2 quad amps/ 4 speakers switch

[Soundfield]

A lot of amplifiers do not like to have open loads on their outputs.

Oh that story again. Perhaps you could list a few of them.

 

[Soundfield]

99.9% of today's consumer electronics comes from China, sad but true.

Yes indeed. Including the OPPO stuff which many on this site hold up as the very pinnacle of construction quality and reliability.

 

[par4ken]

Yes indeed. Including the OPPO stuff which many on this site hold up as the very pinnacle of construction quality and reliability.

As I recall even the boards for the Surround Master are made in China!

 

[MidiMagic]

99.9% of today's consumer electronics comes from China, sad but true.

Which China?

 

[MidiMagic]

Oh that story again. Perhaps you could list a few of them.

These events were about 45-55 years ago. I certainly cannot recall specific brands and model numbers I repaired that long ago. All of the amps had blown output transistors.

I do remember some of the brands.

- Dynaco - I fixed a Stereo 80 and a Stereo 120 that failed in this way (speaker was inadvertently disconnected).

- Sansui - I fixed various models. One failed because the speaker wire had a short in it. Others failed when the wire came off the speaker.

- Carver - I found all of the output transistors shorted on one side of the supply. This was a direct-out amp with 5 output transistors in parallel on each side of a +/-60 volt supply. The speaker was overdriven and had burned out first, and the amp did this with no load.

- Akai - I had a bunch of these, all with the same failure.

- Fisher - Various failures, all in the driver and output transistors.

I did an analysis of failure modes at the time:

The amps that failed because the output was shorted failed because too much current flowed through the push-pull transistors. One of the pair was shorted because of the high current damaging the crystal structure of the transistor. The other transistor was open because the high current flowed until something gave up.

The amps that failed because the output was open depended on the load to keep the transistors operating properly. When the load disappeared (especially if it suddenly disappeared), the voltage swing in the output transistors reached clipping. This often caused a driver transistor to short due to excess current. This then caused the output transistors to fail in the same way as above.

By about 1974, I noticed that newer amps had fewer failures. I also noticed that the amps included fuses or snubber circuits that prevented the wild swings in voltage when the load was wrong.

I did also note that, when the speaker wire opened or shorted, the first thing the owner did was turn up the volume.

 

[Soundfield]

The amps that failed because the output was open depended on the load to keep the transistors operating properly. When the load disappeared (especially if it suddenly disappeared), the voltage swing in the output transistors reached clipping. This often caused a driver transistor to short due to excess current. This then caused the output transistors to fail in the same way as above.

It's remarkably difficult to design an output stage that behaves that badly. They must have put a lot of effort into it.

 

[quadsearcher]

"When the load disappeared (especially if it suddenly disappeared), the voltage swing in the output transistors reached clipping. "
Wait, what? The amplitude of the signal in the output stage increased without an increase of input level or change in gain?
That is not consistent with transistor based amplifier operation.

 

[par4ken]

- Carver - I found all of the output transistors shorted on one side of the supply. This was a direct-out amp with 5 output transistors in parallel on each side of a +/-60 volt supply. The speaker was overdriven and had burned out first, and the amp did this with no load.

I've had that type of thing happen, however the overload itself likely caused the output transistor failure to happen first, then the shorted side caused direct current to flow to the speaker causing it to blow. The whole failure would be almost instantaneous.

 

[par4ken]

Which China?

The People's Republic of China, Taiwan (Republic of China) is no longer recognised as China except by its own government and people.

Interesting that Taiwan used to manufacture nothing but junk but now (judging mainly by ebay offerings) sells higher quality audio parts and components than mainland China does.

 

[skywaveTDR]

People that are not Engineers will not understand this but when a amplifier that is for power output especially a lot of it is worked into an infinite load it really is not designed to do that and when operated where not intended the voltages that were intended to terminate into 8 or 4 ohm loads would oscillate or have other irregular characteristics. There are all kinds of parameters an amp is designed for with Damping Factor, Transient response, Bandwidth, low distortion, crossover distortion and so on. The fact is the load is part of a circuit and when you remove it then not every time but some the amps will respond with bad reactions. Why do you suppose the Technician uses dummy loads on an amp instead of just drive them into air? It is cheaper to drive them into open circuit until you come to the repair cost. If I had a good amp I would not abuse it. Tube amps usually having a transformer in output may not have a reaction.

 

[MidiMagic]

"When the load disappeared (especially if it suddenly disappeared), the voltage swing in the output transistors reached clipping. "
Wait, what? The amplitude of the signal in the output stage increased without an increase of input level or change in gain?
That is not consistent with transistor based amplifier operation.

Remember that I said the first thing the owner did was turn up the volume.

It depends on the interconnection of the transistors and what happens to the current without a load. Remember I said the driver transistor also failed. The output line had a larger swing with no load. This
pulled more current from the driver transistors. One shorted, feeding full base current to one output transistor all the time. When the other output transistor was turned on by its driver, the current became too high. This was not a steady overloading of the output transistors, but an instantaneous one just long enough to damage the circuit.

Usually what they failed to design in is what happens to the currents in the driver and output transistors with a high signal level and without the load being present.

Also remember that there were several different output stage designs, each with its own characteristics:

- Identical transistors (usually both NPN) for control of current from high and low rail. In this case, the driver transistor must have inverting and non-inverting outputs.

- Complementary transistors (one PNP, one NPN), each with its driver transistor of the same type. The emitters are connected to the output.

- Complementary transistors (one PNP, one NPN), each with its driver transistor of the same type. The collectors are connected to the output.

I noticed in later amplifiers that small resistors had been added to the design limit base current.

 

[MidiMagic]

I've had that type of thing happen, however the overload itself likely caused the output transistor failure to happen first, then the shorted side caused direct current to flow to the speaker causing it to blow. The whole failure would be almost instantaneous.

I was there when it happened. The guy had three speakers hooked up. The first was his big speaker, hooked to one channel of the amp. It was on the other side of the stage. Another speaker for the near side of the stage and on the other channel could not handle the full output of the amp. That channel also had a speaker with a 200 ohm resistor connected in series with it to lower the volume for monitor use. I was in the control room with the monitor.

He cranked it all the way up, and the inadequate speaker cone actually caught fire. Smoke came out of the speaker. Then it opened. But I heard the monitor speaker continue for a second or two before the amp blew. After the sound quit, the resistor burned up. The other channel of the amp continued to work.

I was also there when the Dynaco Stereo 120 blew. I didn't know until later that the owner had hooked up only one channel to the PA speakers. There was no input and no output connected to the other channel. I was DJ-ing a dance with two turntables and a mono DJ mixer hooked to the connected side of the amp.

About an hour later, I suddenly detected "smelk", the acrid odor of a blown transistor. But the sound continued to work, so I thought it was something else at the dance that failed (there was also a live band). It was when we started striking the setup that I found that only one side of the amp was hooked up, and that the heat sinks on the other side were quite hot.

 

[quadsearcher]

.