Heaters & Filaments: AC vs. DC

[Bigun]

No need for DC heating unless it's a design that handles very low level signals and is noise sensitive, so DC might be useful for small signal, high gain pents or cascodes.

I have some 6C4C tubes (Russian version of 6A3) and I've read that at this voltage hum is a problem compared with the 2A3.

I've also read that the filament design is such that they are susceptible to hum more than 6A3.

Generally people have been challenged to null out the hum on this tube when used with moderately sensitive speakers.

So perhaps it depends on the details of the design ?

 

[popilin]

I have some 6C4C tubes (Russian version of 6A3) and I've read that at this voltage hum is a problem compared with the 2A3.

I've also read that the filament design is such that they are susceptible to hum more than 6A3.

Generally people have been challenged to null out the hum on this tube when used with moderately sensitive speakers.

So perhaps it depends on the details of the design ?

Hi Bigun
Good to see you around here.
If does not bother you, as nobody answered, I'll do.
The valves that you mention are DHT.
As I have been saying, if you put AC on the filaments, I guarantee you'll have hum.
If you use less sensitive speakers, you'll hear less hum, but still there.

In post # 18 you suggest a bright idea, and even they made fun of you.
As I think and read very slowly, I just saw your post in post # 57 and recognized the importance of your idea.

I humbly suggest that you put into practice.

Best regards
Johann

 

[popilin]

Hi nazaroo
Sorry to answer you just now, but I think slowly and you write a lot, really a lot.
BTW, how long it survives your keyboard? Or use a voice command, or something?

I'll skip the many places where you insulted me, and of all the nonsense you said, I'll just quote one.

(a) Maxwell's equations do not make assertions about cause and effect, but only state relations between parameters (incorrectly in fact, because they don't take into account the velocity of electromagnetic radiation, or Newton's concept of Absolute Space. All this was replaced by Einstein's concept of a limit to the speed of light, and indeed all matter, and a the second idea of Space-time, in which the two form parts of a single geodesic field-structure.)

Your insults do not bother me, what bothers me you do not have the necessary foundations to criticize Maxwell.

Maxwell was not a kind of librarian one day said: well, let's mix the laws of Gauss, Ampere and Faraday and see what happens.
I see that nobody told you about the displacement current.

He was a true scientist, who synthesized the knowledge of the time on a set of four consistent equations and two constitutive relations.
Maxwell's equations are already in the whole thread, if you look at least once in your life, the letter c is the speed of light in vacuum.

I can see that nobody told you that at the time of Maxwell, Newton's concept of Absolute Space and Absolute Time, was the only one that had.
Einstein appeared almost forty years later, not even read about history?
However, Maxwell's equations are Lorentz covariant, but it seems that nobody told you.

In electromagnetic radiation is meaningless to talk about cause and effect, because then and only then, Maxwell's equations are written in another way: no sources.
If you put Maxwell's equations in tensor form, and put a light cone, subsequent reasoning surely destroy cause-effect relation, but I remind you that Maxwell's equations belong to Classical Electrodynamics, put them out of context would be an outrage.
A theory that does not resist the experiment should be discarded, Maxwell's equations are checked from 19th century until now, nobody told you?

If you talk about Einstein, before wash your mouth and fingers.
If with "and a the second idea of Space-time, in which the two form parts of a single geodesic field-structure" you mean Curved Space-Time, no need to be as pompous and garbled.
My English sucks, you're right, but apparently you're not Shakespeare.
The more recent measurements shows that Space-Time is Flat (locally curved but globally flat), as Einstein wanted !!!
But it seems that nobody told you.

It's amazing how much nonsense you can say in just six lines, I do not know if I can continue reading.


As you can see, I am still learning to use Google Translator, I have no money to pay for an English teacher to correct my post, as you suggested.
I'm just a TV repairman.
If you are a physicist you still hide it very well.

Best regards
Johann

 

[trobbins]

With respect to using DC heater for DHT in a 2A3, a quick look at a valve datasheet indicates that a 2.5V variation in effective grid-cathode voltage along a 2500 ohm resistive load line will cause about 5% variation in the anode current at 250V mid-point. This indicates about 2.5% increase/decrease at the ends of the filament in emitted current. Given the life degradation mechanisms of the filament, this variation would most likely be negligible in terms of service life reduction of the filament.

Operating with a 6.3V filament would be pro-rata variation, but again most likely negligible impact on service life.

Note that it would be unlikely that an AC heater is voltage regulated, or even adjusted for nominal voltage at the most common mains voltage that the user experiences; whereas a DC supply would be much more likely to be 'spot on'.

 

[Chris Hornbeck]

I thought these discussions were moderated.

Chris

 

[popilin]

With respect to using DC heater for DHT in a 2A3, a quick look at a valve datasheet indicates that a 2.5V variation in effective grid-cathode voltage along a 2500 ohm resistive load line will cause about 5% variation in the anode current at 250V mid-point. This indicates about 2.5% increase/decrease at the ends of the filament in emitted current. Given the life degradation mechanisms of the filament, this variation would most likely be negligible in terms of service life reduction of the filament.

Operating with a 6.3V filament would be pro-rata variation, but again most likely negligible impact on service life.

Note that it would be unlikely that an AC heater is voltage regulated, or even adjusted for nominal voltage at the most common mains voltage that the user experiences; whereas a DC supply would be much more likely to be 'spot on'.

Hooray! Very good what yours, finally we agree.
Then Bigun does not have to change the polarity of the filament power supply.

What the data you have about 211/VT4C ?, time ago I designed a SE amp and reverse the polarity sounds to me uncomfortable.

Best regards
Johann

 

[MelB]

With respect to using DC heater for DHT in a 2A3, a quick look at a valve datasheet indicates that a 2.5V variation in effective grid-cathode voltage along a 2500 ohm resistive load line will cause about 5% variation in the anode current at 250V mid-point. This indicates about 2.5% increase/decrease at the ends of the filament in emitted current. Given the life degradation mechanisms of the filament, this variation would most likely be negligible in terms of service life reduction of the filament.

Operating with a 6.3V filament would be pro-rata variation, but again most likely negligible impact on service life.

Note that it would be unlikely that an AC heater is voltage regulated, or even adjusted for nominal voltage at the most common mains voltage that the user experiences; whereas a DC supply would be much more likely to be 'spot on'.

Bravo! Now why does a tube heated with AC produce that annoying humm? And why are tubes with lower filament voltage less hummy all other things being equal? Hummm..........

 

[popilin]

Bravo! Now why does a tube heated with AC produce that annoying humm? And why are tubes with lower filament voltage less hummy all other things being equal? Hummm..........

This is because

For an Ideal loop where an AC current i(AC) flows with a voltage U(AC), from Maxwell's equations, the AC magnetic field is given by

B(AC) = [ U(AC) x 10^8 ] / [ sqr (2) * (Pi) * S * f ]

At least Tim got it right, or perhaps already knew and was kidding.
BTW. That makes me happy because I feel useful to help others.
Your TV repairman can serve for something, right?

Best regards
Johann

 

[Jaimo]

AC has advantages - lets work with AC

AC +
Since a filament is effectively a resistor, the resulting voltage drop across a DC powered filament will cause one side of the filament to generate more heat than the other and this will result in more electrons being emitted from the warmer side of the Cathode, reducing tube efficiency.

AC -
In DHT's, AC powered filaments can cause hum - but that's because the AC supply is in the audio range. - Solution - create a high frequency AC heater circuit that is outside the audio band and the AC/DC war can be settled!

Link- Pete Millett's HF Filament PSU
HF filament supply

 

[DF96]

Since a filament is effectively a resistor, the resulting voltage drop across a DC powered filament will cause one side of the filament to generate more heat than the other and this will result in more electrons being emitted from the warmer side of the Cathode, reducing tube efficiency.

Eh? The voltage drop might affect cathode current in a directly-heated cathode, but it won't affect heat. If you feed a bulb from DC (as in a torch/flashlight) does one end of the bulb get hotter? Or have I misunderstood you?

 

[trobbins]

I think Jaimo has applied Maxwells field theory and assumed one end of the filament is at zero potential, and the other end must then be the 'hot' end, and ipso facto...

 

[DF96]

So he thinks that higher potential means more heat? I wondered if that is what he thinks. Not so. Heat is caused by potential drop, not potential.

 

[Jaimo]

My bad- it's voltage drop and not heat that's the issue

I pulled out my NEETS reference and it seems that I misunderstood the paragraph on ac vs dc heaters when i read it some time ago. The filament voltage drop results in one side of the DH cathode being more negative than the other and thus emitting more electrons. This is not an issue in ac heated circuits. Here's the NEETS link - refer to page 1-9

The only Maxwell I am interested in has a silver hammer and was last seen majoring in medicine!

http://http://www.hnsa.org/doc/neets/mod06.pdf

 

[DavesNotHere]

The filament voltage drop results in one side of the DH cathode being more negative than the other and thus emitting more electrons. This is not an issue in ac heated circuits.

thats only true if the heater supply is dc referenced (one side tied to ground) but not in AC referenced DC supply ( one side tied to ground through a .1uf cap)

Rfk is AC resistance not DC resistance ( contrary to popular belief ).

 

[DF96]

No. Rfk is resistance, DC resistance. 'Popular belief' happens to be correct, although sometimes forgotten. When a datasheet says no more than 20k between heater and cathode it means precisely that: it calls it R (Resistance) and measures it in ohms. If it meant capacitance it would say so and call it C. Heater supplies should always have a DC reference.

 

[DavesNotHere]

why don't you build and compare. proof is in the pudding.

 

[DF96]

Why don't you read and understand data sheets? Rkf is a resistance; the clue is given by the R. Any floating electrode in a valve is likely to pick up an unwanted potential and disrupt the valve operation. Relying on capacitor or transformer insulation leakage to set heater reference potential is just bad engineering.

 

[popilin]

AC +
Since a filament is effectively a resistor, the resulting voltage drop across a DC powered filament will cause one side of the filament to generate more heat than the other and this will result in more electrons being emitted from the warmer side of the Cathode, reducing tube efficiency.

Hi Jaimo
Is not my intention to offend you, you've clarified in following post, I quote only to fix ideas.
In the case of a DC powered filament/cathode, what I said, in a simple and crude, is that emission is not homogeneous because the filament/cathode is not equipotential.
In further analysis, we saw how emission affects valve life.
In both cases it is implicit that the filament temperature is constant and homogeneous.

Combining Quantum Mechanics, Classical Electrodynamics and Thermodynamics would be a real nightmare, please do not talk about heat.

AC -
In DHT's, AC powered filaments can cause hum - but that's because the AC supply is in the audio range. - Solution - create a high frequency AC heater circuit that is outside the audio band and the AC/DC war can be settled!

Link- Pete Millett's HF Filament PSU
HF filament supply

The analysis that follows is even more gross and make assumptions that may seem somewhat strange, in order to achieve some useful results.

As mentioned
For an Ideal loop where an AC current i(AC) flows with a voltage U(AC), from Maxwell's equations, the AC magnetic field is given by

B(AC) = [ U(AC) x 10^8 ] / [ sqr (2) * (Pi) * S * f ] .....(1)

Now suppose a filament with N turns, the magnetic field inside the filament, in a crude approximation is

B(fil) ~ B(AC) / N .....(2)

Because the grid is close to the filament in a rough approximation, the grid also "sees" the same magnetic field.

B(grid) ~ B(AC) / N .....(3)

Assuming now that the grid has n turns, after some manipulation, the voltage induced in the grid is of the form

U(grid) ~ (n/N) U(AC) = (n/N) U(fil) .....(4)

Because one end of the grid is not connected, it was crucial the result obtained in post # 36.

Ofcourse, a real grid, only "seems" to have turns, in fact it is almost equipotential, again, using the third Maxwell equation, plus the fact

E = - grad (φ) .....(5)

What we would obtain, in a crude approximation

U(grid) ~ (k/N) U(fil) .....(6)

The previous reasoning is nothing formal, is rough and tricky, just to show that within our beloved valves, is like if we have a transformer that connects the filament to the grid, see (4) and (6)
IMHO. AC powered filaments definitely is not a good idea.
Nor count on me to put an AC voltage of 100Khz, I am afraid of intermodulation, but, I do not know much about Electronics, I'm just a TV repairman, in this discussion is someone who knows a lot and I give the floor, please follow you DF96.

Best regards
Johann

 

[MelB]

Popilin: Most grids are at pretty much 90 degrees to the filament in DHT's. Are you still going to get any significant voltage induction?
(Well except for where the grid turns and comes back the other way...but then the opposite thing is happening on the other side...hummm)
Still in most designs the effect will actually multiply!!! (Both sides of the grid wire where it turns and comes back the other way are alined to induce a voltage into the grid in the same direction)
Ya ...AC is a bad idea in a DHT.
I have to be honest I never thought of the AC magnetic field introducing a voltage into the grid, but the way most filaments and grids are configured.....it most likely does!
(Note: These are just my thoughts, I am however in agreement with Popilin and his conclusion with regards to a voltage induction in the filament)
Further discussion please..........

 

[trobbins]

I suggest the outcome is that AC current and AC voltage induction effects both occur, but with careful layout external to the valve then the total combined effects are as most people have achieved - pretty much negligible.

So for those who are good at carefull layout, then AC is "definitely a good idea" imho. If you are a sloppy constructor, or don't appreciate how best to minimise the affects, then by all means change to DC powering and possibly jump from the frying pan in to another frying pan (so to speak).