Revisiting EF183 at small supply voltage (Vb=6V battery) for amplifying small mic signals (front end) it was observed that the grid assumes a -0.5V when left alone. This was measured by a high Z voltmeter approximately 10MOhms input Z.
When plate (anode) was disconnected from Vb, Vg was -0.65V !!!
So in the absence of any connection (just normal heater) there is a potential difference of -0.65V between cathode and grid! By connecting a micro ammeter between grid a cathode a current of 100 micro amperes was measured, and this in the absence of any power supply other than heater voltage.
Experiment with a EF183 battery pre preamplifier:
Signal path:
Grampian ribbon mic - Pleiades 2.27mH -2.8H (1:32) proximity compensating transformer (8turns to 249turns on Magnetec 073 core - 22nF coupling capacitor to grid - EF183 (triode connected) Vb=6V 6.8M from Vb to Grid (otherwise the electron tube does not have enough anode (electron) current, (Bravo to Hliana) - Altec 15095 output transformer - Realistic Disco Mixer at mic input - attenuating pad 560ohms etc at output - Sennheiser HD-580 Headphones
Sounded nice but disconnecting heater from its supply made the sensitivity of the microphone increasing! while Ia was decreasing! until the temperature of the cathode was so low that one could hear granular emission noise.
A 22Ohms 2W dropping wire wound variable resistor was connected in series with the heaters
At full heater voltage Ia was 300 micro ampere. Things were starting to get really interesting sound wise at 100- 140uA ! This was when maximum gain was observed by having a 440Hz square Seiko metronome at 440z at a couple of feet from mic.
Vh was measured to be 3.14V !. , heater current 180mA !
At this point subjectively the sound on voice was amazing having the quality of the legendary U47 microphone were genius Georg Neumann was using 34V on the 60V heater of the VF14 electron tube! Great detail on the subtlest nuances of voice.
Amp did not seem to work with R (Vb -G) of 8.2M.(why? did not as a sencond step try to reduce Vh when it was connected? (at later experiments it did, optimum Ia current was found to be 105 micro amperes by letting the heaters to cool down by disconnecting heater supply and observing maximum information heard on headphones.
Next high resistors were connected from Anode to Grid (not Vb to Grid) as this was found to give an even more ethereal subjective quality to sound.
By accident 8.2M was connected, max gain was when Anode current was around 105uA !!, Vh=2.6V !!!, Rh=17.5Ohm
2 stages of EF183 was connected one after the other, the 2nd (Ra=1.5K, Ra-g=3.9M). Sound was still clear so need 3 stages.
While doing these experiments it was observed that when connecting 6.8M from Anode to Grid Vg=-0.5V becomes -0.2V when there is plate supply and -0.4V from 0.65V when there is not (open)
Next day:
It seems that the electron cloud around the cathode behaves as traffic masking signal in an information way, not a noise way!
It had been previously also observed that this -0.5V decreases when heater voltage is dropped.Time constant giving same behaviour as sound increasing!
Only heater voltage was supplied to the EF183. Then the usual -0.8V was observed. Another way of looking at it is the cathode is depleted of electrons and becomes positive or grid absorbs some electrons and becomes negative? A similar voltage somewhat lower was observer on the A, g3, g2 combination.
By connecting a uA meter between grid and cathode with no other supply than the heater voltage a substantial current of 200uA !!!(120uA systematic error?) and a current of 3uA between Anode and cathode.
When the heater was connected to the dropping resistor giving max gain, this grid cathode current was increasingly being reduced ascathode temperature was decreasing and final went to 0!! at a time constant of a about 10 seconds similar to the time constant for observing max signal to noise!
No comments:
Post a Comment