|
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 |
7th Jan 2017, 6:29 pm | #21 | |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
Re: Valves in commercial applications for which they were not specifically designed.
Quote:
I've seen it suggested that a throttled-back high-power valve as the LO potentially gives a lower level of harmonics, but I can't remember *where* this suggestion appeared. |
|
7th Jan 2017, 6:41 pm | #22 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
Re: Valves in commercial applications for which they were not specifically designed.
The infamously unreliable Mazda U801 quadruple-power-diode was used in some low-budget TV receivers as a combined HT rectifier and efficiency-diode. This both ran the mains-rectifier diodes close to their maximum current and put the efficiency-diode part under significant PIV stress; the result was not pretty.
|
7th Jan 2017, 7:25 pm | #23 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Valves in commercial applications for which they were not specifically designed.
High transconductance valves are often used as oscillators as this means that the tuned circuit (which may be a crystal) can run at lower power and thus higher stability. I understand that some of the best oscillator valves are video output pentodes.
|
7th Jan 2017, 7:42 pm | #24 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
|
Re: Valves in commercial applications for which they were not specifically designed.
Quite often the motivation is to reduce the number of types used in a set, sometimes to limit the number of types used in the factory. If you load all your usage onto fewer types you can often screw the prices down from the supplier.
Power valves (and RF power transistors) in oscillators are not uncommon. There is a 5763 in the Marconi TF144H for example. Running the oscillator rather hot and applying plenty of attenuation is a recipe to reduce pulling by loads. It can also do nice things to phase noise if you're careful. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
7th Jan 2017, 7:56 pm | #25 |
Heptode
Join Date: Mar 2008
Location: Newmarket, Suffolk, UK.
Posts: 611
|
Re: Valves in commercial applications for which they were not specifically designed.
Many HF amateur radio transceivers in the 60, 70 and 80s used pairs of TV line output valves (aka sweep tubes) as PAs, especially Yaesu and Drake, and also the KW Vespa Mk2 and Atlanta.
Also 4 off in linear amps. 6JS6 and 6KD6 in particular Also some homebrew design and CB commercial amplfiers using European valves like PL509/519 and EL versions 73 Fred G4BWP |
7th Jan 2017, 11:02 pm | #26 | |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Valves in commercial applications for which they were not specifically designed.
Quote:
This used a UCC85, UF89, UCL82, UY85 and a pair of OA79. A standard UCC85-based tuner head was modified to cover the MF band as well. Armstrong also used the ECC85 in unusual ways, this apparently happening with the introduction in 1958 of it ST3 AM-FM tuner, and its Stereo-Twelve AM-FM Radiogram chassis, which shared the same tuner circuit. In the tuner, which had an ECC85-based FM front end, half of an additional ECC85 was used as the reactance valve for FM AFC. The other half was deployed as a 2nd AM amplifier for AM. Armstrong claimed that this reduced modulation rise and its resultant distortion, which was in line with the conventional wisdom about the benefits of non-agc’d final IF stages. But it does look as someone had looked up Martha Stewart’s commentary on uses for leftover valve portions, and had found something for which a beneficial claim could be made. Armstrong also used a pair of ECC85s in the audio section of the Stereo-Twelve. This may simply have been to facilitate the use of the same double-triode type throughout, which the FM front end requirement dictated be the ECC85. The Jubilee AM-FM radiogram chassis, nominally the mono counterpart to the Stereo-Twelve, had an ECC85 of which one half served as the reactance modulator and the other half as an AF amplifier. There was no second AM IF stage in this case; its desirability evidently disappeared with the absence of a spare half-ECC85. So if not “from DC-to-daylight”, the ECC85 was used from DC to VHF, although it was typically presented as being primarily an FM front-end valve. Cheers, |
|
9th Jan 2017, 7:28 pm | #27 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
Re: Valves in commercial applications for which they were not specifically designed.
Another oddity: I recall a US airband 'base station' transceiver where a "Diode-strapped" 6AQ5 beam-tetrode was used as a HT-delay!
The 6AQ5 had its heater fed from the same transforner that supplied 6.3VAC to all the signal-frequency valves. In its anode-circuit was fitted a relay. The "HT" for this bit was the raw 117VAC mains. The relay, when energised, completed the circuit to the HT transformer and its [indirectly-heated] rectifiers. There was a capacitor across the relay to stop it 'chattering' when fed with half-wave-rectified DC. The sequence was - press 'ON' and the transformer feeding the filaments for everything except the HT were energised. When the 6AQ5's filament got hot-enough to provide decent emission, it became a half-wave rectifier and pulled-in the relay [and illuminated a 'READY' lamp on the front panel]. The relay then applied AC to the HT transformer, which also fed the filaments of the HT retifiers - which warmed-up and supplied HT to the transmitter/receiver. I guess they used a 6AQ5 in this application to reduce spares-count: there were also four 6AQ5 in the parallel-push-pull transmitter modulator and another as the receiver output-valve. |
10th Jan 2017, 12:41 am | #28 |
Heptode
Join Date: Oct 2005
Location: Olympia, Washington, USA.
Posts: 663
|
Re: Valves in commercial applications for which they were not specifically designed.
Here in the U.S. one of the "El Cheepo" record players used a 6J5 as a rectifier. I thought it was odd at the time, but it was just post WW2, and tubes were still hard to get. On one piece of stereo gear I saw a 6AQ5 being used as a driver for the outputs which were 6L6's .
During WW2 it was common practice to use an audio output tube as a rectifier here, as tubes were extremely hard to come by, unless you knew a black market supplier. Per my readings;The US Mfgr's even marked some tubes as "Out of Spec" so they could be sold on the civilian market, or marked them "MR" (Material Replacement), so civilians could have their radios working. I have quite a nice stash of "MR" tubes in my collection. Of course, us Hams use oddball tubes for oscillators and outputs on home brew gear all the time. I made a "Lowf'er-160KC CW transmitter back in the early 70's using a 12AX7 as the output tube and the osc. HeathKit company was famous for using Mil. Surp. tubes. They got them cheap and designed their gear around them. |
11th Jan 2017, 7:25 pm | #29 |
Rest in Peace
Join Date: Sep 2010
Location: Ripley, Derbyshire, UK.
Posts: 785
|
Re: Valves in commercial applications for which they were not specifically designed.
I cannot remember where 'collected' this circuit from, but it shows a novel use for an EM87 magic eye as a grid dip oscillator. Pushed to much higher frequencies than it would normally be used at! Saves the cost of a meter, too. . Tony
|
11th Jan 2017, 9:54 pm | #30 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Valves in commercial applications for which they were not specifically designed.
Another example was an early 1950s Philips projection TV receiver, as described by Kerkhof & Werner (1). This used a DF96 battery pentode (triode-strapped) in the picture tube timebase failure protective circuit, chosen because of the very low thermal inertia of its heater, which was fed from an additional winding on the field output transformer. Thus, termination of the heater supply would shut down the valve very quickly. When cutoff, this valve allowed a high negative bias to be placed on the tube grid, thus suppressing the beam. It was biased for cutoff, but normally held off from this condition by positive rectified line flyback pulses. The receiver in question otherwise had a mix of E-, P- and U-series valves, variously on Rimlock, noval and B7G bases.
The apparent commercial risk for the equipment manufacturers – as distinct from amateurs and home constructors – in using valves for non-intended applications was that in some cases the success of this approach could have depended upon valve properties that were typical at the time that the equipment was designed, but were outside of the ambit of those properties that needed to be controlled to assure success in intended applications. Thus these uncontrolled properties might be subject to deleterious drift or even step-change if the valve makers for example made changes for production reasons, and these changes, whilst retaining or even enhancing the primary properties were made without much consideration being given to what effect these changes had on secondary properties. That risk would not have been there for all situations. The above-mentioned example depended essentially on the low heater inertia of battery valves, which was an inherent property of their very low heater power, so highly unlikely to change. And at times the valve makers did take notice of these non-intended uses by equipment makers, and came up with new or improved valves to cover them. Cheers, (1) Kerkhof & Werner; “Television”; Philips Technical Library; 1952; see page 402. |
13th Jan 2017, 9:03 am | #31 | |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
|
Re: Valves in commercial applications for which they were not specifically designed.
Quote:
The RSGB did a CDROM of all of Pat's columns. You can lose yourself in it for years! David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
|
13th Jan 2017, 9:13 am | #32 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
|
Re: Valves in commercial applications for which they were not specifically designed.
The use of active devices in applications they were not specifically designed for is alive and kicking today. The semiconductor industry is still head over heels in love with the cellphone. If you're trying to design anything other than a phone or a laptop/tablet, you'll find a general lack of interest from component firms. What makes it even worse is that parts now usually lack performance data outside the phone and ISM bands.
A few (thank heavens) devices at least still have continuous s-parameter files that don't have gaps between the consumer bands, though the rest of the datasheet gives only band-specific numbers. It's a bit like those maps with 'Here be dragons' marked on blank areas. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
13th Jan 2017, 10:28 am | #33 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: Valves in commercial applications for which they were not specifically designed.
I do like boxdoctor's EM87 grid dip oscillator- very economical and elegant!
|
13th Jan 2017, 11:41 am | #34 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Valves in commercial applications for which they were not specifically designed.
Philco were quite creative with their model 38-7 set, schematic attached.
Instead of using a double diode like a 6H6 for a detector and avc combination, they used a 6J5. They used the grid and the plate as two separate anodes. The 6J5's plate circuit which develops the AGC voltage has a negative bias (from the resistor divider in the centre-tap to ground of the power transformer secondary). The AGC function is effectively uncoupled for low level signals because the 6J5's plate is negative with respect to its cathode, it only conducts with high level signals so this results in a delayed AGC. (one other very interesting & unique feature of the radio is a single additional turn of wire added around one of the first IF transformer's coils, and placed in series with the suppressor grid of the IF amplifier 6k7. Does anyone have a theory on the action of this coil ?) |
13th Jan 2017, 12:26 pm | #35 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: Valves in commercial applications for which they were not specifically designed.
I wonder if it was a gentle way of introducing neutralisation or degeneration in a stage that was expected to provide considerable gain, and to operate consistently over a wide range of control grid bias with AGC action? Whilst pentodes can have very low effective feedback capacitances, I gather that even in the era of the button-base pentode (e.g. EF89) without the extraneous lead-ins/stray coupling of earlier pinch-type construction, a sniff of neutralisation was sometimes found necessary in IF stages.
One of my (admittedly arbitrary) markers of a "good" radio is two or more IF stages, relaxing stage gain and AGC range requirements, but there aren't so many "consumer" AM radios that offered this. The pre-war/wartime German military had a thing about using as few different valve types as possible in equipment, and this can lead to some off-beat/ingenious/possibly compromising valve deployments, such as pentode used as det/AGC/AF amp triode. |
13th Jan 2017, 1:58 pm | #36 | |
Rest in Peace
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
|
Re: Valves in commercial applications for which they were not specifically designed.
Quote:
So perhaps the extra winding on the IFTx connected to G3 gives higher gain for the IF stage? Even if this is the case there must be more too it, otherwise why employ a separate winding? Perhaps it is a way of shaping the bandpass characteristic of the IF stage? |
|
13th Jan 2017, 3:58 pm | #37 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: Valves in commercial applications for which they were not specifically designed.
I would suspect that this connection introduces a little positive feedback (reaction) into the stage, increasing gain and narrowing the IF bandwidth. Unfortunately, gm (g3 - a) is not characterised for most receiving valves; the EF50 was, however.
Leon. |
13th Jan 2017, 6:02 pm | #38 |
Rest in Peace
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
|
Re: Valves in commercial applications for which they were not specifically designed.
I'm not convinced (yet) by the positive feedback argument Leon.
There are 2 connections from the IFTx to the pentode. The first is to g1 which is an input (obviously) and the second is to g3 which is also an input (but less obviously). I would have thought that for feedback to occur, be it positive or negative, an input and an output are required. With two inputs you just have addition (or subtraction depending on phase) of the two signals whose effect must take into account the values of gm associated with g1 and g3. If there is any positive feedback taking place perhaps g3 is becoming an output via the anode to g3 capacitance of the pentode? BTW I am aware that this part of the discussion is taking the thread away from my original topic but I'm happy with that so long as its meets with the approval of the moderators. Last edited by ukcol; 13th Jan 2017 at 6:08 pm. Reason: Added last sentance |
13th Jan 2017, 6:22 pm | #39 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: Valves in commercial applications for which they were not specifically designed.
I'd reckon that the (invisible) output is Cga in the pentode, feeding back to the control grid. And the tertiary winding, connected to the suppressor grid, is phase connected such that the signal fed back via the (larger) anode-suppressor capacitance is in antiphase to the signal fed back via Cga. Having dots on the IFT diagram would have helped...
It's most unusual, and as such meets the requirements of this thread... Unless it's in a manufacturer's application note! |
13th Jan 2017, 6:51 pm | #40 |
Rest in Peace
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
|
Re: Valves in commercial applications for which they were not specifically designed.
Isn't it more likely that any output will be via the (as you say) larger anode to g3 capacitance to the tertiary winding of the IFTx?
Apart from the fact that the value of Cga will be much lower due to g1's greater distance from the anode, surely this path is blocked by the screening effect of g2 which is decoupled to chassis potential at RF? |