|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
16th Aug 2017, 9:11 pm | #1 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
"Short" Superhets.
Is there any general agreement on what constitutes a "short superhet" ??
I've seen the term applied in two distinct cases. [1] A frequency-changer feeding into a single IFT followed by a detector (sometimes a bare diode, more-often a leaky-grid triode, sometimes with a hint of regeneration to liven things up) and then the traditional 2-stage driver-and-power-output audio amp. [2] A frequency-changer followed by a single IF-amp into a diode detector then a single 'high slope' audio power-output-stage (often including a couple of diodes as detector and AGC). Both mean you can build a 'superhet' using 3 signal-frequency valves; I'm familiar with type [1] receivers of the 'midget' type some of which were imported from the US either side of WWII and which sometimes appeared as 'kit' radios on the surplus market. Type [2] is typified by the WWII "Wartime Civilian Receiver" and the Ekco AD75. There's also what I know as a "Super-gainer", the amateur-radio equivalent of [1] which was essentially a regenerative receiver designed to tune over a narrow bandwidth, with a 'down-converter' in front to translate each of the amateur bands to the narowly-tunable IF [which could thus be made to gently oscillate to allow CW/SSB reception without 'hand capacity' or swaying-antenna effects]. There was also the "superregen-superhet" that was used extensively in WWII-era RADAR and IFF sets: a frequency-changer bringing VHF/UHF signals down to around 30MHz where they were resolved by a fixed-frequency superregen receiver. This had the advantage of providing isolation between the superregen-stage and the antenna, superregens otherwise being notorious for radiating wideband noise around the frequency they were tuned to. |
16th Aug 2017, 9:21 pm | #2 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,788
|
Re: "Short" Superhets.
The vast majority of British short superhets dispense with the audio preamp. Pye were particularly fond of this arrangement in the 40s and 50s, and even used it in relatively upmarket radios and radiograms. Initially the diodes were put in the output valve (EBL31), later moving to the IF amp (EBF80).
The type with the missing IF amp was very rare in Britain - in fact, I can't think of a single example. They weren't uncommon in some countries though. |
16th Aug 2017, 9:27 pm | #3 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
Re: "Short" Superhets.
The type with the missing IF-amp I generally associate with Pre- and early-WWII-era US 'midget' radios (along with resistive line-cords) as a Depression-era precursor to the classic "All American Five". I've seen several sold-in-the-UK 'kit' radios using this layout too, as opposed to the more-usual 1-V-1 approach.
I've also seen schematics for US radios in which there *is* an IF-amp but it uses a resistive, or untuned-inductive RF-choke load rather than a proper second IFT. |
16th Aug 2017, 9:36 pm | #4 |
Heptode
Join Date: Jul 2008
Location: Hereford, UK.
Posts: 719
|
Re: "Short" Superhets.
The sobellette is probably the most likely to be encountered example of (1), Cossor 484 is another example
|
17th Aug 2017, 8:33 am | #5 |
Rest in Peace
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
|
Re: "Short" Superhets.
There are designs that follow the type 2] design above but add an RF amplifier in front of the frequency changer. They are not short superhets as I understand the term.
An example is the Philips 170A |
17th Aug 2017, 9:21 am | #6 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: "Short" Superhets.
Ever ready 5014 is another British example of (1).
And Pye 47C another example following Colin's remark. Regarding Colin's comment, I've never understood the reasons for doing that! Once you have gone to the trouble of changing the frequency, if you add an extra valve, amplifying at AF is the cheapest way of using it to get more gain; adding an IF stage is the next cheapest and of course gives you extra selectivity. Using it at RF is the most expensive way (extra section on the tuning capacitor isn't cheap; extra wavechange switching and doubling-up on coils for EVERY waveband), and the gain benefit is least! The only plus is much better image rejection. Or have I missed something? |
17th Aug 2017, 12:05 pm | #7 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: "Short" Superhets.
Lower noise from the frequency changer and better AGC control.
Leon. |
17th Aug 2017, 1:01 pm | #8 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: "Short" Superhets.
Definitely lower noise, yes I missed that! Thanks Leon
And yes I can see better AGC action, as with no AF voltage amplifier the detector must be handling lots of volts and we wouldn't want it overloading by being biased hard back. Whereas the RF stage, handling small signals, wouldn't care. But that wouldn't be much of a selling point, surely? Or was it! |
17th Aug 2017, 1:37 pm | #9 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: "Short" Superhets.
I wonder if the RF-stage-and-high-slope-AF-output recipe had utility within a dozen or so miles of a multiple sender site, where very strong signals many kHz away were of more concern than strong ones just a few kHz away? Good RF selectivity, though not much overall gain. (Speaking as someone who once lived just a few miles from the Burghead site!)
|
17th Aug 2017, 4:58 pm | #10 |
Heptode
Join Date: Mar 2011
Location: Dorset, UK.
Posts: 519
|
Re: "Short" Superhets.
I do recall seeing a set with 6A8 frequency changer, 6B8 if/af amplifier plus detector in a reflex circuit followed by a conventional 6V6 output. A very good example on how to make a single valve amplify two different signals simultaneously.
It had partial AVC applied to the 6B8 and full AVC to the frequency changer. I think the 6B8 had a semi variable-mu characteristic, very suitable in a reflex configuration. This set did not appear to suffer from gain issues and compared favourably with a conventional line-up. Rich
__________________
To an optimist a glass is half full; a pessimist half empty; an engineer twice as big as need be! Last edited by Marconi_MPT4; 17th Aug 2017 at 5:07 pm. |
17th Aug 2017, 5:10 pm | #11 |
Octode
Join Date: Feb 2009
Location: Bletchley, Buckinghamshire, UK.
Posts: 1,205
|
Re: "Short" Superhets.
That Philips had a short wave band. The preselector RF stage meant it worked much better than that on other domestic sets, in which image responses were a big problem.
|
17th Aug 2017, 5:26 pm | #12 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: "Short" Superhets.
Many radios have too much audio gain, so the volume control needs to be kept down. Useful for marketing, but poor for using. Pye (and others) realised that you could get away with less gain for domestic broadcast reception.
|
17th Aug 2017, 6:48 pm | #13 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,788
|
Re: "Short" Superhets.
Unfortunately there's no ideal compromise. An audio preamp results in more gain than is really needed, but doing away with it means the radio is very dependent on gentle AGC operation, with a sort of 'cliff edge' effect with weaker stations. The Pye short superhets do work perfectly well in the strong signal areas typical of UK BBC listening, which I suppose was all most customers wanted.
|
17th Aug 2017, 9:12 pm | #14 |
Retired Dormant Member
Join Date: Apr 2003
Location: Salt Spring Island, BC, Canada
Posts: 368
|
Re: "Short" Superhets.
The american midget short superhets such as the Arvin 243T used 4 tubes 12SA7, 12SQ7,50L6 and 35Z5. Close examination of the schematic reveals that the 12SA7 is also reflexed as a IF amplifier so the overall gain is better that you would expect from so few tubes.
John. |
18th Aug 2017, 4:26 pm | #15 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: "Short" Superhets.
It is a pity that the 6Q7 (and its ancestors and offspring, such as 6AT6/EBC90) contained a relatively high mu triode (70). Less gain would be more appropriate for its usual application.
|
18th Aug 2017, 5:45 pm | #16 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: "Short" Superhets.
The 6R7 (pin compatible with the 6Q7) was available to meet applications requiring a lower mu. Remember though, that US output tetrodes such as the 6V6 and 6L6 had poor gms compared with their European counterparts and needed more drive. The requirement for a gram (phono) input would also have favoured a high mu triode, as would the welcome addition of negative feedback around the audio amplifier in better sets.
If anything, there was a trend in the US for even higher mu triodes such as the 6AV6 (100). As has been previously pointed out, audio gain is cheap and gives the set an artificial impression of high sensitivity to an unskilled user. A Philips 170A (3 valve + RF stage) can seem fairly dead to the uninitiated, who perhaps may be unused to advancing the volume control. Leon. |
20th Aug 2017, 2:48 am | #17 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: "Short" Superhets.
The 6BF6, with a 16-mu triode, was the miniature successor to the 6R7. It was introduced just after the 6AT6, but before the 6AV6. So, the valve makers at least were catering for all tastes. There was also a corresponding trio with high-perveance diodes, and in the American series at least, various other double diode-triodes designed primarily for TV and FM applications, but also usable for AM receivers. These had triodes of diverse characteristics, and usually three, sometimes two separate cathodes.
It does seem that the short superhet design involves a choice amongst a number of trades-off. The RF stage with a third gang certainly does help when you are next door to a set of regional transmitters. Also, an RF stage with agc applied allows the possibility of using only fractional agc on the IF stage without unduly compromising the overall agc curve, this reducing the incidence of modulation rise distortion when receiving strong signals. A third gang used to provide bandpass input tuning (as favoured by Murphy in the post-WWII era), without an RF amplifier, would also provide similar benefits except for those associated with agc and noise reduction, the latter more applicable to HF than MF and LF. Reserve AF gain would allow tapping down of the demodulator output in order to minimize distortion due to differential AC and DC loading, as covered in this Murphy item: These charts, the first attributable to Murphy and the second to Marconi, show some of the benefits that accrue from using an RF amplifier: If one assumes that a basic and simple form of the short superhet is mixer/oscillator, IF amplifier with demodulator and agc rectifier, and high-slope output valve, then successive valve lines-up reflecting the various stages of post-WWII miniature valve development might be: ECH42, EAF42, EL41 ECH42, EBF80, EL41 ECH81, EBF80, EL41 ECH81, EBF80, EL84 ECH81, EBF89, EL84 Insofar as high IF gain would seem to have been required with this arrangement, and probably with not much to spare to allow for tapping down of the demodulator output, neutralization of the IF stage may have been preferable, to allow very high gain without the passband tilting that is an early sign of regeneration. But that was probably an unwelcome complexity in what were simple receivers designed for customers for whom low price was more important than performance. The arrival of the EBF89 c.1955 with its higher-slope pentode, still with very low Cag, may have eased IF stage design. After this, perhaps the ECL82 (c.1956) could have been considered in place of the EL84. Then there would have been reserve AF gain, although maybe marginal output power. The ECL86 (c.1960) would have solved that problem. Cheers, |