Successful Old Dynakit Transistor Audio Amp Repair Tale

Thread Starter

crutschow

Joined Mar 14, 2008
28,201
I just finished repairing my old Dynakit Stereo 120 basic amp after a sudden failure.
The 120 was one of the first good solid-state amps (all discrete of course) and I built mine in the mid 1970's, which I've used ever since. It was a relatively simple kit to assemble, since all the circuitry was on prewired and tested PCBs.
It's an old amp, and I suppose I'm sort of nostalgic about it, but it sounds great (no apparent "transistor sound") and served me well, and I'd hate to have to junk it. The only previous failure was a bad electrolytic in the signal path.

It powers some good sounding HSU bookshelf horn speakers (along with a powered subwoofer) for music at my desktop. A couple weeks ago while listening to some music at a low level while on my computer, there was suddenly a loud crack/pop sound from one of the speakers, and then everything instantly went dead...except for the sub still thumping away. :(

Fortunately, being a kit, it came with a manual (which amazingly I still have), fully describing its operation with complete schematics, detailed parts list, and node voltages. I removed the amp's cover to check voltages and determined there was no voltage out of the regulator (the amp was well designed with a series regulator for the 75V supply). At first I thought it was a regulator failure since the voltage from the diode bridge was good at about 90V, but after removing an output connection to test the regulator I found the regulator was okay. The regulator has fold-back current limiting, so a short was likely causing the zero supply voltage, and the likely candidate for that was the two output transistors connected in totem-pole fashion directly between the supply voltage and ground.
Indeed, one of the two amps did have two shorted power output transistors (<1Ω collector-emitter). It's likely one initially shorted, which then put the full voltage on the other, causing it to also blow. Apparently it was just an end-of-life type wearout failure (after only 50 years :D).
My troubleshooting was helped by having a Variac for the AC input so I could slowly bring the voltages up during testing and repair with less worry about zapping something else.

The original transistors were listed as 2N3055's in the parts list, but were apparently selected devices as they had a Dynakit part number stamped on them with a stated 90V Vcer rating, whereas standard 2N3055's are rated at 70V. It is thus quite likely that run-of-the mill 2N3055's (especially the cheap Chinese knockoffs) would be marginal. Fortunately I was able to find an upgraded version of the 2N3055 (MJ15015G) by a reputable vendor, which has essentially the same characteristics expect for higher voltage rating (Vceo=120V) and better safe area.
(It was a serendipitous find as I just happened to see it listed while looking at the 2N3055A ON Semiconductor data sheet.
Interestingly, they were about the same price as the 2N3055A, around $6.50US).

I was concerned that the transistors' shorting may have damaged something else, but after replacing them (along with a 0.47Ω emitter ballast resistor that I smoked during my initial troubleshooting), the amp works good as ever.

I also bought a couple spare transistors in case the other channel has the same failure in the future. :rolleyes:
So for about 35 bucks in parts (including the spares, resistors, new TO-3 thermal isolators, and thermal grease) I got it working again.
 
Last edited:

danadak

Joined Mar 10, 2018
4,057
I have the amp and preamp and tuner, have not powered them up since late 70's I think.

I have a Tek curve tracer, could screen some 3055's, but a part with overall better
safe area probably more prudent.

I will save your posts, looks like an excellent reference for when I blow mine up.

Regards, Dana.
 

wayneh

Joined Sep 9, 2010
17,155
I’ve been cleaning up my Dynaco SCA-80Q. My basement workshop suffered a mouse infestation last year and my old friend was not immune. It was chock full of shit and it makes me want to puke when I think of it. I’m past it now, after literally hosing it off in the sink. It’s working pretty well sound-wise but I may need to replace the pre-amp pots.
 

Yaakov

Joined Jan 27, 2019
3,755
I remember my father building a Dynaco amp and tuner in the late 60s. The tuner, I think used the big cap in the PS as the coil form for the choke. You wound a piece of insulated wired around it.

It was during this build that I first burned myself with a soldering iron. I was intensely attending to the mechanical connection and reached for the iron, picking it up by the wrong end. It took longer than it should have for me to notice I’d done that. The sound and smell registered before the pain!

Being the laser-focused type that I am (or maybe I am just “special”), I did it twice. But, I did learn on nature’s second lesson and haven’t had more than a solder splash burn since.
 

wayneh

Joined Sep 9, 2010
17,155
The tuner, I think used the big cap in the PS as the coil form for the choke. You wound a piece of insulated wired around it.
My amp has that exact configuration in it, two big coils on two big caps. The wire they supplied is quite stiff - I still have a piece leftover - and I remember winding those coils by hand. It was quite challenging to keep it neat.
 

Thread Starter

crutschow

Joined Mar 14, 2008
28,201
On the Stereo 120, the coils wound on the two big output caps are part of the output high-frequency low-pass filter (roll-off above 500,000 Hz according to the manual) to minimize RF interference and any high frequency instability.
 
Last edited:

bertus

Joined Apr 5, 2008
21,680
Hello,

I have been using the 2N3773 in stead of the 2N3055.
They are rated for 140 Volts and 150 Watts.
Just have a look at the datasheet.

Bertus
 

Attachments

Thread Starter

crutschow

Joined Mar 14, 2008
28,201
I have been using the 2N3773 in stead of the 2N3055.
Those look good and similar to the MJ15015G I used except for a somewhat higher voltage rating and high-current beta, and possible lower gain-bandwidth (the data sheet is not clear on that).
But I wanted to stay with something close to the 2N3055 except for the voltage/safe-area rating, since that was the transistor characteristics the Dynakit circuit was designed for, and the MJ15015G uses most of the same characteristic curves on the data sheet as the 2N3055a.

That's not to say the 2N3773 wouldn't have worked as well since the transistors are just used in a Darlington emitter-follower and complementary Darlington (Sziklai pair) emitter-follower output configuration with negative feedback, so should be pretty tolerate of any (likely) small differences in the operating characteristics.
 
Last edited:
Top