from the pin layout on the OD503A-P. The NPN transistor on the left
is actually a complete mirror image of a normal power transistor and of
the right hand side of the
OD503A-P. i.e. the pins are in the order B, C,
E rather than E, C, B. So this means, that when each of the the Toshiba 2SC5359's
are mounted, the 1st and 3rd pins must be swapped around rather than
soldered directly to the board like the
2SA1987 on the
right side. Probably the neatest way to so this switcharoo is to
break the tracks on the under side of the board to the Base and Emitter of
the left side transistor and solder some wires on to cross them over.
Remember though that the connection needs to be good and able to carry 15
Amps, so thin lead wire will not do! If anyone else has done that
and can send me a picture to put
here, I would appreciate it.
Click here to see the SE-A3 Service Manual.
Thermal Runaway
At
the end of the day, under the covers of the SE-A3 as described above are
bi-polar power transistors. They are what they are and they are NOT
MOS-FETS because those just weren't invented back then. You need an
SE-A100 for that! Bi-polar
power transistors have a negative temperature coefficient of resistance
which means that as the temperature increases,
resistance reduces and more current flows leading to more heat and so on.
There are of course temperature detection and biasing arrangements for
this case in the design, but because the power transistors are doubled up
in the SE-A3, if one of the pair gets a bit too hot and approaches this
state, thermal runaway occurs in the device fast, and it starts to
hog current that was intended for its parallel mate and Pow!
This is usual the fate of the Technics DLPT OD503A-P transistors, mostly
because of poor matching. Another thing that can cause this problem
is the way the transistors are mounted. As indicated on the diagram
to the left, the power transistors are mounted onto the heat sinks with
insulating mica plates, but what it does not indicate is that
it is vital for a liberal amount of
Silicone Heat Sink Compound to be used on BOTH sides of
the mica plate! Over the years since the SE-A3's were produced, if
they are stored or operated in very low humidity, the original heat sink
compound
used by Technics can actually dry up leading to poor thermal
conductivity between the transistors collector plate and the heat sink.
Then, the first one to get too hot, starts off the chain reaction
described above. So if you actually do have an SE-A3 with the
original OD503A-P transistors, it may be a good idea to check the transistor mountings to make sure there is no trouble coming. Also
of course, if you are fitting the new Toshiba transistors, make sure to
use plenty of a good quality
Silicone Heat Sink Compound when you put it all back together.
Having said all that, the front line of defense to this problem is clearly
written on page 2 of the SE-A3 user guide: "Because this unit is a
high-power amplifier, its temperature will increase when it is used for
long periods of time. Take care therefore, not to obstruct the
ventilation holes (located on the top panel) by placing other audio
equipment, books, etc. on top of this unit". Are we all
understanding about this now?.. The SE-A3 is NOT for
rack mounting and there must be NOTHING on top of it!
Be good to your old SE-A3 and it will keep going for a lifetime! If
you look closely at the pictures at the bottom of the page, you'll see
that some of those SE-A3's have had the dust covers over the power amp
heat sinks removed and in the
SE-A3MK2 design, these are not fitted either. I would say that
is a good idea if you are concerned about thermal problems with a stock
SE-A3. Its easy to clean out the inside with a compressor once in a
while like with all other amps.
SE-A3 Meter Lamps
 The
correct original part for the meter light pilot lamps is part number XAM43P,
order number NR57515-01 and you may be able to order them
here. They are very easy to replace compared to an SE-A5 and there is
no soldering required. All
you have to do is remove the top cover.
Then there are 3 screws to remove from the top of the white lamp cover. The lamps are then exposed to be changed.
SE-A3
Normal / DC inputs

The Technics SE-A3 is equipped with two sets of line level inputs marked "DC"
and "Normal". The "Normal" inputs are coupled by a capacitor like with an
ordinary amplifier. In this case, the amplifier operates as a normal AC
amplifier with a low end cutoff frequency of about 2Hz. As a capacitor is
a reactive device that can introduce phase changes, this is considered to be
undesirable, so the SE-A3 has been designed as a DC amplifier where there are no
capacitors in the signal path at all! So using the
"DC" input gives a
completely flat frequency response right down to 0 Hz with no reactive phase
effects of any capacitors! One possible issue with doing this is in case
of a fault in the device connected to the input. If for some reason there
is a DC component in the input signal, this will result in the speaker cones
being pushed and held there which will cause the voice coils to burn because of
the very low DC resistance compared to about 8 ohms of AC resistance (or
inductive reactance), although the internal output protection circuit will
operate to save the speakers from this. So if it is desired to operate with such an
input, the "Normal" inputs are provided to ensure compatibility and
are simply coupled to the "DC" inputs via a capacitor which eliminates
the DC component of the signal. When ever possible,
the amp should be operated using the "DC" input and only one set of the inputs
should be connected at one time. Some other Technics power amp designs
have just one set of inputs with a switch that has the "Normal" and "DC"
settings.
SE-A3 Specifications
Rated minimum sine wave RMS power output 20Hz~20kHz
both channels driven with 0.002% total harmonic distortion 8 ohms. |
200W |
Rated minimum sine wave RMS power output 20Hz~20kHz
both channels driven with 0.003% total harmonic distortion 4 ohms.
Note that because of impedance monitoring the 4 and 8 ohm values are
always the same. |
320W |
1kHz continuous power output both channels
driven 0.001 total harmonic distortion 8 ohms. |
220W |
1kHz continuous power output both channels
driven 0.003 total harmonic distortion 4 ohms. |
350W |
Dynamic Headroom 8 ohms. |
1.1dB |
Dynamic Headroom 4 ohms. |
1.8dB |
Total harmonic distortion at rated power 20Hz~20kHz. |
0.002% |
Total harmonic distortion at half power 20Hz~20kHz. |
0.001% |
Total harmonic distortion at half power 1kHz. |
0.0003% |
Transient Intermodulation Distortion. |
unmeasurable |
Frequency Response DC ~ 20kHz |
+0dB - 0.1dB |
Frequency Response DC ~ 300kHz. |
+0dB - 3dB |
Input Sensitivity |
70mV (1V, IHF '66) |
S/N (IHF A) |
100dB |
S/N (IHF '66) |
123dB |
Input Impedance |
47k ohms |
Low Frequency Damping Factor 8 ohms |
200 |
Low Frequency Damping Factor 4 ohms |
100 |
Total Load Impedance |
4~16 ohms |
Meter Reading Range 8 ohms |
0.0001W ~ 300W |
Meter Frequency Response from 10Hz~20kHz at more
than -40dB |
+/-1dB |
Meter Frequency Response from 10Hz~10kHz at less
than -40dB |
+/-1dB |
Meter Attack Time |
50usec |
Meter Recovery Time |
750 ms (0dB to -20dB) |
Power Consumption |
1150W, 1400VA |
Dimensions (WxHxD) |
16 15/16" x 8 3/16"
x 19 31/32" |
Weight |
32.5 kg (78.3 lbs) |
SE-A3 Pictures & Reference
Technics SE-A3 Literature
|