A simple stereo Amp aroud a 2N3773

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have four old Toshiba 2N3773 transistors (along with two new Motorola’s; six total 2N3773) from a 1970-s era Sunn Coliseum Lead guitar amplifier. I also have a pair of their PNP counterparts, the 2N6609 made by ON Semiconductor (Motorola?). In addition, I have any number of old 70’s-era Marantz and Pioneer receivers from which to scavenge parts from, and two 460VA transformers from a couple of early-1990s Technics integrated amps. I am thinking, instead of tossing ways a bunch of still good parts from old equipment, I would put them to use in a new, simple, clean sounding stereo amp. Obviously I would welcome any suggestions anyone might have, or some schematics of amplifiers that use the 2N3773 and 2N6609. It needn’t be pretty, but it does have to sound good.
 
--------------------------------------------------------------------------------
And the Saga Continues ...
Now we have Toshiba branded 2N2773 power transistors. This in itself is interesting, as a search on the Toshiba site reveals that they don't even seem to make this transistor! It would be unusual for a Japanese manufacturer to make a "2N" device at all, but doubly so since this is a very old device now, and seems to be discontinued by just about every other maker.

Again, these have all the "earmarks" of counterfeits - and naturally enough someone was caught out, and his amp failed with these transistors installed. If you happen across any of these components, be afraid - be very afraid!
 
These transistors were removed from an old 1970s amp. There was not much counterfeiting back then, if any. China had no semiconductor industry then. They are probably OK.

Most of the cost in an amp is in the power transformer, filter caps, chassis and heat sinks. The semiconductors are fairly cheap by comparison. You may want to consider using some better parts for improved sound quality and reliability since it really won't affect cost much. The 2N3773s are very slow and not capable of dissipating as much power as their modern counterparts. That said, I use them for repairing Dynaco 120s without problems.
 
can i give +/-50v to this circuit

Audio Amplifier 90-100W with 2N3773
Crimson's CE1704 uses a single pair 2N3773 (Toshiba) in quasi complementary for a 100W into 8r0 amplifier, using a 35+35Vac 250VA transformer. It's listed in their catalogue as a recommended set of components.

I have had 6 of these amps (CE1704) over the decades I have never damaged any of them.

I have tested two CE1704 to 200W into 4r0 using a 35+35Vac 625VA transformer. That exceeds the specification of 170W into 4r0 (1704) and did not damage the amplifiers.
I would not use the 3773 at >+-65Vdc due to their Vce0=140V
 
Last edited:
Back in the day before I fully understood transistor ratings, I rebuilt a Phase Linear 400 with real 2N3773s. I am still amazed it worked fine (±85V!).

The Crown D150A would do 90W/8R with one pair of 2N3773, about 130W/4R. They would blow up after a while, I think the heatsink arrangement was less than adequate.

I think the counterfeit transistor thing started with the distributors, not the manufacturers. I was buying un-branded 2SD555s from a distributor, no markings at all. I called them up and told them they were fake, they asked 'how do you know', I replied 'because they are PNP'. After this their parts all had brands and numbers on them. Their MJ15024 looked like Motorola, but would not run at ±80V. They did run at ±57V. They also sold me some Toshiba branded 2N3773 that would not run at ±67V (Crown DC300A), and I assumed they were fake too.

Can a Toshiba 2N3773 meet JEDEC spec for that part number? I doubt it. Real RCA and Motorola parts were rated at 200°C, I never saw a Toshiba part of that vintage rated above 150°C.
 
Something not generally known is that from the mid 80's Toshiba made a lot of BJT parts for Motorola, particularly motor drive modules with IGBT's but also many of the regular 2N and MJ series parts with the Motorola logo's on.

I expect they may well make the MJL and NJL parts as the numbers tally with the 2SA / 2SC parts Toshiba used to make before introducing the 'C5200.

Most Jap spec sheets are more conservative with power ratings, in particular Sanken, but most of the others too.
Very few will rate a TO220 part beyond 40 watts which is probably a good real life dissipation level for this case, vs 80 and 90 watt ratings for US and European parts. This conservatism could explain the 150 degree temp rating.

(I've seen TO220 fets rated over 200 watts by IR and others, presumably clamped to a super flat liquid cooled block of copper.)

The 3773 is a rugged beast with very strong SOA but isn't too fast. I got over 100W from a 3773 / 6609 pair many years back.

Never got to see how long they lasted as the amp powered speakers on a dodgem ride. Every few months the speaker wires would get dropped onto the 70 volt DC grid powering the cars and blow the earth wires in the amp usually taking out the silicon in the amp stage.
Never thought to put a fuse in the speaker live and ground return, but with the ride operator paying to have it fixed each time and money being tight my focus may have been elsewhere....
 
Back in the day before I fully understood transistor ratings, I rebuilt a Phase Linear 400 with real 2N3773s. I am still amazed it worked fine (±85V!).

If you screen modern (4MHz fT) 2N3773's to 200V Vceo, they are better than any of the original PL400's transistor options (FPL909, XPL909, RCA1B05, 2SD555). Only when they went to the 15024/15025 pair in the later compelmentary version did they get "better". I used to use Vceo-screened 2SD424's as replacements in both the PL400 and CS800 because they were only $3 a pop in 10+ lots and didn't blow up.

Can a Toshiba 2N3773 meet JEDEC spec for that part number? I doubt it. Real RCA and Motorola parts were rated at 200°C, I never saw a Toshiba part of that vintage rated above 150°C.

2SD873 was Toshiba's original "2N3773". It was rated 175C. The beta curves look like they were pulled right from the original Moto data sheet. The s/b curve, however, looked like a copy of the one for the D424 which has the breakpoint at 70V. Real D424's back in the day would routinely withstand the 2N6259 torture-test (2.5A at 100V for 1 second).
 
I have four old Toshiba 2N3773 transistors (along with two new Motorola’s; six total 2N3773) from a 1970-s era Sunn Coliseum Lead guitar amplifier. I also have a pair of their PNP counterparts, the 2N6609 made by ON Semiconductor (Motorola?). In addition, I have any number of old 70’s-era Marantz and Pioneer receivers from which to scavenge parts from, and two 460VA transformers from a couple of early-1990s Technics integrated amps. I am thinking, instead of tossing ways a bunch of still good parts from old equipment, I would put them to use in a new, simple, clean sounding stereo amp. Obviously I would welcome any suggestions anyone might have, or some schematics of amplifiers that use the 2N3773 and 2N6609. It needn’t be pretty, but it does have to sound good.
Was it the amplifier (include schema) from the attached jpg files?
about
http://www.diyaudio.com/forums/soli...-models-quasi-complementary-power-output.html
you will find more schematics from vintage commercial products, also suited for the 2N3773.

Back in the early 80's (that is before many of you were born!) I designed Pulsed ATE power supplies using Motorola 2N3773.
We used to abuse the hell out of them but never had to use the stock of spare devicea we prudently stored. Indestructable, AFAIR.
Which brand was the developer of the 2N3773? Maybe first release from RCA or Motorola? Older or newer than 2N3055?
 

Attachments

  • Sunn Coliseum chassis.jpg
    Sunn Coliseum chassis.jpg
    38.8 KB · Views: 1,336
  • Sunn Coliseum Slave card.JPG
    Sunn Coliseum Slave card.JPG
    267.5 KB · Views: 1,304
  • Sunn Coliseum Slave Schematic.jpg
    Sunn Coliseum Slave Schematic.jpg
    244.8 KB · Views: 1,863
Last edited:
MJL21193 can do just under 2 amps at 100v, but has drooping gain with increasing Ic.

Sanken 2SA1216 and 2SA1295 and their complements the 2SC2922 and 2SC3264 both can do 1.5A at 100v and have good gain linearity and are both fast and tough.

Other than that not much that I know of.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.