On Tuesday, 12 November 2013 20:40:29 UTC+11, John L Stewart wrote:
Here is something to puzzle over in your quitter moments! Cheers to all, John L Stewart +-------------------------------------------------------------------+ |Filename: Triode UL.jpg | |Download: http://www.audiobanter.com/attachment.php?attachmentid=361| +-------------------------------------------------------------------+ -- John L Stewart
I've had another look at the schematic file in
http://www.audiobanter.com/attachment.php?
The cathodes of "driver triodes" are direct coupled to OP triodes. So therefore we are left to assume the OP triodes are drawing grid current, because if they are not, the driver triodes would have no Ia and no Ik so would not drive anything.
The use of IST is "incidental" and not necessary, and any balanced input signal will drive the circuit. Unfortunately, I see extremely little detail in your schematic, no tube types, no Vdc, no Vac basically no bloomin nuthin..
But lets assume driver triodes could be a pair of 6SN7 parallel wired, we put 15k cathode resistors on to a negative rail, maybe -150Vdc, and maybe they'd be Ik at about 10mA each. Assuming OP triodes have negative bias with no grid current, then Ra looking into anodes of 6SN7 would be Ra + [ (U + 1) x Ra ] = 320k, so that any if OPT tranny UL tap voltage = 100V, then 0.315mA of feb back current flows in 320k, so in fact the cathode drive voltage would have 15k x 0.315mA = 4.7Vac, and if THD = Vd at OPT tap, then 0.047Vd appear at OP grids, and this is amplified say by 3.5, in OP tubes are 300B and error signal at 300B anodes = -0.146Vd, which does very little to reduce the +2 x Vd at anodes. To be useful FB, you need the error signal = maybe -10Vd, so that the THD reduction factor = 0.2.
Pardon me for the boring reasoning, but how come you didn't include a reference to what someone else has thought about the detailed workings? If there is a mistake in my figurings, perhaps you'd like to point them out to me.
Right now now I have completetion work on an oscillator I begun 2 months ago.
I wanted to include a Schmitt Trigger square wave circuit to make a reliable good looking square wave even at 2MHz, the top F for my new wien bridge oscillator. I have tried many types of two transistor schmitt triggers, and seems its very difficult to get an acceptable square wave above 500kHz, for +/- 10pk volts. I made a nice voltage amp with bjts, gain 4, symetrical pnp and npn circuit, only 6 bjts, and easily got 0.3Hz to 12MHz at 7Vrms.
I tried C-mos 4093 and although it has stability and works up to 1MHz, the rise time is slow, ie, the harmonic content of the square wave is no more than 3MHz, and I want 10Mhz.
So I am trying more things, including small package switcing mosfets to get better HF performance that the 4093 data sheets don't mention, - because it is so poor. I figure its easy to amplify a square wave with +/- 2.5Vpk to get the final voltage I want, and to get the F content to survive going through attenuator network and output buffer to get 50 ohms Rout for this gene.
I have to say the variation of F with tuning caps is smooth and jitter free, and so much better than using any pot.
So, I am occupied, while being useless to ppl wanting dirt cheap masterpiece amps. NOW, they all must learn a lot and build their own, because none of them want to pay anyone proper wages for hundreds of hours of work.
Two things in life give it meaning - Usefullness, and Happiness. and the first is not very possible, leaving me only the second issue to get right, and its enough to occupy me.
Maybe none of us ever figure out why Existance exists, or what is Infinity, or Eternity, although we think we know these exist.
Somehow the Speed of light seems very slow. Because how does one end of the Universe ( that we know of, could be many more ) know what's happening at the other end of Universe RIGHT NOW? It implies stuff moves at speed = SOL raised to power of infinity, or some incomprehensible speed. God does not play dice, but whatever God is, then It hides itself easily maybe because our gear is far to slow to detect the really quick stuff going on.
But if I get a really good square wave at 1MHz ( with H up to 10MHz from 2 transistors then I'll let you all know via my cluttered web-site.
Patrick Turner.