|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
89.164.186.207
Hi to all tube enthusisasts,
I need a little help understanding the attached circuit. It's supposed to be an excellent sounding high voltage regulator. Can someone please explain how it works. Input from power supply is 310VDC. Power supply is 6x5, C (50uf), L (20H), C (50uf). I'm not sure about the output voltage. I guess it should be arount 90-110V.
It's intended to drive one 6922 tube. If there is a room for improvement, please let me know.
Follow Ups:
What about if resistors are rearranged like this?
If things are arranged like that, the 0B2 doesn't serve any purpose.
I would imagine that the actual circuit uses the gas regulator to set the grid voltage on the 5687, then the regulated B+ can be drawn from the cathode.
This supply is for something which can only impact rather than improve the sound : a tubed SPDIF buffer
May as well build it as-is and use for the intended purpose of being the power supply of a totally useless/worthless gimmick that some audiophools just have to have . I can almost see it now , an IT coupled 300B SPDIF buffer . All you need to do is market it , forget deveopment , just make it look pretty and give it a silly name . You could make a heap of cash...
Al
Well yes it can easily be done, I'd reach for a 12TA7 or 6C4 myself. After all SPDIF at 44/48K is standard NTSC video bandwidth, 6mhz, and tubes were used there for many years even inout old TV sets of the 60s.But why??????
What possible sonic advantage can a tube SPDIF buffer have?
P.S. It would most likely have to be the dreaded cathode follower. Remember that SPDIF uses a 75ohm terminated transmission line. You need to drive a volt into that.
Edits: 02/25/15 02/25/15
I'm not an expert in designing circuits, but I have more than 20 years of experience in the audio hobby. I've owned and listened to the very best CD players and transports (Esoteric, Wadia, CEC, Audio Note...) during that time, and I also find it very hard to believe that this magical Cd performance comes from a fairly simple regulated supply and 6922 spdif buffer as shown at the attached picture. All the other ingredients are off the shelf CD pro2 with cheap chinese controller and clock. If someone has an idea why I'm getting such results, please explain. The sound of this machine gives me an insight into the music similar to vinyl. It's hard to describe, but music is more alive and it flows. No oscilloscope so no way to measure the signal coming in and out of the buffer, but when it sounds like this I find it not too important how the signal looks at the screen.
FWIW, a 75 ohm load really needs a SRPP, not a standard follower. That's the sort of load SRPPs were originally intended to drive.
--------------------------
Buy Chinese. Bury freedom.
x
Oh dear, I hadn't noticed. What a horrible idea!
This (something like this) circuit works. I've heard it and I want it. It's sounds better than any other regulators or stabilizers I've heard so far for this purpose (spdif buffer in a CD transport). The guy who draw me a sketch which I used to draw a schematic was really sloppy so the sketch he draw is hard to read. This guy died, so I can't get the explanation from him.
What if R1 and R2 are used only as a voltage divider to which center we connect 5687 pin 8 (ground center tap). So let's get that out of the schematics for now.
I made a new schematics. R5 says 120-150K at the sketch. Could this work?
Question #1
What voltage output do you want?
Question #2
What tube is V2?
Question #3
Do you want it a variable output supply?
DanL
1. 90-100V
2. OB2
3. Yes
Here you go...
R4 is the voltage adjust pot.
DanL
Thanks a lot. I made this and it works amazing for the purpose. Whatever someone might think, sound quality went up quite a bit with this regulator.
You find a schematic for some obscure tweak or project. You build it and it always sounds great. Been there myself since the 1970s.
Not trying to be sarcastic here but it wouldn't hurt to analyze how the circuit is making the improvements you are hearing.
We actually see and hear with our brain, not our ears and eyes. They just collect the data. It's our mind that makes that data good or bad.
Yes, this version of the basic circuit will work.
With a fixed voltage (to ground) at the grid of the 5687, the output impedance of the regulator will be approximately 1/Gm of the tube (which depends on the cathode current and plate-cathode voltage). With roughly 200 V across the triode and 10 to 30 mA current, that output resistance ranges from 200 to 100 ohms (for a single triode).
Personally, I would break R3 into two 1k to 5k resistors in series, with a good film capacitor from the junction to ground. The 0B2 can handle 0.1uF directly across the tube, and the resistor after it will reduce instability caused by the shunt capacitance: the series resistance before the capacitor will improve the high-frequency noise filtering.
Thanks for explaining this to me (us). Will the effect of adding a film cap across OB2 be the same as adding a cap accross the zener diode? I never liked that, it sucks the life out of high frequencies. How is a value of this cap related to high frequency roll-off?
Putting a capacitor directly across a Zener diode, which shows a very low resistance when conducting, is not very efficient as a filter. A series resistor between the Zener and the capacitor improves the filtration dramatically (so long as the rest of the circuit allows the series resistance). Gas VR tubes have a maximum capacitance specification, typically 0.1 uF, above which they are liable to relaxation oscillation (cool flashing light, but otherwise useless).
But this is powering an SPDIF buffer is it not? Nothing you can do to this power supply can affect the audio frequency response as being suggested. Round the signal off too much or apply too much ringing and you will get errors. That is hear clicks and pops.
Want to "liquify" the high end just a tad? Well with digital audio you will need to learn DSP programming. Because tweaking a cap in the power supply ain't gonna do it!
The 0B2 can't be connected directly across B+. If the sketch really shows it wired like that, toss it. Maybe someone can figure this out if you post a scan of the original work.
--------------------------
Buy Chinese. Bury freedom.
I'll try to find the original sketch and post it here.
The power supply is not grounded in the diagram; Triode_Kingdom has apparently assumed that pad2 and pad4 are connected.
If the power supply negative is connected to pad4 instead of pad2, and the power supply is 310 volts, then it meets the ignition voltage but cannot supply enough current to keep the gas reg tube lit, even if the 5687 is removed. If you solve all those problems, you still have only about 45 volts output, not enough for a 6922.
I kind of think this must be taken out of context.
"If the power supply negative is connected to pad4 instead of pad2"
Paul, Pad 2 is clearly labeled as being connected to PS ground. On that basis alone, with Pads 1 & 2 connected across the PS output (the last cap), the voltage developed at the divider is insufficient for the OB2. I agree this has probably been taken out of context, or... it could be a piece of garbage posted to the Internet by someone who didn't finish it or didn't know how. I run into this junk constantly now when I'm searching for prior art and techniques. People will also take a commercial schematic and relabel it with voltages and attributes contrary to those described by the original publisher. It's become just as bad as unsubstantiated opinions about sound quality. If it's not from a known legitimate source, it simply can't be taken at face value.
--------------------------
Buy Chinese. Bury freedom.
Something's wrong. Based on a 310V input, this doesn't provide sufficient voltage at the R1/R2 junction to ignite V2.
--------------------------
Buy Chinese. Bury freedom.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: