Friday, May 4, 2018

Why are there potential differences between electrodes of electron tubes even when nothing is connected to them



A Telefunken EF183 vari μ pentode electron tube was used.


A 10ΜΩ input impedance digital voltmeter (YFE YF-3200 Taiwan) was connected to various electrodes through small length aligator cable clips. The input impedance is 1000MΩ.


This experiment was performed under the sun.


The black (negative) lead was connected to cathode, and the plus lead to control grid.


A potential difference of more than 300mV was measured. The time constant using the 10MΩ voltmeter is quite big. It takes a few seconds for the voltmeter to show the max value.


The max value approaching the inter electrode potential difference was changing as clouds appeared under the sun.


When the setup was brought indoors a typical reading was 20mV. (Later addition: even indoors a reading of more than 100mV can be taken after a few minutes. This apparatus behaves a bit like a photon counter).


When the setup was in a dark room, a typical reading was only 1 mV.


The grid was always more possitive than the cathode.


A potential difference could be observed between most of the electrode combinations, possibly between all of them.


The smallest time constant and higher potential difference was possibly observed by connecting the voltmeter between cathode and anode. Anode was more positive than cathode.


Are the electrodes loosing electrons (photoelectrons) as photons from the sun strike them?


Would a Pleiades V6 (3.9Volt at anode) pre preamplifier still operate without the high Megohm resistor between anode and grid to reduce the negative grid potential between grid and the cathode when the cathode is heated by the heater current and losses electrons to the electron cloud?


Perhaps this effect explains the change of gain, if memory is correct, on the Pleiades V(-1) preamplifier, UCC85 or ECC82 log tailed pair when lights where tuned on. (The secondary of the input transformer was connected to each grid so both grids were floating. The amplifier was operating with a small anode voltage, 12V for the ECC82. Can this effect be used for automatic gain control? This attack time if grids are free or unreturned must be very low. Only a few electrons are needed to charge such a small grid capacitance It is remembered that the lights turned on were fluorescent. As they were flicking at turn on a corresponding sound could be heard reproduced. Assuming this is not due to electromagnetic interference. This was with UCC85 operated at 48V) . All safety precautions should be taken with any voltage.)


As sun is setting and light fades out at the Pleiades lab the electron tube, cathode and anode, connected to the hi Z digital voltmeter setup can be seen resting on top of an EKO CL-61 Italian electronic organ.


Now when the digital volt,enter is turned on a slightly negative voltage can be seen to increase. Typically Only a few millivolts.


The setup is lifted and approached to the window and the voltage becomes again positive and typically 40mV.


Returning on top of the organ lid. After sometime the digital multimeter is turned on and the reading is again slightly negative. The rotary dial is set to ohms measurement. This would send a voltage to the electron tubes terminals so at their capacitance will be charged .


Indeed when returning to millivolts measurment a reading of a 3mV millivolts can be observed. The reading stays almost steady. The capacitance keeps the voltage constant and the 10MΩ load impedance is very slowly discharging it.


Now a led torch is shined on the electron tube and the potential difference increases to 10mV. This voltage again remains very slowly decreasing.


The photoelectric effect during daylight time was obscuring the more subtle capacitance charging effect.


It is getting late. It is now about an hour after sunset in Athens and it was decided to continue playing taking advantage of the much less ambient light. I like Chopin - Gazebo is playing on the Pleiades 2N3053 Philips AD5046N setup.


It is time to connect a heater voltage. For safety a fuse in series with battery should always be used.


The minus lead of the voltmeter is connected to the cathode of the EF183, the plus lead to the anode.


A LED torch was shinning next to the electron tube until the voltmeter showed about 20mV. The torch was taken away. The reading was near 15mV and very slowly decreasing.


Then 5V were connected to the pins 3,4 ie the heaters.


As soon as the temperature started to further increase, after about 10 or more seconds the voltage was starting to quickly drop. (Is this because the cathode was starting to emit electrons and was becoming positive due to lack of them?). 0 was reached and then the reading was going negative to about -40mV?.


It was decided to switch off the heaters to see what would happen then. The batteries are too stiff on the battery holder and none could be removed. All were tried and time was spent not looking at the reading in the darkened room. The heaters are still on and now the reading is 40mv positive?! This cannot be explained. Did the potential difference difference started to increase while the electron tube was further warming up? (Later addition, this is a systematic error, bad battery holder contacts, as soon as force was applied to the batter is contact resistance changed and with it the cathode temperature? What is working is that outing force in the batteries the interelectrode charge readings change abruptly).


The heaters are finally turned off by disconnecting a pin aligator cable clip.


The voltmeter terminals are short circuited to get back to 0 volts. We had a capacitor charged by the effect of the heater voltage.


Now the heaters are reconnected, the cathode to anode connected voltmeter shows a positive voltage. But when the torch shines on the electron tube the interelectrode voltage becomes -40mV and decreasing. Ie exactly the opposite of what was happening when a heater voltage was not connected. Why?
Is this why all electrodes become negative relative to cathode when heaters are on? For example the grid bias becoming negative when heater voltage is supplied. Is this because all these experiments were carried at daylight? Would no anode to grid resistor be needed as on Pleiades V6 battery powered when the electron tube is inside the octagon box where no photons would find their way to it? Measurement of the potential difference between cathode and control grid has not yet be done in dark with parameters cathode temperature and light striking the electron tube.


It was decided to continue playing. The voltmeter was connected to cathode and control grid. The potential differenc is nearly 0V. Then the torch light is on the electron tube. A 20mV potential can be observed. Grid positive wrt cathode. Then a 5V heater voltage is applied to the heaters. The potential difference start to decrease and gradually reaches -700mV. The grid is negative with respect to cathode. Or the cathode positive wrt control grid by 700mV. Shinning light or not did not make any difference save for perhaps reducing this by perhaps 20mV. It again can be seen how negative the grid becomes by just connecting a heater voltage to the electron tube. Hence the need of the Pleiades bias from anode to grid by an external high Megohm resistor. This makes the operation of the Pleiades V6 microphone pre preamplifier or other amplifiers possible with just 4 volts at the anode.


But why is the anode mV potential so much affected in polarity by the torch light?


No comments:

Post a Comment