***Phil replied to my article........
There is an artical posted here about tubed phase shift oscilators at : -
http://www.rfcafe.com/references/pop...lectronics.htm
Its an old 1958 article with a hell of a lot of info to digest because
unlike so many modern descriptions of anything technical, the old guys could
say a lot, and not leave out very much. I think my father's generation were
better communicators than many of today's website creators who just post a
stupid brief plagerised picture of something, then think they are God's gift
to the Universe of Information.
But I digress.
I do have a lot of 3 gang tuning caps from old radios which will never be
used for radios again, and so its possible for me to mount maybe 8 of them
in a box with so I'd get 3 caps with variable range 8 x ( 45 - 450pF ), ie,
360pF to 3,600pF.
Has anyone else built a decent phase shift oscillator with wide range of F?
** Not me.
But I do have an all valve Wien Bridge oscillator using a dual gang
condenser, taken from a 1940s radio. Owned it since I was 15 years old.
Tube line up: 2 x Mullard 6AM6, 1 x AWV 6BQ5 & 1 x Miniwatt 6V4
rectifier.
Three ranges: 9 to 218 Hz, 185 to 2158Hz & 1889 to 21,899Hz.
Two 18 megohm resistors used in the low frequency range.
The 6AM6s are wired as triodes.
***Yes, sounds like an AWA standard telephone tech's oscillator used by the PMG.
That was my first WB oscilator, with 20Hz to 20kHz, 3 ranges, and PP transformer at output from a 6SN7.
***I gutted it and rebuilt it with TWO 2 gang tuners, and made it go from about 2Hz to 200kHz. WB Circuit uses 6DJ8 diff input pair, 6CA7 gain pentode, then 6CM5follower OP stage. Goes well, and with series grain of wheat lamps, 'cos the thermistor was stuffed, along with the 3 pos switch when I examined it.
Stabilisation is by an R53 NTC vacuum bead thermistor.
The condenser is electrically isolated from the chassis and lives in a steel
box of its own for shielding.
***Common practice, because the cap chassis is NOT at 0V potential.
The shaft is driven by a backlash free, 60:1 worm reduction drive and a 200
point scale. Verrry silky.
***I have numerous old drives from that era. Some are works of art, others are cheap crap, from the days when we made things in Austraya, and hadn't yet gorn astray.
Once warmed up, it is remarkably stable. THD is a modest 0.2%, pure 2nd harmonic.
Gave it a major renovation about 10 years back - including MF resistors,
new tag board, all electros and many poly film caps etc.
Nostalgia is a real bitch.
***Ah, but we can learn from our father's generation. We can do better than they could. Much of the modern world disgusts me, and the gadgetry invented daily does not seem to make anyone much happier, so why be a slave to something like a mobile phone with 2 million transistors inside, so you yabbayabba to ppl who yababababa back, mostly about nothing meaningful.
***I quite like overcoming the shortcomings of the previous generations. They mostly did the best that they were allowed to do with available time and available wages from penny pinching bosses. Real good tubed test gear which is far more complex and difficult to understand was mostly made in USA by HP and only affordable by the rich, or military / govts. Its somewhat hard to feel nostalgia for my HP 606 HF oscilator 6 ranges from 50kHz to 65MHz, but it uses tuning gangs, by far the best I've ever clapped eyes on, and coils, and a PP oscilator, full AM mod. Somebody gave me a pile of such junk which is too good to strip down and maul, and the 606 is quit handy, although huge, about 2 cubic feet. The one I got had a few minor problems and a few crook tubes but once I fixed all faults it runs fine. Warms the shed a bit in winter.
***After reading the 1958 article on phase shift oscilator, I made one up, with 6BX6 / EF80 and one 6CM5 output buffer. It oscillated, but many problems occurred which the 1958 authors didn't dare mention. In fact, I take bak my praise for those authors because in fact very little analysis of understanding is offered at all, just like the ****wits you see after Googling "phase shift oscillators" There is not one single schematic with all working voltages at every electrode shown, but just blather without depth.
***For wide F range between 2Hz and perhaps 2MHz, maybe 3 parallel 3-gang caps are needed to life the C value so the R need not be too high for LF, while also giving low R value at HF not lower than say 820r so that the phase network isn't a horrible load to drive, even with tubes. I've converted a 1950 radio chassis to oscillator, with the phase network driven with 1.2 6DJ8 CF, and then its output feeding 1/2 6DJ8 CFF, so the network is buffered at both ends, and the 6BX6 can drive it effortlessly. The FB network needed is a combination of NFB and PFB using 10k0 plus 1k0 pot in what is a shunt network with oppositely phase signals at both ends, so that just the right amount of drive goes to the 6BX6, which has gain = 47. BUT, it remains to be seen what limits the Vo level and perhaps I have a use for the spare working P54 thermistor I have. I've never owned a phase shift oscillator, and the idea appeals, and I have lots of tuning caps. Its possible to have 4 x RC sections in the network and some designs use a follower buffer after each RC, so that the difference between network Vin and Vo is much less, so less amp gain is needed.
***I searched online for graphs of typical F response for 3 casacaded RC sections and found nobody offered a decent graph, so I made up a couple of networks and plotted the response to find where the phase shift = 180 degrees, and what the ratio was for Vin / Vo. Its not exactly 29.
Anyway, I build gear based on what I find and proove to be true rather than build junk based on inadequate ramblings of online junk-minded idiots. There are usually 200+ things to sort out when starting from scratch to design and build something and unless you sort out each issue, it just won't work properly.
***I'll try to read all the URLs ppl have posted up about such old gear. There's often some trick or two or three that you find lurking in the schematics that one would never think of by oneself.