A question of feedback

This week, Dave Wilson’s son asked him some tricky questions about the inner workings of the operational amplifier.

‘Nothing capable of being memorised is history’. – R. G. Collingwood

Earlier in the week my son phoned up. He had a problem. Despite the fact that his teachers had provided him with some basic information about that bastion of analog computing – the operational amplifier – it seemed he was having some trouble understanding just how that classic piece of analog wizardry actually worked.

The operational amplifier, of course, has been with us for a long time. The first such devices were introduced as tube devices in the early fifties by the old Philbrick company. But it wasn’t until the 1960’s, with the introduction of solid-state devices developed by analog genius Bob Widlar at Fairchild Semiconductor and then National Semiconductor that they really started to get the attention that they deserved. Then, through the 1970’s, as they dropped in cost and improved in variety, even designers of non-military equipment could afford to take advantage of their rather amazing characteristics. Today, of course, they are regarded as commodity items.

But what makes these beasts so important to designers of everything from missiles to hi-fi? Well, three things. First off, their gain is extremely high – ideally infinitely high. Secondly, their input impedance is enormous – ideally infinite. Thirdly, there’s their tiny output impedance, which ideally would be zero.

When one takes those characteristics into account, it’s not hard to see that the voltage gain of a circuit comprising just one operational amplifier and two resistors – one input resistor and one very important feedback resistor – is set entirely by the ratio of the two resistors.

Once these points had been made to the prodigal son, he was happier. His questions had been answered.

But the question of feedback remained on my mind. Almost every op amp circuit developed today uses some form of feedback. Yet the whole concept of using feedback in an amplifier wasn’t developed until the 1920s by Harold Black while he was working at Western Electric in New York on the problem of how to build a series connection of hundreds of telephone repeater amplifiers using directly heated triode tubes.

Today, many regard Black’s work in developing the feedback amplifier as important as the development of the aircraft engine. Surely, It was truly the enabling technological breakthrough that allowed Bob Widlar’s op amp to make it big at all?

Sadly though, neither Bob nor Harold made it big onto the GCSE curriculum this year. They didn’t even get a mention. And that’s a pity, because history can teach us everything, as well as giving us an insight into the future.

Dave Wilson


The Engineer Online