I assume you are talking about a vacuum fluorescent display? (VFD) A little off topic, but these are definitely best heated by AC, and as you have discovered don't work properly on DC. In most consumer grade stuff there is just a 3V filament winding on the power transformer to take care of this task.
Kevin
Kevin
One old trick is to light a light bulb via induction from a Tesla coil. It makes for a pretty spiffy presentation, however, the filaments never last too long. Magnetic effects can cause mechanical vibrations within the filament. Though not harmful at normal powerline frequencies, at 50 -- 100 KHz (typical for Tesla coils) these become destructive. If you look at how the filaments are arranged in CRT power diodes (1B3GT for example), these are always straight lengths of oxide coated wire. This not only holds down the inductance so that enough current can flow with reasonable filament voltages, but also prevent destructive vibrations.
Interesting idea about using a high freq AC for the filament. I wonder how high you can go before it causes damage.
Anyone try this on audio tubes?
Yagoolar,
I tried to read those articles last night. The one on lowering the filament noise had me stumped with the two drawings. I don't understand why he put the cathode resistor and bypass cap on the filament supply center tap instead of on the cathode of the tube.
I've seen the idea of bypassing the electrolytic cap with a smaller film cap done on the CMOY headphone amp for the power input.
The other articles introduce a feedback loop, not in the classical sense though. Use the noise to cancel the other noise. I think I'd prefer to remove the noise instead, so I'd probably go for the idea in the first article, once I figure it out!
Anyone try this on audio tubes?
Yagoolar,
I tried to read those articles last night. The one on lowering the filament noise had me stumped with the two drawings. I don't understand why he put the cathode resistor and bypass cap on the filament supply center tap instead of on the cathode of the tube.
I've seen the idea of bypassing the electrolytic cap with a smaller film cap done on the CMOY headphone amp for the power input.
The other articles introduce a feedback loop, not in the classical sense though. Use the noise to cancel the other noise. I think I'd prefer to remove the noise instead, so I'd probably go for the idea in the first article, once I figure it out!
kusojiji said:
I would be cautious about that high frequency in audio because you may cancel hum but introduce RF, especially if you use square wave. Wires act as antennas then
. That practice is common in powering Roentgen tubes filaments, so it is easy to go into standby by reducing the width of pulse and then quickly go back into operation. That is also designed for filament protection - one never has cold filament. But I would not use those tubes for audio 
Which drawing have you been referring to?
That is not about filament induced hum, if I understand your words properly.
If I recall correctly this issue of high frequency ac versus dc heating was handled thoroughily in a back issue of VTV. (Disclosure: Vacuum Tube Valley, I am not associated with them, but have written for them in the past and may do so again in the future.) The conclusion was that the benefits of ac heating were not discernable in comparison to properly implemented dc heating. The problems they encountered with uneven filament heating amongst other things was interesting. They used an RF frequency of 2MHz I think, and used a lot of shielding. The experiment I think was well conducted and the article well written.
I have used both low frequency AC heating and DC and to be honest don't hear any difference other than a somewhat lower noise floor for DC heating. I generally use current sources to heat my 300B's and usually include some 0.47mH - 1mH chokes in series with the filament lines where they enter the amplifier chassis. (I use 2 chassis designs with supplies remote to amplifier.) Cathode return is usually by a pair of 25.5 ohm resistors to a 10 ohm current sampling resistor to ground. I compensate for the 100mA of resistor current in the design of the current source.
The Ronan board is popular, but frankly the additional voltage regulators are not required if you use sufficient filter capacitance in the supply. Take a look at National or Linear Tech app notes for how to configure. I use LT1085's.
AC heating works well at filament voltages of 2.5V and below (not in pre-amps however), have built several 45 and 2A3 based amplifiers with center tapped filament transformers with no hum issues.
Speakers are >103dBspl @ 1W so any hum is quite audible.. Current amplifier combined hum and broadband noise is <0.5mVpp at the speaker outputs. The hum component is buried in the noise floor.
Kevin
I have used both low frequency AC heating and DC and to be honest don't hear any difference other than a somewhat lower noise floor for DC heating. I generally use current sources to heat my 300B's and usually include some 0.47mH - 1mH chokes in series with the filament lines where they enter the amplifier chassis. (I use 2 chassis designs with supplies remote to amplifier.) Cathode return is usually by a pair of 25.5 ohm resistors to a 10 ohm current sampling resistor to ground. I compensate for the 100mA of resistor current in the design of the current source.
The Ronan board is popular, but frankly the additional voltage regulators are not required if you use sufficient filter capacitance in the supply. Take a look at National or Linear Tech app notes for how to configure. I use LT1085's.
AC heating works well at filament voltages of 2.5V and below (not in pre-amps however), have built several 45 and 2A3 based amplifiers with center tapped filament transformers with no hum issues.
Speakers are >103dBspl @ 1W so any hum is quite audible.. Current amplifier combined hum and broadband noise is <0.5mVpp at the speaker outputs. The hum component is buried in the noise floor.
Kevin
Yagoolar,
The article on Directly Heated Filaments has two circuit diagrams showing the cathode resistor, the bypass cap, and the second bypass cap on the center tap of the transformer. I don't understand why the resistor and caps are shown on the transformer and not the tube.
This article speaks of using a small film cap to help speed up the response of the bypass electrolytic cap. I've seen this applied in articles on headwize.com. I'd give you a link, but the website seems to be down right now.
Kevin,
Thanks for the info. I'll try searching for the article. I hope the circuitry involved is not too complex.
The article on Directly Heated Filaments has two circuit diagrams showing the cathode resistor, the bypass cap, and the second bypass cap on the center tap of the transformer. I don't understand why the resistor and caps are shown on the transformer and not the tube.
This article speaks of using a small film cap to help speed up the response of the bypass electrolytic cap. I've seen this applied in articles on headwize.com. I'd give you a link, but the website seems to be down right now.
Kevin,
Thanks for the info. I'll try searching for the article. I hope the circuitry involved is not too complex.
kusojiji said:
Two things you need to consider. DC current which flows thru both halves of the winding and that is why the bias resistor and bypass cap are connected to the transformer center tap. AC currents flowing in both halves of the transformer is the other issue. They should be balanced in CT point. If they don't - you have hum.
If you don't use CT xformer (as in Steve's article) you should use 2 cathode return resistors. Kevin has pointed that earlier.
That one is obvious. I guess
My apologies, everyone. At first, I couldn't understand why people started giving me info about hum, even after I had reported that I had REDUCED my hum to what it is now. It used to be around 7 to 10 mv!
I started searching around and found that while 5mv was minimal, <1mv was acceptable! What a shock!
Thank you very much for the great info!
I started searching around and found that while 5mv was minimal, <1mv was acceptable! What a shock!
Thank you very much for the great info!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.