Mosfet Identification Help Request

[Enzo]

What if you had a 9v battery circuit, and you wanted to add a -9v circuit to it. WOuld you look for a -9v battery?

A negative regulator doesn't need to be made for that job. DO you have a independent transformer winding for your negative supply? If you make a basic supply and regulate it, at the end you wind up with a + and - lead. Usually we ground the - lead and have +V. But if the transformer is not already grounded, you can just as easily ground the + lead and have -V. Still using a positive regulator IC.

As to specs, this is just a power supply voltage regulator, no distortion or freq response to think about. WHat is the raw voltage you are starting with? Your +v knocks 240v down to 180v which would seem like 60v. But when power is first applied, empty caps can seeem like a momentary short, and that is zero volts, so I would want the MOSFET to have at least the 240v rating. SOmeone else may disagree. Looks like your -V starts with the same 240v. And then current. Well, how much current does the circuit need? What is three of four times that much? Pretty much any similar type transistor that meets the minimal specs should work.

[bob p]

Well, the problem isn't solved yet ... I have been digging but I can't find any suitable integrated solutions for high negative voltage regulation. I've even spoken to Tech Support at TI and they're unable to help.

It looks like I could be headed back toward mosfets/opamps in voltage stabilizer circuits.

can anyone provide specs for the 6761 and the 8401?

[bob p]

Look at the end of this thread: http://music-electronics-forum.com/t15120/

I had found the TL783 which takes up to 125v directly
http://focus.ti.com/lit/ds/symlink/tl783.pdf

And Tedmich found several interesting resources. All potentially helpful to you

Thanks Enzo! It looks like the TL783 circuit would have a lower parts count, and my needs fall right in the range of its specs:

Vi - Vo = 125V
Io = 15 to 700 mA
Tj = 0 - 125 C

Best of all: $2.02 shipped from Digikey.

[J M Fahey]

most of those devices in working order are selling in the area of $1000.

All I can say is OUCH !!!
No offense taken and wish you good luck with your project.
Still think that the "mechanical" part will be a PITA, if you want to keep it on the same high level as the electronics proper part.

[bob p]

Sorry, I didn't mean to cheat you. Those pages that I linked to were not from the company site. The company doesn't share their schematics. Those pages were from a fan site that is maintained independently. Offhand, I don't know who the site's owner is. I found the interview by Googling for the designer's name.

If I took your comments about obsolete discrete semis and mistook them to refer to an obsoleted circuit design, that was my mistake. I agree that modern semiconductors would be a better choice. If its possible to find modern substitutes for both the 6761 and the 8401 that would make the PSU circuit build-able. I guess this thread will stay open until I have the definitive list of replacements for all of the obsoleted parts, or an encapsulated solution to replace them.

WRT hunting and buying an original one -- most of those devices in working order are selling in the area of $1000. Dead ones still command a ridiculously high price, because most buyers aren't aware of the problems in repairing a PSU board that's based on obsolete components that can't be sourced. Both working and non-working units command prices that are too high and there aren't any "deals" out there.

That brings us to the prospect of a DIY build. The audio signal path is a damned simple circuit that just about anyone here should be able to build. In the best case scenario a well built DIY might sound similar to the $1000 used gear. In the worst case scenario a well built DIY will sound like vintage audio from the golden age. Its not that hard to build circuits that are on-par with the 1950s products, so if the recommendation is to buy "golden era" amps, to me that's equivalent to a DIY recommendation. To me, DIY seems like a no-loose proposition. Its a worthwhile as an experiment, anyway.

Thanks for your helpful posts.

[J M Fahey]

1) Ok, now you've thrown on the table a TON of data that you didn't before. It sort of changes the rules of the game, in the middle of a game, doesn't it? I feel cheated.
2) I was neither critizicing the use of tubes (quite the opposite) nor the use of regulated power supplies (same consideration). Where did you read that?
I *did* feel uncomfortable with somebody using a design with *components* that were obsolete in 1995. Imagine what I think of doing the same in 2010 !!!
3) I am *sure* there is no advantage in using a "6761" MosFet as a series pass regulator instead of, say, an IRFP450 or another modern MosFet.
Maybe an OP176 is better than the humble TL071 I suggested earlier ( as in 0.001% voltage regulation instead of 0.01%) , but I'm sure there are *many* fancy modern Op Amps that far outperform it, specially the Laser trimmed ones which are spec-adjusted individually before encapsulation, a technique that was not available in 1968.
4) No, I don't believe that a NOS , '68 vintage semiconductor is "better" than a modern one.
I agree that , say, a 1956 Telefunken E83CC is "better made" than a 2010 Chinese 12AX7, although both are nominally the same, but I do not extend that analogy to semiconductors.
4) I think that if you like that sound, fine.
May I suggest that you hunt and buy an original one.
A clone will *not* sound the same, period.
I'm not talking about the electrical schematic itself, but to all-important factors such as lead dress , actual "mechanical" grounding and the like.
Not to speak about components as switches, connectors, the chassis and cabinet, etc.
5) I am not doubting your construction skills at all, on the contrary, I recognize and respect your years of experience.
In fact, I'm sure that if you had available a kit with all the original parts (say, from a factory closure auction) , you would build a *perfect* Preamp.
Buying parts here and there, and, specially, not having the "mechanical" stuff, won't help.
6) Anyway, Friend, it's your project and I sincerely wish you the best.

[Enzo]

Look at the end of this thread: http://music-electronics-forum.com/t15120/

I had found the TL783 which takes up to 125v directly
http://focus.ti.com/lit/ds/symlink/tl783.pdf

And Tedmich found several interesting resources. All potentially helpful to you

[bob p]

Forgot to say- unlike tubes, a P-chan mosfet can regulate your negative rail. It's tough getting electrons to flow backwards in a thermionic device!

jamie

Thanks for your ideas. I'm looking at the first schem that I posted and you're right -- the bias supply has an "8401" in the voltage stabilizer circuit, and my understanding is that its a P-channel MOSFET (though I haven't found the data sheet yet).

[bob p]

Now that I know what it's all about, I simply doubt that that amplifier, preamp, and all the conceptual package sound any better (or worse) than any other good audio equipment.
I find it a marketing gimmick.
Don't waste time on them.

I wouldn't call them a "gimmick", unless I was willing to concede that any tube amp was a "gimmick" back in the 1970s, 80s and 90s when tube amplification systems were considered an oddity in a marketplace that was so heavily dominated by solid state gear. Only people with "affectations" bought them.

Personally, I think that the references to the "golden age" overlook the important fact that the guy who was responsible for those designs that we're talking about during the 1970s-1990s was one of the "golden age" designers of the 1950s-1960s. He was NOT one of those new kids on the block who entered the marketplace late in the game and designed a "gimmicky" feature to establish a foothold. He was one of the elder statesmen who was using the best solid state technology he could, in order to allow technology that everyone else had abandoned to compete directly with the most hideously expensive solid state gear in the rather extreme "high end" marketplace that was being dominated by young whippersnappers like Mark Levinson and Nelson Pass.


To put things in perspective for myself, I've compared my ARC VT100 to some vintage tube gear that I own. Two of my most prized pieces are an old H-K 250 and a Marantz 9. As you know these are two well respected amps from the "golden era." They're nice amps, but they are quite limited in terms of what they can and cannot do. They're not even close to being in the same performance class as the VT100. Its a 100+100 watt stereo Class AB amp running an octet of 6550s with cathode follower drive circuits and fully regulated voltage supplies. I don't think its necessary to go into supporting arguments (and I hate arguing about stereo stuff anyway), but suffice it to say that in a high powered application where you don't want the supply rails to sag, there are some legitimate reasons to say that a traditional Pi-filter amp just won't compare to an amp that employs fully regulated voltage supplies.

A couple of years ago R.G. and I had a thread going where we were discussing the power supply for a "monster" 300W Class AB amp that I wanted to build. The power supply became a real problem because as you know, an array of 6550 going into maximum current draw can wreak havoc on a PS rail that doesn't have low impedance and isn't well regulated. Suffice it to say that for a monster amp (or even for a big amp), regulated supplies have real merit.

Of course, its probably overkill to put the same degree of regulation in a preamp, but isn't overbuilding part of what the hobby is about?

[bob p]

I had thought that *you* had pulled a circuit from a 1968 vintage "Electronics" Magazine, which was directed to NASA engineers , IEEE fellows and the like (I read them at the University) , I almost fainted when I saw *that* was used in a 1995 design!!

Its interesting that you say that. Looking back on 1995 from 2010 with the advantage of retrospective vision, the designs do look antiquated. But given the passage of enough time its easy to lose perspective on the history of evolution in amplifier design. I'd rather not morph this thread into a debate about antiquated technology, because we are talking about vacuum tubes after all! To put things in perspective, let's consider this quote that I dug up from one of the stereo magazines:

"You've set the audio industry back 20 years!"

That was a fellow engineer's response upon seeing vacuum tubes in William Z. Johnson's amplifier at a hi-fi show in 1970. Back then, the vacuum tube had been totally abandoned in favor of the transistor. In fact, the only newly made audio components based on vacuum-tube technology you could buy were guitar amplifiers. Glowing bottles were embarrassing reminders to the engineering community of just how "primitive" electronics technology had been just a short time ago.

Imagine that engineer's surprise if he could have known that, instead of setting the audio world back 20 years, Bill Johnson's pioneering return to the vacuum tube paved the way for a tube renaissance that would reach full bloom 20 years later.

One reason that we still have new production vacuum tube audio gear today is because there were people who returned to vacuum tube technology at a time when the industry had abandoned it in favor of transistors.

Thankfully back then there were a select few people who realized that SS devices weren't all they were cracked up to be. They saw that vacuum tubes still had relative strength in some applications. One of these people was the designer we're talking about, William Z. Johnson. He recognized the strengths and weaknesses of both tubes and solid state devices, and started a company building hybrid amps at at time when nobody else was even interested in tubes.

Back in 1970 he was the only person with the foresight to use the new 1968 "NASA-spec" solid state voltage stabilization circuitry and apply it to consumer vacuum tube audio amplifier power supplies -- and he did this at a time when the rest of the industry was abandoning the vacuum tube altogether.

The idea of using solid state regulation devices to support vacuum tube amplifier technology was a novel idea -- it was something that nobody else was interested in. He continued to produce his amplifiers using these design features through the 1970s and 1980s -- a period that was admittedly very lean for vacuum tube audio companies! He registered a lot of patents along the way, and he continued to use similar power supply designs in the 1995 circuit we're talking about. He continued to manufacture tube amps while other manufacturers were rejoicing in the glory of the transistor.

Fast forward to the 1990s. His company was one of the only American tube audio companies that was still in business. They had been producing high quality vacuum tube amps continuously for decades, and their products were on-par performance-wise with the competition's best solid state offerings. Because a few companies insisted on producing expensive, high-end vacuum tube audio components, we saw a renaissance in interest in consumer audio tube amplifiers in the 1980s-1990s. That only happened because designers like this one carried the torch and produced "obsolete" designs like the one we're talking about when nobody else was willing to bother. After they proved the market would support tube audio, and a lot of "me-too" companies came along to share in the profits. Suddenly everyone wanted to jump on the bandwagon.

I can't fault a forward thinking guy who came up with the idea of using that brand-new 1968 military/NASA technology on a consumer audio amp in 1970 -- especially if you consider that he did this at a time when everyone else was abandoning the vacuum tube altogether. He continued to use the technology for decades, as he made obsolete vacuum tube technology completive in a high end consumer audio marketplace that was being dominated by transistors. He was one of the few people who continued to pioneer that trail at a time when it was very unpopular to do so.

Fast forward to the 21-st century. Tube consumer audio is in a great revival period, and new companies are coming out of the woodwork to share in the profits to be made in the tube marketplace. We're all on this forum because we love the sound of tubes. Its hard to deny that this guy continued to pioneer new advancements in tube technology at a time when everyone else was abandoning it. He's built vacuum tube audio amps non-stop for the past 40 years and his company is still in business and profitable. That fellow is VERY old now. He just sold his company because he's too old to run it. I can hardly blame this guy for using "obsolete" designs in 1995. He was the leader in the industry at the time when those designs were bleeding edge. There's nothing wrong with an old man of the golden era sticking with designs he's familiar with. Especially if he's one of the few men who were willing to carry the torch while everyone else was running the other way.


Back to that "antiquated" voltage stabilizer circuit -- can you tell me if there were suitable integrated / one-chip devices around in 1995 to eliminate the need to build this type of circuit with discrete components? I can't say that for sure. What I can tell you is that I spent this evening at the National Semiconductor website looking for packaged solutions to this voltage regulation problem and I came up empty handed. Maybe I just couldn't find what I needed, but to me its looking like these antiquated designs are still one of the best solutions available. My only other option would be to stack up a pair of voltage regulator tubes and bleed off the extra voltage with Zeners. In comparison, my method looks like a hack.

[imaradiostar]

Forgot to say- unlike tubes, a P-chan mosfet can regulate your negative rail. It's tough getting electrons to flow backwards in a thermionic device!

jamie

[imaradiostar]

Seems like a mosfet and a stack of zeners make a pretty decent regulator if you're willing to throw away a little energy as heat- wire the mosfet up like the old series pass tube voltage regulators with a high value resistor to B+ and a smallish cap and your zeners connected to the grid...er um, gate and ground. Sure, the source will end up a few volts in the opposite direction from its tube counterpart but it makes little difference. The capacitance at the gate should (if I recall correctly) be multiplied by the gain of the mosfet. Should have good enough regulation to run any preamp that glows.

jamie

[J M Fahey]

WHy build discrete when they already package it in a TO220?

Maybe because the TO220 solution costs 50 cents and the discrete version can be sold for U$ 2500.
I had thought that *you* had pulled a circuit from a 1968 vintage "Electronics" Magazine, which was directed to NASA engineers , IEEE fellows and the like (I read them at the University) , I almost fainted when I saw *that* was used in a 1995 design!!
It's like building (now) an amplifier with SMT construction, DSP, etc. , but with some cloth covered wire, a couple paper-oil caps , a few carbon composition resistors and a 2A3 direct heated triode for Mojo. Gimme a break !!!
Now that I know what it's all about, I simply doubt that that amplifier, preamp, and all the conceptual package sound any better (or worse) than any other good audio equipment.
I find it a marketing gimmick.
Don't waste time on them.
If you want to build something "special", build a *real* classic, such as a Harman Kardon, Scott, Marantz, Quad, etc. from the "golden age", say from 1959 to 1968.
besides, you'll feel more comfortable around that.
Good luck with your project.
(And don't worry about that balanced input, just drive it unbalanced or build an NE5534 inverter; it will NOT hurt your sound, believe me)
Whatever sound you hear today, has gone through zillions of Op-Amps, although I admit that the most critical ones probably were discrete)

[bob p]

We were talking not long ago about high voltage regulators. Seems to me it might have been someone trying to invent a 48v phantom supply, but the LM317 he was wanting to use wasn;t up to the voltage levels... or something. I recall finding some newer adjustable regulator IC that was good for high voltage, but I don't recall just how hugh a V, nor the part number. In any case, the circuits posted above look more or less like the internal circuit of some Vreg IC anyway. WHy build discrete when they already package it in a TO220?

Enzo, I'm not up to date on what's current in Vreg ICs. I guess I should stop throwing those tech articles from FTM away with the junk mail. If there are packaged devices that will eliminate the entire voltage stabilizer/opamp circuit, I'm more than willing to use them.

If anyone knows of a suitable device and can point me in the right direction, I'm more than willing to look at the applications data sheets. thanks.

Almost forgot -- this application is a preamp with all miniature twin triodes, so the B+ will be at or below the anode voltage limits of ~ 250-300 V. The schematics I was looking at use regulated plate voltages of +180 VDC and regulated bias voltage supplies of -90 VDC.

[bob p]

Sorry that I didn't give you enough information. It wasn't part of the original post because I was originally just looking for MOSFET identification. I didn't realize that I was looking at such a dated PSU. If it would be better to re-design the supply with modern parts, then I really do need to tell you more about the application! Sorry about that!

This is the follow-up to those PMs we've had about the balanced tube amplifier. My objective is to build a HiFi preamp with a balanced signal path to drive that balanced amp we've been talking about. In that application zero sag is an objective. The design is still in flux as I shake out the criteria for a phono input stage, so I can't give you the simple answer of X milliamp at Y volts DC. At least not yet.

I haven't been able to find many good examples of a fully balanced stereo preamp. This is the most recent design I've been able to find, so I've been using it as the starting point:

ARC LS-22 Line Stage (balanced)
Description & Specifications
Preamp signal path
Switching circuitry
Power Supply

This is already an "obsolete" design even among the tube HiFi people, as the 6922 seems to have fallen out of favor in deference to the 6N1P. As you've mentioned, the PSU seems to be a bit dated as well.

I was planning on keeping the basic circuit topology and making the following changes to the circuit:

-- Replace the 6922 with 6N1P; the net result would be that the total 6.3v heater current will approximately double.

-- Add a phono input based on traditional designs; at this point I'm thinking that this would add 1 or 2 more miniature twin triodes per channel.

For the phono input I had thought about borrowing a traditional implementation borrowed from this circuit:

ARC SP-8 Phono Preamp (unbalanced)
Description & Specifications.
Schematic.

It uses a straightforward tube based phono input stage, seen on the left side of the schematic. I can't see the point in building a phono amplifier as a stand alone component, so I'm planning on adding the phono input to the line stage I mentioned above.


I understand that the current trend in tube HiFi involves a hybrid approach, using JFET as the first stage in a tube phono amplifier because of noise considerations. I am somewhat limited in my ability to entertain the JFET kind of design because of the poor availability of schematics. So far I've been thinking about using a traditional tube design with a fully regulated supply. I would be willing to entertain a JFET design if anyone could point me in the right direction.

At this point a rough estimate for the all-tube implementation I would say that the total total tube count will be approximately 12 medium mu twin triodes, maybe 10x 6N1P and 2x 12AX7.


I'm not married to these circuits. There aren't a lot of options out there for fully balanced circuits, and this is what I've been able to find. The circuits appear to use signal paths that are nothing at all out of the ordinary; the strengths of these designs seem to be primarily attributable to PSU regulation. Although the PSU circuits may be dated, these are the most recent designs I've been able to find.

I was thinking that given the relative simplicity of the signal paths, something like this could be a good DIY project. Now the PSU looks like its going to complicate things.

I'm not married to this approach. I'm open to suggestions and other alternatives if you think there are better ways to solve the problem. Thanks for your time.