Question about Frequency Changer Valves

[julie_m]

This is driving me crazy. There must be an answer .....

Why, in the usual triode-hexode mixer arrangement, is the hexode G3 driven from the grid of the triode? Surely if G3 was driven from the anode instead, it would have even less influence on the oscillator frequency? I always thought that the idea of the separate oscillator in a mains set was to minimise any interaction between the aerial and oscillator signals; so if G3 is driven with the triode grid, and not from the anode, surely this creates a possibility for unwanted interaction?

[WME_bill]

Frequency changer valves. I suspect the answer to your question lies with the internal construction of the valve and ease of manufacture, as well as the electrical performance. There is an interesting section in Langford-Smith (4th Edn, 1955,p 996-999), where the construction and performance of 6J8 (X61M) triode heptode, 6A8 pentagrid, 6K8 triode hexode, 6SA7 pentagrid and 6BE6 pentagrid, and the merits of inner grid oscillator injection as compared with outer grid injection are discussed.
If you haven't got these pages to hand, I'll scan them. wme_bill

[kalee20]

Some triode-hexodes (eg ECH81) have g3 brought out to its own pin, so you can connect it wherever you want.

If g3 be connected to the triode anode, it'll have a significant positive voltage on it, which isn't what you want as it'll divert much of the hexode's current.

Whether linking g3 to triode grid or anode, from a frequency stability point of view, is a good idea, depends on the oscillator circuit. Often, the triode is connected with a tuned-grid stage so yes in principle you could be right. But sometimes the grid circuit is untuned and only loosely coupled to the oscillator tuned circuit, which is tuned anode configuration.

There again, you also have the possibility of inner grid injection and outer grid, as WME bill observes above.

[Synchrodyne]

As already mentioned, the ECH81 had no internal connection between gt and g3, allowing some flexibility in the way it was configured. To some extent that would have been because it was designed for the era of FM-AM receivers, in which the heptode was independently used as an FM 1st IF amplifier. Also, the car radio version, the ECH81, was sometimes used in pairs, one being the customary oscillator-mixer, and the other being an RF amplifier (heptode) and 1st audio amplifier (triode).

Notwithstanding the lack of an internal connection, in some consumer applications, including those involving HF towards 30 MHz, the ECH81 was used as a frequency changer with the gt pinout directly connected to the g3 pinout. So that arrangement, which mimicked earlier ECH-type valves with their internal connections, was evidently perceived as having adequately low oscillator pulling in those situations.

In more exacting HF receiver applications, the separateness of the two parts allowed the heptode to be used as an externally fed mixer, further reducing oscillator pulling. There was some recent mention of the various ways in which the ECH81 was used here: https://www.vintage-radio.net/forum/...=38395&page=12, #239.

By the way, insofar as it was not unusual for oscillator anode voltage to be fed back to the oscillator grid, and setting aside for now the previously mentioned DC level problems, coupling the triode anode to g3 might not have provided full relief in terms of avoiding getting signal voltage into the oscillator loop.

Cheers,