• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Universal filament regulator

Disabled Account
Joined 2010
As you say there is a difference with AC and DC heaters, its interesting that the DC components in the supply of the heaters can also make a difference!

This is 50Hz and again the heating and cooling is not the issue.

I think you would have to show a modulated idle current to prove this, I know it shows it in the link. However many here will argue the AC/DC heater issue also. I will look more closley at the link!
Best wishes.

Regards
M. Gregg
 
You are right - and Dmitry Nizhegorodov's work proves it.

His "study" doesn't "prove" anything. Yes. He posted some oscilloscope shots on a website and attached some graphics. But not anything I would consider documentation. It's not even clear what the oscilloscope shots show, what the vertical or horizontal scales are. Gimme a break.

His study shows that even when you null out the fundamental signal (with the usual pot) you get 2nd harmonics of the filament current at the output!

Of course you do. You create an artificial ground reference for the signal that couples capacitively into the tube. It's intuitively obvious that this will make the interference signal twice the frequency at half the amplitude. No surprises here.

Check out his measurements and see for yourself how clearly he proves that this is a modulation effect and not a thermal effect.

Again, he doesn't prove anything. He merely shows poorly documented results of some measurements.

Rod, rather than postulating, hypothesizing, and hand-waving, would you please share YOUR results? Specifically, I would like to see THD+N vs frequency and output power with an "amateurishly implemented DC supply" and your constant current source. If this "current mixing effect" really causes the distortion to skyrocket like you claim, it would be easy for you to measure and show the difference with/without your regulator. Any computer sound card along with free FFT software would be able to pick this up. Without hardcore data, your posts are really just fear mongering and pseudo-science.

I don't understand why you are pushing this issue so hard. Last I posted anything related to DC filament supplies, I recall you did the same thing. Shouldn't it be up to the individual designer which circuit topology he chooses to use? What's your angle there? Just curious...

~Tom
 
His "study" doesn't "prove" anything. Yes. He posted some oscilloscope shots on a website and attached some graphics. But not anything I would consider documentation. It's not even clear what the oscilloscope shots show, what the vertical or horizontal scales are. Gimme a break.

Read what he says carefully, and it makes perfect sense. The scope shots show a residual 2nd harmonic of the filament voltage, and their phase relationship to the incoming filament supply waveform. The output levels are separately graphed, and presented in dBV, and plotted for various frequencies... I have attached the output graphs so you don't have to search for them.

What Dmitry presents us with is a study of output magnitude and phase of DHT amplifiers, when the filament is fed from a selection of frequencies from 20 to 6000Hz.

This is useful - to spell it out - to show that :

- there is a current-modulation effect on the output of DHTs;
- 2nd Harmonics are generated
- the effect is frequency independent.
- the phase relationship is stable and unchanged from the incoming signal.

The last two items are of great interest, since they prove that this is NOT a simple capacitive coupling effect.

Of course you do. You create an artificial ground reference for the signal that couples capacitively into the tube. It's intuitively obvious that this will make the interference signal twice the frequency at half the amplitude. No surprises here.

Intuition proved wrong by the measurements shown!

If it were a cap-coupled signal, the effect would increase with frequency, since the load at the grid is a resistor (50 to 100K).

What's more, the 8pF or so of input capacitance of the 2A3 is over 800Mohm of reactance at 20Hz. The grid has a typically 100K pulldown. And yet -40 to -60dBV of 2nd harmonic was measured at the amplifier output.

I would be interested to see your calculation that shows how such a feeble coupling could give such a large output as the -40dB V measured. For normally configured 2A3 amplifiers you'd get less than -140dB!


Rod, rather than postulating, hypothesizing, and hand-waving, would you please share YOUR results? Specifically, I would like to see THD+N vs frequency and output power with an "amateurishly implemented DC supply" and your constant current source. If this "current mixing effect" really causes the distortion to skyrocket like you claim, it would be easy for you to measure and show the difference with/without your regulator. Any computer sound card along with free FFT software would be able to pick this up. Without hardcore data, your posts are really just fear mongering and pseudo-science.

I don't understand why you are pushing this issue so hard. Last I posted anything related to DC filament supplies, I recall you did the same thing. Shouldn't it be up to the individual designer which circuit topology he chooses to use? What's your angle there? Just curious...

~Tom

NATURALLY, every designer should make his/her own choices about design work.... where did I try to step in the way of choice? But in an open forum, if you present a solution, and label it optimum, you have to be able to accept criticism of it. Please see the difference between criticism and arm-twisting - everyone is free to ignore either of us, if they don't like what we say.

Tom, my Angle is simple. The forum is here to present ideas, debate the origin of problems, and discuss solutions. A big part of that is constructive criticism, and critical analysis. Filament heating is one of my pet topics, and there is no shortage of DIYers here who can vouch for the effectiveness of my solutions. I have offered some criticism of a solution you presented, and you seem to have interpreted it as an attack, and say that you are "not interested" in discussing the cardinal points of my analysis.

As for presenting measurements, what reward would I have for all that considerable labour? I presented Dmitry's measurements, and you tried to misrepresent and belittle them, effectively labelling him a charlatan, despite it being a sincere and useful analysis.
 

Attachments

  • all-tubes-data-plots.gif
    all-tubes-data-plots.gif
    20.5 KB · Views: 708
As you say there is a difference with AC and DC heaters, its interesting that the DC components in the supply of the heaters can also make a difference!

This is 50Hz and again the heating and cooling is not the issue.

I think you would have to show a modulated idle current to prove this, I know it shows it in the link. However many here will argue the AC/DC heater issue also. I will look more closley at the link!
Best wishes.

Regards
M. Gregg

How do you distinguish 2nd order harmonic noise from electrical induced emission from second order harmonic noise from thermal induced emission?

H. J. Van Der Bijl in The "Thermionic Vacuum Tube-Physics and Electronics" clearly explains the relationship between temperature and emissivity of at filament/heater. If the ac power through the the heater/filament does cause thermal changes it will be at a second harmonic of the line frequency and therefore indistinguishable from emissivity changes from second order harmonic emission from voltage e-field changes.

Furthermore, he explains the difference in emissivity from one end to the other of a heated filament/heater, and how a potential difference along the heater effects emissivity along the heater (his analysis was done on an AC signal iirc, but equally applies to DC).

Finally, given that there is a correlation between switcher noise on the filament and noise at the anode, what is the relationship between the 8mV 500KHz-800KHz switcher noise and audible sub-harmonics?

Are there any?
 
Last edited:
How do you distinguish 2nd order harmonic noise from electrical induced emission from second order harmonic noise from thermal induced emission?

This question is addressed by Dmitry's tests: the thermal effects will be subject to a thermal time-constant, so increasing the frequency should decrease the effect. The results of his test were that the output was constant, regardless of frequency - which suggests zero thermal effect, 100% electrical modulation.

H. J. Van Der Bijl in The "Thermionic Vacuum Tube-Physics and Electronics" clearly explains the relationship between temperature and emissivity of at filament/heater. If the ac power through the the heater/filament does cause thermal changes it will be at a second harmonic of the line frequency and therefore indistinguishable from emissivity changes from second order harmonic emission from voltage e-field changes.

Furthermore, he explains the difference in emissivity from one end to the other of a heated filament/heater, and how a potential difference along the heater effects emissivity along the heater (his analysis was done on an AC signal iirc, but equally applies to DC).

This is a relevant fact, and contributes to unwanted artefacts in the DHT output.
These add to this conduction-generated noise effects.

Finally, given that there is a correlation between switcher noise on the filament and noise at the anode, what is the relationship between the 4mV 500KHz-800KHz switcher noise and audible sub-harmonics?

Usually, a switching chip responds to changes in the input voltage by altering its PWM duty ratio. If there is noise and ripple in the raw dc supply, the PWM ratio bounces along to this waveform. An amplitude-to-pulsewidth conversion in other words
This means it is delivering energy to the output filter in pulses with low-frequency content, not just the 800kHz carrier frequency. The amount in the residual just depends on the quality of the supply input, and the (analogue) loop response of the switcher.

This is added to other noise leakage paths, eg:
during the ON time portion of the PWM, the input is connected to the output by a low impedance path, allowing other conducted emissions through to the filament in a piece-wise fashion.
 
Because he sells pc boards of his design.

Bingo! I would have preferred if he'd be willing to admit this himself. If his customers can hear a difference between constant voltage and constant current, more power to them (and him). But don't claim there is a *HUGE* quantifiable difference without backing it up with data. That's all I'm asking.

Some people also swear that they can hear a difference when they suspend their speaker cables on myrtlewood stands. Any other kind of wood or other inferior dielectrics just don't sound the same.

~Tom
 
The scope shots show a residual 2nd harmonic of the filament voltage, and their phase relationship to the incoming filament supply waveform.

All I see on that page is the blanket statement that some residual of the heater waveform will be present somewhere in the amplifier. I'm not contesting that. In fact I'm supporting that.

You make the assumption that what he is measuring is due to this "current mixing effect" that you have thought up. Based on my understanding of how vacuum tubes work I just don't see that argument as being rooted in anything that goes on physically in the tube.
Yes, the cathode current provides electrons for the filament coating to release. The filament coating releases electrons because it's heated by the filament current (DC or AC). But the process by which these electrons are released to the space-charge cloud surrounding the cathode is completely random. It's a stochastic process. To argue that you can distinguish one electron from another and follow this one electron as it travels happily from the heater wire to the anode is absurd and not supported by science.

NATURALLY, every designer should make his/her own choices about design work.... where did I try to step in the way of choice? But in an open forum, if you present a solution, and label it optimum, you have to be able to accept criticism of it. Please see the difference between criticism and arm-twisting - everyone is free to ignore either of us, if they don't like what we say.

Rod. At no point did I say my solution was optimum. If I did, please point to it. I am willing to accept criticism if it is rooted in science. I am not willing to accept fear mongering, snake oil, and pseudo-science that is deliberately targeted to drive customers towards one solution over another.

You claimed earlier that a DC heater supply would cause all sorts of nastiness and result in severe degradation of the THD of the amplifier. You mentioned THD in excess of 4 %. Please back this up with data. If you are unwilling to take this data, would you allow me to borrow one of your regulators? I am perfectly willing to take the measurements for you and provide a true A/B comparison.

I'm measuring 0.05~0.06 % THD+N with DC heating at 1 kHz, 1 W into 8 ohm. This is measured on this 300B amplifier speaker output using an HP8903A audio distortion analyzer. Once I get further with this filament heater design, I will be posting THD+N vs frequency and output power just like I have for any of the other amplifier designs I've posted here. I will also be posting complete schematics of the amplifier and filament regulators as I have for my other designs.

Enjoy.

~Tom
 
You make the assumption that what he is measuring is due to this "current mixing effect" that you have thought up. Based on my understanding of how vacuum tubes work I just don't see that argument as being rooted in anything that goes on physically in the tube.
Yes, the cathode current provides electrons for the filament coating to release. The filament coating releases electrons because it's heated by the filament current (DC or AC). But the process by which these electrons are released to the space-charge cloud surrounding the cathode is completely random. It's a stochastic process. To argue that you can distinguish one electron from another and follow this one electron as it travels happily from the heater wire to the anode is absurd and not supported by science.

Why do you think the current mixing occurs in the space-charge? You are imagining it if you think anyone claims that.

Filament current and Anode-cathode current mixing occurs in the substrate wire of the filament, which conforms to Ohms law, just like any other piece of wire.

I have explained it numerous times already, so any interested readers can look further up the thread for the workings of it.
 
Gentlemen, with all due respect...

This thread is about Tom's switcher and I'm very interested in its practical aspects. The theoretical arguments about its merits vs other schemes should be discussed separately in other threads.

Tom, thanks for sharing your work and please continue to share as you finalize your regulator.
 
I'm measuring 0.05~0.06 % THD+N with DC heating at 1 kHz, 1 W into 8 ohm. This is measured on this 300B amplifier speaker output using an HP8903A audio distortion analyzer. Once I get further with this filament heater design, I will be posting THD+N vs frequency and output power just like I have for any of the other amplifier designs I've posted here. I will also be posting complete schematics of the amplifier and filament regulators as I have for my other designs.

Tom, this is the heart of the trouble.

If you want to design amplifiers based on a single parameter - THD, that's your choice, but to me that's no better than snake oil.

Unless you design for a listening experience, the outcome is not fit for purpose.

Single-parameter judgement is suspect at the best of times, but THD?

Most importantly, why bother with DHTs? Why not delete the 300B from the schematic and fit a nice LM3886, or even a class-D switcher. I mean everything sounds the same - but the measurement just gets better!

you could get a 35dB improvement in THD, lower cost, less heat to get rid of... The chip-amp sounds like a clear winner to me.
 
Gentlemen, with all due respect...

This thread is about Tom's switcher and I'm very interested in its practical aspects. The theoretical arguments about its merits vs other schemes should be discussed separately in other threads.

Tom, thanks for sharing your work and please continue to share as you finalize your regulator.

Thank you. I would like to see this thread return to a build thread as well. I will continue to post data, schematics, and such as it becomes available.

~Tom
 
An interesting thread... I'm also in the process of designing a 300B amp using Western Electric (300B) tubes. So this (thread) has been a good read and some food for thought. What I've not seen is a discussion of the various DHT types, i.e., the filament structure, alignment within the tube itself and what coverage the filament has within the confines of the grid/plate structure. Not all (DHTs) share the same filament/cathode construction and as such some heater drive circuits may work better with one type vs another.

As an example, I'll use the 45 triode and then compare the original single plate 2A3. The filament design of the 45 is very simple. It uses a single filament wire which is arranged in a "M" configuration. The two top peaks of the "M" are tethered above the grid/plate structure and the two bottom ends of the "M" plus the center peak (or dip) are tied to wire supports on the bottom of the tube below the grid/plate structure. The filament voltage is applied to the two bottom ends of the "M". This arrangement gives the 45 a symmetrically aligned filament within the grid/plate structure. Also note that less than the entire filament wire is contained within the grid/plate structure, so any calculations and/or theories which consider the entire length of the filament wire and it's associated heating voltage as interacting with the grid/plate structure (current flow) would be inaccurate. With with the 45, you have 4 segments of the filament wire which can interact with the grid/plate structure. As the 45 has a symmetrical filament, using an AC heating voltage "should" be a preferred method, as the the AC signal for heating should cancel itself out as the ends are out of phase with each other. Using a fixed DC balance (as the filament wire should have an even cross-section and linear resistance) is logical and effectively balances the filament as a cathode. A balance pot to null any remaining AC component (which is caused by uneven coating of the filament wire, a slight physical alignment error, etc.) should result in quiet operation, assuming good quality tubes. By contrast, using a DC heating voltage skews the bias towards one end of the filament and any ripple component can not be reduced by the symmetrical filament. In the case of a 45, I would lean towards an AC heater supply as it looks to be a better fit and feasible to obtain quiet operation.

The single plate 2A3 has a different arrangement. The filament is a "series-parallel" arrangement per the RCA manual. It's filament has two "sets" of wires. Each set is arranged to make 4 vertical passes thru the grid/plate structure held by a tension bar across the top with a pair of pusher springs. Note it's not a "M" structure like the 45. The heater voltage is applied to the ends of the wire (like the 45). The two sets are sitting side by side within the grid/plate structure, i.e., one set covers the left side of the grid/plate structure and the other set covers the right side. Once again, the entire run of the filament wires are not contained within the grid/plate structure, so any calculations/theories are as above with the 45. The parallel reference is how the two sets are tied together. Where the two sets meet in the middle of the tube, the ends are tied together as one filament contact. The two outer ends of the filament sets are then tied together as the other filament contact. As a result, it's impossible to heat a SP 2A3 with AC and get anything close to an acceptable output noise level.

The Western Electric 300B has a similar filament arrangement and can not be heated with AC and achieve an acceptable output noise level. So, heating either tube will require a DC filament supply for quiet operation. As the purity of the DC can affect the performance of the tube, attempting to calculate it is difficult, as it's hard to measure what percentage of the filament is within the grid/plate structure. This makes it difficult to accurately calculate distortions, noise, etc. which are the result of filament supply noise. Whether constant current or constant voltage is used, you can probably make a good case for either but results may be more subjective than measurable.

Most designers use some common criteria for measurements and specifications which result in proper performance given a set of conditions. Among these measurements are: frequency response, output power, distortion, signal-to-noise, etc.. For my own amplifier designs, one goal is to achieve a minimum of 80dB signal-to-noise ratio referenced to 1-watt RMS output. Having very low THD, wide frequency response, flat power bandwidth, square wave response, etc. are important as well. Once meeting these specifications, I think it becomes a difficult discussion to quantify what audible distortions are attributable to the filament supply. Note that I'm not tossing stones at anyone's filament supply design, I need to have one at some point as well. The point being that one type of filament supply may not work best with all types and/or brands of DHTs.

Regards, KM
 
For my own amplifier designs, one goal is to achieve a minimum of 80dB signal-to-noise ratio referenced to 1-watt RMS output. Having very low THD, wide frequency response, flat power bandwidth, square wave response, etc. are important as well. Once meeting these specifications, I think it becomes a difficult discussion to quantify what audible distortions are attributable to the filament supply.

That lines up very well with my design goals actually. In addition to the measurements you mention, I usually measure the output noise spectrum of the amplifier with the input grounded. This allows me to detect, quantify, and hopefully minimize any residual hum/ripple/EMI resulting from layout, power supplies, etc.

When I design circuits, I tend to work block by block. I design, build, test, and characterize each block individually. But in the end, it's the system performance that matters. I.e. the performance of the completed amplifier is what matters.

~Tom
 
This little discussion gave me an idea about a different kind of filament supply. Instead of using straight DC or low frequency (50/60Hz) AC how about this.

A split supply of +/-5 volts and a basic mosfet bridge. Now this is just a concept so in reality you'd want a driver chip with some logic so both mosfets can't turn on at the same time. The 7404 is configured as an oscillator and then one gate is used to invert the signal.

I bet this hasn't been tried before :D
 

Attachments

  • dual fil supply.pdf
    37.1 KB · Views: 251