• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

6sn7 alternative

Status
Not open for further replies.
Brit01,

Thanks. I've used LED bias in Beta Follower stages in DACs with 6N1P and 6DJ8/6922s, line level preamps with 6SN7s, phono stages with EC8010s, 12AT7s (paralleled and single plates), and believe it or not 12AX7's in the "RCA" circuit! In each case, it's worked out to my satisfaction. I've altered the LEDs I use to suit bias needs naturally. THE IR LEDs from Radio Shack work well for those 1.2V biases.

When cap coupling to a second stage, I like to use battery bias on grids, connecting the cathodes directly to ground.

I was most concerned about operating at 0.5-0.7mA. So, a little test jig was in order to measure DC voltage drop with changing current. I don't have the numbers in front of me, but if memory serves, there was less than 0.1V difference in DC voltage drop between 0.5mA and 5mA. I do remember thinking that bypassing was unnecessary.

A friend has actually listened to several different LEDs and proclaims one to sound better than the others. Can't say I've gone there, yet. I think it was an Agilent product, but my memory's a bit fuzzy on this.

The good news is that they are quiet enough for 1st stage phono work. About to try a 6GK5/6FQ7 Beta Follower 1st stage phono with LED bias, cap coupled to a battery biased (1.57V Lithium AA battery on the grids thought a 3.01Meg resistor) 2nd stage of a 6AQ8/ECC85//6FQ7 Beta Follower.

As you can tell, I'm sold on LED bias and Beta Follower variants. Thanks to your countryman, Morgan Jones.

For your amusement, I attached a copy of the schematic of a complete preamp I just put together for a friend who insisted on "normal" (LOL) tubes. nothing odd like 6C45PIs, 6GK5s, etc.

Another strange effect is that "rolling" tubes seems to have less of a change on the sound than with resistive loaded, resistor biased circuits.

Keep in touch. Please let me know how your projects turn out.

Stuart
 

Attachments

  • smallerschematic.jpg
    smallerschematic.jpg
    96.4 KB · Views: 531
Hi Stuart,

those suspenders look cool. I assume that your microphonics are caused by mechanical vibrations through the chassis, probably from the transformers? I get the feeling my case is the same because holding the tube usually alleviates the singing. In another posting it was suggested that they could be a result of radio frequencies from a nearby tower. Anyway I did my best to isolate the pcb. I didn't go so far as putting the Aikido in a separate box but mounted the transformers and chokes on 3/8 inch of cork and then 3/8 inch or rubber mat. I suspended the pcb using the same material as washers and I sat the chassis on a rubber mat. I guess this is all part of the compromise involved using tubes. The frustrating thing is that it doesn't always occur. I've considered how long the tubes are on, switching input/output tubes, what else is on in the house and on the circuit, music playing/sources disconnected etc. I haven't been able to establish a trend because it occurs in all situations. There is some good advice in the posts above, I'll start with Kevin and Gregg's advice and post any good outcomes.

thanks,

g
 
Gordy said:

There is only one electrical specification for 6SN7, 6CG7, etc.. All 6SN7s are the same, all 6CG7s are the same, and 6SN7 and 6CG7 are functionally equivalent. The difference that you hear is based upon different MANUFACTURING TOLERANCES. Those that are manufactured exactly to the datasheet specification will perform differently to those manufactured at + 20 % tolerance.

I wouldn't classify totally different plate structures, height to width ratios, the presence or lack of wireframe supports, mica reinforcement etc as being just "manufacturing tolerances".

While electrically they may be functionally the same when connected to a tube tester, it's pretty commonly accepted that not all tubes (within the same type) sound the same.

Doesn't matter if you're talking about 6SN7's, 12 AX7's, EL84's or anything else. Some people did the job "better" than others, and within the variation in manufacturing methods, folks prefer certain makes and versions of particular tubes more than they do other variations or makes of the same tube.
 
Gordy said:


There is only one electrical specification for 6SN7, 6CG7, etc.. All 6SN7s are the same, all 6CG7s are the same, and 6SN7 and 6CG7 are functionally equivalent. The difference that you hear is based upon different MANUFACTURING TOLERANCES. Those that are manufactured exactly to the datasheet specification will perform differently to those manufactured at + 20 % tolerance.

Hi Gordy,
While I understand exactly what you are trying to say your statement is something of a gross oversimplification. Differences in electrode design, mounting arrangements, geometry, metallurgy, and cathode chemistry result in measurable differences between different manufacturers and in many cases different plant and production vintages from the same manufacturer. AND big shock, not only are the differences measurable, but in low/no feedback designs they are generally subtly to not so subtly audible.

Carefully comparing datasheets between manufacturers will sometimes illustrate slight differences in transconductance, inter-electrode capacitances and other parameters which hint at things that might result in slightly better or worse performance in a given circuit.

U.S. Hifi manufacturers purchased large quantities of imported Telefunken and Mullard tubes not just necessarily because they were cheaper (which in many cases they were) but also because they were more consistent, had measurably better linearity, and were quieter.

Edit: Note that most parameters other than mu (in triodes) are subject to pretty large parametric variation even with relatively tight manufacturing tolerances. And, um tolerances particularly on transconductance were (and are) quite wide. No tube is manufactured to the exact specification listed in the data sheet, it is merely a representation of approximately what can be expected in a typical "bogey" device. (Statistically speaking it could either be the average or possibly statistical mean a large quantity of the device tested, and acceptance criteria would probably be about +/- 3X the calculated sigma if practice was similar to what is done today.)
 
kevinkr, DrewP; clearly I did over-simplify! (Probably 'cos I am just a simple country boy at heart). Anyway, my underlying sentiment was to say that manufacturers did not invert the half-finished product and dribble-in some 'lushness' or 'warmth' or other 'sound quality' as many seem to think, and that differences are always attributable to actual measurable parameters.


...and with that Gordy decided to re-read his Morgan Jones, and slinked off back into the shadows...
 
Keep in touch. Please let me know how your projects turn out. Stuart

Cheers mate.

Good info there.
I was trying led bias on some 12AU7's and 12AY7's that everyone seems to hate. Must have found a sweet spot on the 12AY7's for some heavy metal. I liked the sound anyway.

But will revert back to some 5687's on a WCF I'm playing around with now.

Also eager to start the 6080 amp. I got my hand on some GE 6SL7GT/6080's the other week.😀
 
Gordy said:
kevinkr, DrewP; clearly I did over-simplify! (Probably 'cos I am just a simple country boy at heart). Anyway, my underlying sentiment was to say that manufacturers did not invert the half-finished product and dribble-in some 'lushness' or 'warmth' or other 'sound quality' as many seem to think, and that differences are always attributable to actual measurable parameters.
Gordy, I'm in full agreement that there's potentially a way to measure everything we hear in the performance of a component, but it's painfully obvious that at present we don't have the facilities to do so.

We still can't do it with simple things like resistors and capacitors and get a consensus on precisely which characterisitics are responsible for the way that they sound and the ways in which all the perameters interact, so to me it's no surprise that we don't have all the answers with active devices such as tubes.

Until we understand how this whole business works, I guess folks will stick to reverse engineering those examples that consensus indicates "sound" the best. (or folks will keep chasing the overpriced NOS originals).

I'm pretty new to this game too, but I've been there during a couple of tube rolling sessions where the differences have been disturbingly obvious
 
Gary H,

Regarding your questions of post #22: I have no idea what causes these differences in sound resulting from microphonics. And it's not just in the tubes! Other passive components are clearly affected by chassis construction. I wish I knew what the mechanism was that causes noticeably different sound from the same circuit with different mechanical construction.

Over the past couple of years I have been treated to the impossible: sonic differences from power cords, connectors, etc. All sorts of things that for decades i assumed just worked or not, without contributing to "the sound".

My phono stages have been built with both integral and separate power supplies, i.e., the transformers, power switching, rectification, etc. in another enclosure. In all cases my preamps seem to benefit from all sorts of vibration damping/isolation. YMMV, as they say.

My mentor is a veritable phono stage building dynamo. He seems to build innumerable variations of circuits, trying all manner of possibilities. It seems everything affects the sound, including things that may not seem obvious. Whether the differences will appear to you depends on lots of things, such as what you listen for; your sonic priorities.

Some of the discoveries are predictable, some were genuine surprises.

FR-4 sounds tizzy. Build the circuit on 1/4" Teflon board. Chassis mount sockets help make this a very easy task. Use nylon, not steel standoffs.

Teflon sockets are superior to ceramic and Micanol. The ceramics I've used for several years and countless projects are worse than good old mica sockets🙁

What you build the circuit into (or on top of), the chassis material, makes a difference as well. I typically build on plywood boards to try out circuits, much better than thin aluminum or steel. Maple sounds better across-the-board. Surprisingly, Delrin sounds the best we've heard yet. I must stress, the exact same circuit, using the same tubes, one built on particle board shelving (the prototype), one on maple butcherblock and one on 1-1/2" Delrin. The maple seemed to have a bit nicer midrange, but the Delrin was more extended in the highs and had much tighter and more dynamic bass. All three preamp versions being compared were using the same cables, the exact same separate power supply, and the same tubes. I don't pretend to know why the base material makes a difference, only that it does.

Don't forget cones (sorry, even similar looking ones from different manufacturers sound different), the shelf material your system is perched upon, etc.

The most bizarre difference: cones under a passive line stage sound different!

I guess the lesson to be learned by us all, is that in trying to evaluate changes in your rig, you really must control all parameters and be aware of the changes. My ears usually work pretty well, but sometimes I find it helpful to have some of our DIY club members over to help evaluate a change. Since we all tend to concentrate on different aspects of the sound, someone may notice a change, better or worse, that you missed.

When your system reaches a certain level of resolution, these "mechanical" resonances/differences really become pronounced. I truly wish it weren't so. My life would be so much simpler.

Stuart
 
Sounds like this thread has gone the way of "chasing the mystical tube sound".

I will throw my $.02 in.

Let's look at it this way. IMHO most of the "tube sound" characteristics are in our heads. What I mean by this is our sense of hearing is made up in a good portion by our brains not our ears.

Has anyone ever seen the emails that go around with complete paragraphs written with all the letters out of order? You can still read it even though it looks like jiberish? That is because our brain and eyes see the group of letters as a common word rather than just a group of nonsense letters.

I beleive our ears and brain work the same way. Since most solid state amps are designed with computers and the components have very accurate specifications they can be designed to have very flat and "accurate" responses. I think our "ears" do not really "like" the flat accurate response of the SS stuff.

Just like any musical instrument there are "nuances" to the sound. They can be caused by resonances in the wood of a guitar body, slighty "out of tune" strings, etc.

To me the "tube sound" is characterized by a "natural" overall sound. Subtle changes in the response through the frequency range, slight accentuation of certain notes or sounds.

I agree with an ealier poster that given the proper testing and hours upon hours of subjective listening we could probably define the electrical properties that casue this.

Why would we? If we could predict to absolute certainty what the outcome of our projects and designs would be then why would we ever build them?

"Bottleheads" are like car guys, more than half of the satistfaction is in the "build" and "tuning" not in the final result.

Too me, at least IMHO, the final outcome of the project is less than 25% of the satisfaction in building the amp. How many people EVER get to listen to an amplifier that they built?
 
Not sure what to say...

...I only know that I have only heard a very few solid state rigs that I could listen to other than for background sound. On the other hand, good tubed components allow me to listen and be entertained for long periods without thinking about my system. I am drawn to the music. I guess to me if the scientists can not find out what it is that many of us hear with regards to different cabling systems, different tubes, different topologies, etc, maybe they need to look deeper into how they are trying to quantify rather than disregarding what we hear.
 
Curly,

You are 100 percent right!! There are certainly things we are "hearing" and in all honesty I don't want the "scientists" to mess with it. I feel that we are like an "exclusive club" membership requires some effort on our part. If the "mass production" types get ahold of the "sound" then it is "over" for us. As a "car guy" also it is like "kit cars" to me. Anyone with a decent tool box can build one. To me that is not a "hot rod" anymore than a microwave pizza is a "pizzaria" pizza!
 
Curly,

After re reading my post I see what you mean. My intent was to say that there truly IS a "tube sound" but that it would be almost impossible to measure the TRUE causes of this. Our brains and senses are so complicated and we understand so very little of them that a test method would be impossible.

It would be like trying to develop a "sure fire" new sandwich for McDonalds. They spend millions of dollars and countless hours developing new products and still have no gurantee that it will sell well.

In a very over simplistic way think about shaving. Do you use an electric or blade razor? What type? How many blades? What shaving cream? You can go on and on and never come up with ONE perfect answer to the perfect shave right?

This is what I meant by "chasing" the mythical tube sound. To each listener there is a different set of characteristics of "the sound"

Most of us can probably tell a SS amp from tube amplification in a "blind test". But could not tell one tube amp from another in the same blind test. This leads me to beleive that the "sterile" almost too accurate sound of most solid state stuff is the reason. Just like comparing a Honda to a Harley. The overly engineered Honda just doesn't have the character of the Harley.
 
Thanks everyone for the input,

as for me, my preference for "tube sound" was made the instant I first heard one. Don't know why and I don't bother discussing it any more since most hi-fi people I know go for SS and give me a good rational thrashing.

Stuart, thanks for the perspective. I have everything on a spruce plank but inside a cabinet. I'll try fiddling with some of the things you mentioned.

To be clear, I'm not searching for better tube sound, I'm extremely happy with the 6sn7s and JBs Aikido kit. I was under the impression microphonic singing was like feedback in a PA, something that just won't do. Now I'm getting the idea that this is a characteristic of tube audio and that it is the builder's job to mitigate it by improvising design techniques and tube rolling. I'm also getting used to the idea that my projects will never be perfect and for this subject I'll start a thread at the DIYobsessive/compulsive forum.

thanks all,

gary
 
For me, the whole deal is partly academic, partly emotional and exceedingly confusing.

One of my first eye opening hifi experiences was listening to a high end Teac/Esoteric front end feeding Bryston 2B NRB monoblocks into Acoustic Research M1's.

Amps were class B but the whole show sounded stunning.

A few years later playing with a Marantz PM80 which could switch from 100w in class AB to 25 watts of class A. Even on quiet passages, the class A version was so demonstrably superior that I'd have happily tolerated the lack of headroom.

Flipside that I'd heard 100W Electrocomaniet class A amps (Meridian front end source and B&W Matrix 801's) and been utterly unmoved.

Sugden A21a blows me away with how it sounds.

Have a close friend who designs in both tube and SS and I've been there when he's been blind switching amps that he has made from a 100 watter solid state to 100w of tubes (PPP 6550's). Despite there being 4 sets of "golden ears" in the room, none of us could reliably pick which is which.

Brings me to where I am now. I know that I usually prefer the sound of class A, I know that I usually prefer simpler circuits to more complex ones. Intellectually I have issues with repairing something that isn't there anymore so I'd prefer to avoid feedback if possible.

My present amp build will have 3 active stages: gain, splitter and output triodes. Input and splitter will be LED biased, outputs fixed biased. Class A. Low power ceiling but an instant recovery from overload conditions.

I still haven't figured out what path I'm going to end up on. Have heard lots of pricey stuff that sounded great, but a lot that didn't really do much for me.
 
Broskie has a new blog article about the possible issues of dc heater

But sure, ccs and vccs have made a tour round the world
Is it really that good, or just a cure for less good designs

I mean, there have been a great amount of things overhyped in the name of audio
And half the world is on its knees for years, only to find later that it was BS

So, whats up and down 🙄

btw, why isnt it possible to do a regulated AC supply ?
 
Status
Not open for further replies.