Why did Ge transistors persist as output-stages?

[G6Tanuki]

Originally Posted by M0XNA Neil

I have an old (and still working!) Elpico 50W PA which has Germanium TO3 transformer coupled output stage. The chassis looks like it was originally designed for valves and still has the 'HT choke'.

Germanium-transistor transformer-coupled P-P I have a weakness for - having worked on a range of gear that used eight Newmarket NKT404 transistors as the HT inverter and AM modulator [feeding a QQV06/40A valve PA].

As shown here: http://www.g3zud.com/vint_rad_eqt/pye/f206.jpg

[Synchrodyne]

An interesting case is that of the Eddystone EB35 Mk III receiver of circa 1975. This was virtually a complete rework of the earlier all-germanium EB35, which dated from the germanium era. The new circuitry used a mix of fets and silicon bipolar transistors and a TBA570 IC, except for the output stage, which was based upon an AC187, AC188 germanium pair. It wasn’t a case of simple carryover of the preceding output circuit, either, as in this case the output pair was driven directly from the TBA570. By 1975 Eddystone had used both discrete silicon bipolar and IC output sections in some of its other solid-state receivers.

The Valvo application note (1975 May) for the TBA570 shows both AC187, AC188 and AD161, AD162 output circuits. Possibly Eddystone was simply following one of the AN options. Nevertheless, one is left with the thought that as late as 1975, Ge outputs were still in some way seen as technically advantageous for receivers operated primarily from dry batteries, 9-volt in this case.

Now Eddystone was known for doing odd things at times (e.g. look at the strange valve mix in its 820 FM-AM tuner; presumably types in stock rather than ideal fits) and for hanging on to older practices (e.g. Rimlock valves in the 1960s 670C and 840C models, presumably carryover from their predecessors), but it would be difficult to put together a credible rationale along these lines in the case of the EB35 Mk III. Even relative cost is questionable, considering that Eddystone was using silicon output stages elsewhere in its range.

The TBA570 was by no means an early analogue consumer IC; it was second-wave if not second-generation. So its by-design coupling with Ge output stages could be described as intergenerational, although perhaps not to the same extent as those late-1960s RCA TV intercarrier ICs that were designed to directly drive valve output stages.

Cheers,

[Synchrodyne]

The 1970s combination of the TBA570 IC with germanium output devices to some extent may have been a carryover of previous practice with the TAD100, as shown in the attached Mullard advertisement from 1968.

Cheers,

[Synchrodyne]

It's interesting too that new germanium output devices were being released even after silicon had started to displace germanium. The Motorola 2N5324/2N5325 pair was an example, as shown in the attached item.

Cheers,