I am trying to understand the earlier statement about a 220k current source.
First, where did the 220k originate, what post #?
Was the meaning of the 220k a rating of the current source impedance?
The higher the impedance, the better the current source.
And what was the statement about a 220k current source (impedance?) that was only 1 mA.
That requires a 220k resistor that returns to 220 Volts.
You can have a solid state current source that has a 220k impedance, and only has 10V across it, and that is a 10mA current source. That is all possible at the same time.
That solid state current source is the 'equivalent' of a 220k Ohm high voltage power resistor with a 2,200Volt DC supply.
Of course, the solid state current source is much more efficient than the real 220k resistor and real 2,200V supply.
But the effect during normal operation is just the same.
Just do not use the resistor and high voltage to be a plate load for a 12AT7, unless you always have fully warmed filaments on the 12AT7 before the 2,200V supply is active, or you will have 10mA arcing in the 12AT7.
Likewise, a current sink in the cathode using a 220k high voltage power resistor and 2,200V supply when the tube is cold, would cause similar 10mA arcs.
The solid state current source with a 220V supply will not have such arcing, even if the 12AT7 is cold when the current source is turned on. The current sink would cause arcing if the tube was cold and the filament was returned to ground.
First, where did the 220k originate, what post #?
Was the meaning of the 220k a rating of the current source impedance?
The higher the impedance, the better the current source.
And what was the statement about a 220k current source (impedance?) that was only 1 mA.
That requires a 220k resistor that returns to 220 Volts.
You can have a solid state current source that has a 220k impedance, and only has 10V across it, and that is a 10mA current source. That is all possible at the same time.
That solid state current source is the 'equivalent' of a 220k Ohm high voltage power resistor with a 2,200Volt DC supply.
Of course, the solid state current source is much more efficient than the real 220k resistor and real 2,200V supply.
But the effect during normal operation is just the same.
Just do not use the resistor and high voltage to be a plate load for a 12AT7, unless you always have fully warmed filaments on the 12AT7 before the 2,200V supply is active, or you will have 10mA arcing in the 12AT7.
Likewise, a current sink in the cathode using a 220k high voltage power resistor and 2,200V supply when the tube is cold, would cause similar 10mA arcs.
The solid state current source with a 220V supply will not have such arcing, even if the 12AT7 is cold when the current source is turned on. The current sink would cause arcing if the tube was cold and the filament was returned to ground.
Last edited:
I dont' follow this. Surely if the heater is assumed grounded, and the cathode load is a tuned circuit which is likewise grounded, then Rhk forms a parallel damping resistance. What circuit configuration are you imagining?
I said "added resistance" i.e. resistance from the heater circuit to ground, typically. However, in most RF circuits the heater will be grounded. The cathode will almost certainly have a low resistance to ground, either a bypassed resistor or an inductor. Hence the 20k requirement will be met. Rhk is in parallel with whatever is connected from the cathode to ground. If this is a problem, then just add some capacitance from cathode to ground to swamp Rhk.
Please read the thread starting from post #42 on.
"....lower than1mA " because the max operationvoltage between heater and cathode in some tubes can be as high as 200V (disregarding power tubes for series-regulators the PC92 is a rare exceptions) with allowed peaks of max 315V.
You say it.
So, what was all that "swamping with C" about if it is allready swamped by Rhk?
Why could you not accept the Rhk 20kohm limit set by Philips?
beats me....
I don't understand what it is that you don't understand.
I accept that there has to be a limit on external resistance between heater and cathode, but I believe this is all about setting up bias potentials (so no floating electrodes) and possibly hum in audio circuits, not anything to do with the intrinsic Rhk affecting RF circuits. If that means that I am disagreeing with the manufacturers data sheet then so be it.
I accept that there has to be a limit on external resistance between heater and cathode, but I believe this is all about setting up bias potentials (so no floating electrodes) and possibly hum in audio circuits, not anything to do with the intrinsic Rhk affecting RF circuits. If that means that I am disagreeing with the manufacturers data sheet then so be it.
I think I'm shifting my position. I too now think it has little to do with damping resonance circuits, and is actually all about hum.
A large resistance between cathode and ground (which by implication is between heater and cathode) inevitably invites more hum voltage in an audio amp.
The same is true for an RF circuit, except the 'hum' is not an audible problem but a modulation problem -it modulates the gain of the stage and therefore wobbles the tuning (I guess that would qualify as phase noise if it happens to be an oscillator stage), or creates sidebands.
So I think it's probably the same 'hum leakage' problem either way; the datasheets just put it into AF or RF language to suit the different readerships.
Last edited:
Finally something I thinck we all can agree upon, but I would like to point out that in sensitive measuring equippement there may also heater - cathode "diode-effects" (and not so "linear" and time variant DC-leakage behaviour in general) be of importance.
Dr. Slevogt made me aware of this during the developement of a Dk-meter.
In my youth I was lucky enough to work with top scientists in Germany at the premises of WTW and Fraunhofer Institut.
Dr. Slevogt made me aware of this during the developement of a Dk-meter.
In my youth I was lucky enough to work with top scientists in Germany at the premises of WTW and Fraunhofer Institut.
Last edited:
Can you elaborate on what may be exhibited by "diode-effects" ?
In a typical grounded heater mid-point configuration, and even the very vintage single-ended heater grounding configuration, any portion of the heater 'filament' will cyclically vary from being positive with respect to nearby cathode tube, and then negative. In simple V style heaters, any heater portion also sees its mirror portion of the heater nearby, and the nearby cathode wall may well see both a filament portion that is negative and a portion that is equally positive with respect to the cathode wall. Similarly for the double helix wound heaters, where any portion of heater is much more spatially mirrored by its other 'half'.
In a typical grounded heater mid-point configuration, and even the very vintage single-ended heater grounding configuration, any portion of the heater 'filament' will cyclically vary from being positive with respect to nearby cathode tube, and then negative. In simple V style heaters, any heater portion also sees its mirror portion of the heater nearby, and the nearby cathode wall may well see both a filament portion that is negative and a portion that is equally positive with respect to the cathode wall. Similarly for the double helix wound heaters, where any portion of heater is much more spatially mirrored by its other 'half'.
Last edited:
Suppose Uh 6,3V 50/60Hz midpoint grounding.
One end of the heater is at -4,44Vpk in respect to ground.
Cathode at, let as say, +2V
The hotter (than cathode) heater wire emmits some electrons that will be drawn to the 6,44Vp more positive cathode = "diode" forward current.
At the same time the other end of the heater will be at +4,44Vpk in respect to ground
and electrons emmitted by the heater will be repelled by the 2,44Vpk more negative cathode = "diode" reverse current.
Even isolated "alumina-clad" heaters emmit electrons to the as "anode" functioning inner side of the cathode depending on temperature, + or - voltage diff., impurities and contamination.
That is how I understood the explanation given to me by Dr.Slevogt.
One end of the heater is at -4,44Vpk in respect to ground.
Cathode at, let as say, +2V
The hotter (than cathode) heater wire emmits some electrons that will be drawn to the 6,44Vp more positive cathode = "diode" forward current.
At the same time the other end of the heater will be at +4,44Vpk in respect to ground
and electrons emmitted by the heater will be repelled by the 2,44Vpk more negative cathode = "diode" reverse current.
Even isolated "alumina-clad" heaters emmit electrons to the as "anode" functioning inner side of the cathode depending on temperature, + or - voltage diff., impurities and contamination.
That is how I understood the explanation given to me by Dr.Slevogt.
Last edited:
Correction:
This should read rectified instead of halfwave rectified...
Only when the heater is grounded at one side is it halfwave rectified AC (pulsating direct current)
Measurements of heater to cathode resistance don't show a generic diode characteristic.
Heater-cathode DC conduction resistance
Some additional comments on page 4 of hum article.
https://dalmura.com.au/static/Hum article.pdf
It all gets messy to interpret/measure near 0V difference between heater and cathode, because the heater always has a voltage along its length during normal operation, and the heater to cathode voltage varies at any instant at any point along its length (for AC heater supply), and there are a few ways for heater filaments to be deployed inside a cathode.
But my main observation is that there is no 'diode' characteristic relating to the emission current.
Heater-cathode DC conduction resistance
Some additional comments on page 4 of hum article.
https://dalmura.com.au/static/Hum article.pdf
It all gets messy to interpret/measure near 0V difference between heater and cathode, because the heater always has a voltage along its length during normal operation, and the heater to cathode voltage varies at any instant at any point along its length (for AC heater supply), and there are a few ways for heater filaments to be deployed inside a cathode.
But my main observation is that there is no 'diode' characteristic relating to the emission current.
Last edited:
Please notice that in the paper from Klemperer that has been originally posted by you, the h/k rectifying effect is also mentioned
https://dalmura.com.au/static/Heater cathode insulation performance.pdf
Last edited:
Yeh the paper by Klemperer is pretty amazing - for starters it was prepared way back in 1934-5, and he had the resources of the RCA lab to prepare a wide range of test setups from which to gain a better understanding of indirectly heated cathode-heater insulator operation.
He was able to prepare heaters using abnormal conditions, so as to investigate characteristic operation from results that show obvious extremes. Certainly, distinct diode like behaviours were observed for one extreme (figs 8 and 9). His typical curve 4 is not symmetric, and shows a reasonably abrupt change around 0V difference, which could be described as diode like over the range of a few volts either side of zero, although as a general characteristic it does not show a defined asymmetric blocking character.
My measurements also weren't showing a defined diode like character, and similarly showed a variety of responses. In general they did show the generally advised benefit of operating with a voltage difference magnitude beyond about 10-20V, with no clear generic winner as to positive or negative being better. And similar to microphonic variation between tube samples, the tube to tube variation can easily be at least 1-2 orders of magnitude different with respect to insulation resistance.
He was able to prepare heaters using abnormal conditions, so as to investigate characteristic operation from results that show obvious extremes. Certainly, distinct diode like behaviours were observed for one extreme (figs 8 and 9). His typical curve 4 is not symmetric, and shows a reasonably abrupt change around 0V difference, which could be described as diode like over the range of a few volts either side of zero, although as a general characteristic it does not show a defined asymmetric blocking character.
My measurements also weren't showing a defined diode like character, and similarly showed a variety of responses. In general they did show the generally advised benefit of operating with a voltage difference magnitude beyond about 10-20V, with no clear generic winner as to positive or negative being better. And similar to microphonic variation between tube samples, the tube to tube variation can easily be at least 1-2 orders of magnitude different with respect to insulation resistance.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.