Well, with a series string, you can feed it with a CCS which will eliminate current hogging. It makes for more efficient regulation as well. The question is, at equal currents, how equal is the voltage drop across the filaments? Some tube types are super consistent, some aren't, and I admit to limited experience with the 1V jobbies.
Three terminal regs are absolutely suitable for heater/filament supplies.
Three terminal regs are absolutely suitable for heater/filament supplies.
They were not good enough for Hovland. Bob Hovland has been at it since the late 50's. It's not about data it's about sound. You know this, I'm sure.Well, I guess after 45 years of building and designing this stuff, I'm a beginner.Now if you have actual data showing a problem with 317-type regulators on heater supplies, I'd love to see it.
Well since you won't trust Hovland I guess you will just have to trust little me
. My first tube circuit run in with the little three terminal friends was with a 6072 mic preamp I did to start Requisite Audio. The little three terminal friend did a good job at making a stiff voltage but it made the tubes sound a bit more tilted to SS. The founder of Requisite Audio liked that and it stayed, albiet with a 2ohm resistor added to limit current rush. So if you like a SS blanket for your tubes go with it. It does make things easy! Since the author is concerned with power efficiency why not SMPS? I did it on a guitar amp filament to reduce weight (since you are a musician also I'm sure you understand about lugging around heavy guitar amps). It did work well enough and less energy waisted to boot.
These guys sum up my feelings about filament regulators pretty well http://www.space-tech-lab.com/Q6.html
Last edited:
Not true, I agree with SY on this and so does Rudolf Moers in his new tube book.
I went through several configurations on the filament supply. This circuit is simple and works quite well. I have been using it for quite some time and many, many hours of trouble free operation.
Because it is relatively simple does not mean it is not effective, it controls the current and the voltage very well.
Rob
Posting schematic
AudioXpress has granted permission to air the schematics. My overall decisision is to wait until this isue of aX is off the shelves before I post the schematics here.
I would have been more inclined to post them immediately but a couple of people here accused me of posting for my sole benefit and to sell magazines which I felt was terribly unfair and uncalled for.
They will have to wait for the schematics until the curent issue of aX is off the shelves.
The rest of you can PM me for the schematics and I will post them privately with the condition that you do not post them anywhere without my permission.
Again, when the June issue of aX comes out I will post for all to peruse.
Rob
AudioXpress has granted permission to air the schematics. My overall decisision is to wait until this isue of aX is off the shelves before I post the schematics here.
I would have been more inclined to post them immediately but a couple of people here accused me of posting for my sole benefit and to sell magazines which I felt was terribly unfair and uncalled for.
They will have to wait for the schematics until the curent issue of aX is off the shelves.
The rest of you can PM me for the schematics and I will post them privately with the condition that you do not post them anywhere without my permission.
Again, when the June issue of aX comes out I will post for all to peruse.
Rob
Robert,
I try to make published designs more accessible to readers, certainly experimentation is in order and I'd love to see what you come up with. The LM78xxx's have about a 40mv noise floor if I remember which I think is good enough for filaments although DH might see a lower noise regulator benefit.
Rob
Accessibility is a noble thing. I'm a bit like this guy at times when I read another "Hey look what I built!" article in an AA publication. So much of what is published completely disregards the strides made in the late 80's through the 90's. It's like Audio Amateur is in some sort of dark ages.
It's not just noise but that is a good place to start. I wish I had more time to spend on regulators too. Take a look at an Audio Amateur author who precedes you and read. Also, relating to filaments, think about perveance and emissions (not the gas from a car or a post bean eating person). Nothing happens in a tube until that cloud of electrons is boiled off the filament and once those electrons go to the plate more electrons must be delivered to the cloud or the tube does not amplify. The devil is in the details as to how those electrons get delivered to the filament. Try adding a simple shunt current source pulling even just a few milliamps across ALL regulated supply lines. The LM334 is a good easy way to do this for supply voltages up to 40v and this device can be used to bias an op-amp into class A operation to boot. And IF YOU MUST go to the 3 terminal reg dark side again consider Linear Technologies devices.
I disagree. First, NS regulators work fine, if you want to go to LT that's your decision.
This simple circuit allows the current to build and not exceed the maximum current of the tube and then lock on the 1.4 volts that these filaments require. Electrons flow, I assure you.
I built this, listened to it for over a year, had others listen to it and compared it to other designs of my own along with other builders work. You, on the other hand, looked at a partial schematic.
I'm happy to have readers improve my designs. If you think you have send me the schematic, I will be happy to bread board it and try it.
For future reference, assumptions about my background, hype, silly you tube videos, "dark side" analogies, etc., etc. do nothing to impress me.
Rob
Last edited:
I've looked at the design and don't quite get it!!
You take a high level signal of approx +- 2v, attenuate it though a 100k pot so you can use a 1T4 biased at -0.4volts, grid leak 10M, then Trioded, running at B+ 67volts, pass it through some MHz Opamp, followed by another MHz unity gain Opamp Buffer.
From my understanding, get rid of the 1T4!!
I'm gonna get Backballed again, but I notice SY is a contributer in the same edition
Regards
John
You take a high level signal of approx +- 2v, attenuate it though a 100k pot so you can use a 1T4 biased at -0.4volts, grid leak 10M, then Trioded, running at B+ 67volts, pass it through some MHz Opamp, followed by another MHz unity gain Opamp Buffer.
From my understanding, get rid of the 1T4!!
I'm gonna get Backballed again, but I notice SY is a contributer in the same edition
Regards
John
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.