SWR SM 400 burned resistors on power amp

[chrisdski]

I’m a new student to solid state amp repair. I picked up a used SWR SM 400 bass amp. Stereo output with 2 power amps. Right channel works great. Left channel was distorted with pops. Did the usual testing suspect components pulled out of circuit (diodes, transistors). Voltage coming into the power amp was correct and scoped a clean DC current. Replaced one transistor that was questionable with exact match. Sound did not improved so I tested the 4 power transistors (2sd555). All checked out ok diode test NPN. Just for hoots I got 4 new ones and swapped them in as they screw onto their mounting plates. Fired it up and it worked and sound great. For about a minute. Then it popped thru the speaker with smoke coming out the questionable power amp. When I pulled the power amp 2 resistors had burned up. They seem to feed the 4 power transistors. The power transistors are toast also. I’m at a loss how I should proceed at this point. I have attached pictures and the closest schematic I could find though mine is an earlier model than schematic date.

[Jon Snell]

Looks like the 100R resistors clamping the bases to the emiters have blown. I would check all semiconductors and replace where required, check the 0.1R resistors, replace the 100R resistors and 250mA fuses then purchase and fit MJ21194 output transistor purchase all spares from a reputable seller; Mouser, RS, Farnel etc NOT fleabay as 2SD555 have been obsolete for a very long time.
Set the preset to maximum resistance before powering it up then set the quiescent current to produce an average of 2.5mV across each 0.1R resistor.

I would guess the most likely cause of failure was fleabay components. I could be wrong though and thermal runnaway could also destroy the output stage.

[glebert]

Where did you get the power transistors from? There are a TON of fake transistors, many coming out of China. They may work for very low power application but fail far below the specs of a legitimate part.

edit: Jon beat me to it.

[J M Fahey]

The original transistors were good/originals.
The ones you bought are fake.

Just for hoots I got 4 new ones and swapped them

Don´t "fix" what´s not broken or you´ll break it.
Now it´s worse than before.

[g1]

Here is the older schematic. 2SD555 mentioned in notes.

[nevetslab]

A side note. I like your PCB mounting fixture. Haven't seen that one before.

[J M Fahey]

It´s a hard to find if at all not made for MANY years transistor, but EBay sellers sell bucketloads of them, as in 4 pairs for 12 bucks , 5 pairs for 16 bucks, I guess you bought some of those.

Results?:

https://www.ebay.com/p/1853294017

Most relevant reviews

• 1 out of 5 stars
  by johnj822Nov 27, 2018
  
  Burned out
  
  Install these parts and burned out immediately.
  Verified purchase: Yes | Condition: New

https://www.ebay.com/p/2147201315

Most relevant reviews

• 1 out of 5 stars
  by pmkkingNov 02, 2017
  I tried 3 using theese 3 times to repair my amp. The. first time it worked for 15 minutes and then pop and it stopped working, Blew up many components
  
  The second time Smoke as soon as turned on, blew many components. The third time a bright flash and the 80 volt trace going to these transistors was blown off the board. These transistor are not what they are claimed to be. I will not buy any more parts from China on Ebay. Thank you: Patrick M. King
  Verified purchase: Yes | Condition: New

• What else?

[nosaj]

A side note. I like your PCB mounting fixture. Haven't seen that one before.

[chrisdski]

Thanks all for the reply’s, very helpful. Thanks for the proper schematic, very helpful. Today was resistor and diode day. On the attached picture I have circled the 2 burned 100ohm resistors and a 22ohm resistor (measured .1 ohm, maybe I’m picking up thru the inductor? Will pull and test) that will be replaced, and a single diode under spec in circuit that will be pulled and replaced. All other resistor and diodes checked to specs. Yes, the power transistors were an eBay purchase. I’ll contact them and see if I can get my money back.

[chrisdski]

Quick update. I got my money back on the eBay power transistors. Replacement transistors and resistors from Digikey. I replaced the two burned 100 ohm resistors, the 4 power transistors, and one of the 2N5416 transistors. Every transistor was checked out of circuit and passed. The bias resistor was set at max resistance. Powered up and works good. Next, I have to adjust the bias.

[nevetslab]

Powered up and works good. Next, I have to adjust the bias.

Typically you'd adjust bias for about 400mV across the 100 ohm emitter resistor of Q7, maybe a bit more. Looking to just begin getting current flow in the output stage...maybe 10mV across the emitter resistors.

[chrisdski]

So an update: I got the power amps to work, but they were intermittently doing a ground loop I couldn’t figure out. I ended up getting some new Chinese power amps off eBay that have similar specs. The only difference being the amp was supplying 60V +/- while the power amps required up to 55V +/- with 50V ideal. So I wired in a 1.8Mohm 3 watt resistor, which on my VM showed output voltage dropped to the 50V desired. But when I wired it into the power amps, it would show basically no voltage going into the power amps. I don’t understand why it didn’t work (too much resistance?). What would be a better solution to achieve a workable voltage drop? Anyways, I wired the power amps directly without the 1.8Mohm resistors, and the power amps worked great with input voltage at 56V+/- I kinda feel it will work as is, with some tolerance probably built in, but I would appreciate some knowledge as why my original thinking is incorrect, and what would be a better solution to obtain the voltage drop desired. Thanks. Chris.

[Helmholtz]

So I wired in a 1.8Mohm 3 watt resistor, which on my VM showed output voltage dropped to the 50V desired. But when I wired it into the power amps, it would show basically no voltage going into the power amps. I don’t understand why it didn’t work (too much resistance?).

The voltage drop caused by a series resistor depends on the load current.
If you only connect your DMM (probably having an input resistance of 10M), the load current is only microamps and the voltage might drop by 10V.

But with a real load current even as low as some mAs, the voltage after the resistor will be close to zero.
So yes, way too much resistance.

As the current demand of the power amp strongly varies with signal, dropping the voltage with resistors won't work.

[chrisdski]

Thanks for the reply. Another question- on the original power amp (schematic attached), what is the function of the AC line coming into the amp from the power bridge rectifier? I have seen some info about “speaker protection circuits”, and I am wondering if that is its function? SWR PA early.PDF

[Mick Bailey]

That's a soft-start circuit which prevents possible speaker pop when the amp is switched on.

Dropping the power amp supply rail voltage is more involved with a solid-state amp due to the higher current compared with a tube amp. How you achieve this depends on the space inside the cabinet and the method used. The problem with regulating the voltage down, or using a MOSFET dropper circuit is the heat dissipation - such methods effectively convert the voltage drop into heat and require a lot of heatsinking to prevent the regulator from burning up at full output. I can't think of a really easy way to do this that doesn't require a lot of space.