Tube Lab  

Greetings! I'll try and work my way through your questions quickly, some you will need to research. 1. No once the cathhode bias is set the tube is operating in class A and is essentially on 100% of the time, so the when a signal is applied there's no significant shift in bias or emissions (mA). Remember bias is DC and the signal is AC, and you can't alter the bias of a tube with an AC signal. 2. Ahhh yes this has been tried and many think a hardened power supply is the solution for great sound. To understand why this isn't the case, you need to think of an amplifier (any preamp, power amp) as a complex valve that turns on the power from the wall outlet. In my design work my approach is to create a minimum amount of restrictions between the amp and the house mains power. Yes we need to filter that power, but in listening tests I've found that the unregulated minimalist power supply gives audiophile grade sound - and that's all that matters! When you first start designing audio amps, you look at all the amazing circuits people have come up with and want to use it all. But the reality is the opposite, go simple, short signal paths, no fancy add on circuits, quality components with a design that is engineered from first principles. This is my approach and it has created the highest quality of sound I've ever heard.

Well that was a taste of good things to come. Our goal is to call for Test Builders as soon as the production PCB's arrive and we get our newly installed CNC up and running so we can make the top plates quickly and efficiently. The component parts we're missing are arriving almost every day - in fact while I was typing a bulk order of the coupling capacitors arrived!

I have one original mono Miles pressing and yup you guessed it - Round About Midnight - and I just learned that was his first album for Capital Records, which means it was recorded at "The Church" on 30th Street.

@@tubelab194 That makes it double cool. It's too bad that studio was demolished. Not only was it sound engineers dream but to think about how much American musical history took place under the roof of "The Church" is beyond phenomenal. History is a non-renewable resource.

@@jeffbrooke4892 agreed, but at least we still have all of these wonderful recordings to enjoy.

C9 is or was a low pass high frequency trimmer cap that RCA engineers incorporated in the circuit to compensate for the very high output impedance (~22kohm) of the last gain stage without a CF (cathode follower) stage. This is easily understood when you realize the circuit was designed with a console stereo in mind (I grew up with just such an RCA console and probably the same phono circuit!) so the phono OUT would only have a few inches of hookup wire to mess with the EQ. With a CF stage added to the circuit, and now very low output impedance the EQ isn't affected at all by the RCA patch cord length. But the original RIAA equation doesn't include a high frequency roll off. All designers deal with this in various ways and my solution was to bench test the value of C9 till I was happy with the roll off. This has two benefits, one the EQ has a defined upper frequency that passes before a 1st order roll off, but just as importantly it brings down the noise floor substantially without affecting the quality of the sound. In fact as a critical listener I'm very sensitive to high frequency noise and do everything I can to get it below my hearing threshold and my business/design Partner's hearing (my Son Charles is 25 yrs younger and he has amazing hearing).

@@tubelab194 Thanks for such detailed reply.. I stand corrected.. Have a great week.

@@ceylonmooney thanks for noting this, we made a revision the the schematics and forgot to put a new link in place. By the time you read this the links should be active.

The 12ax7 and 6sl7 are very different tubes electrically, yes they're both high gain twin triode but the similarities end there. Any circuit with a critical EQ stage (like the RIAA) is going to be affected by the plate resistance of the preceding gain stage. So you must use the 6sl7 or direct equivalents in my circuit for the equalisation to come out flat. Now you could build the original RCA 12ax7 phono preamp, that's what my circuit is based upon. Caution though, it took me months of lab work and listening tests to get the circuit where I want it, so even though RCA published a schematic doesn't mean you won't need to do a lot of technical work to get it perfected.

That's a fact Jim, ...indeed and it needs lots of time listening , perfectioning and voicing such a delicate circuit ....@@tubelab194

Thanks Scott for your Eagle eyes, you are correct we both managed to produce a final schematic with an error, hopefully that's the only one! We'll have a revised schematic in the store information downloads shortly.

@@tubelab194 Of course, I'VE NEVER made a mistake! 🤣 You believe that, right? Really nice work, by the way.

see the explanation above to Jesper. Thanks for your thoughts on the design.

Normally when people say "I hate to be a bother but...", they actually mean exactly the opposite! Asking design questions about a RU-vid video that is all about a new design is almost always expected and often results in some constructive dialogue. Your calculation of the current passing through V2 is correct at 0.75mA (E/R=I), which in turn means the plate dissipation is .01125 watts. I've used this operating point in another Kit design and found the CF (cathode follower) to work well and more importantly, to sound great.

@@tubelab194 Please help me understand why this low current was chosen because it makes no sense to me. (particularly on a 6SN7 tube which has a DC Cathode Current limit of 20mA and a peak of 70mA - the curves on the final page of the GE 6SN7GTB data sheet don't even extend to that low standing current) Are you expecting to be directly driving the grid of a preamp tube in another schematic or is this a standalone device?

@@jasonagee385 No this obviously feeds into a standard control preamp. The CF stage doesn't require much current to work well and in bench testing I've found this operating point to work well and sound great. Not every design decision and circuit will be text book, in this case the scope, stability and sound all sent me in this direction. The final arbiter on my bench is two fold, first does it work and is it stable and second does it sound good.

Normally you don't want high voltage DC on the input grid of the tube. Some tubes like the E80cc can handle this and we have a direct coupled Kit Preamp that does exactly that. Is it possible to direct couple this circuit - maybe - but given this is a Universal design and users will want to experiment with different compatible tubes in the CF stage, it's much safer using a coupling cap. And I love these Vishay caps for this application, sonically they're very neutral and transparent.

Thanks for your thoughtful comments. Nothing strange about it, the design was 6 months in development and we thought about, designed and built many versions. Everything in audio is a compromise, and we try and make as few as possible. In our experience if you're going to combine the channels in one twin triode the CF stage is the place to do it. Listening tests have confirmed the engineering decision, channel seperation is excellent, better than any phono preamp I've ever heard. We're currently playing with a new preamp platform that allows us to listen to and test literally dozens of different miniature 9 pin twin triodes and we've confirmed that a tube built with a shield between sections can easily handle both channels and maintain excellent stereo seperation.

I’m not going to argue with your experimentation, but for analog you could have basically went with a single tube in both stages as channel separation in vinyl is the limiting factor. I guess there is the issue of L/R channel separation and also circuit/power supply isolation so maybe you’ve empirically identified a balance between the two.

@@rambler3080 correct, and it's important to note that we design everything from first principles. Nothing is taken for granted, sometimes we think something will work better, but testing on the bench and "live" listening says the opposite. Most tube preamps and power amps manufactured in the last 50 years are just copies of old designs with either very little or no development work. It's fine to say "I think" a better approach is seperate tubes for each circuit, but until you actually build the circuit (we must have built and tested 20 or more versions) you don't know. Some things I've learned in audio design include don't automatically do what everyone else is doing - most of the time they don't have a clue what they're doing, always bench test and follow up with lots of critical listening. Only when it passes the bench and listening tests is it a viable product.

This isn't a paper design, it's an actual working prototype that's been months in development. The circuit works perfectly, is very low noise and follows along a design\build path that we've developed over the years. Years ago I thought the same and built a prototype preamp with a floating DC voltage and it turned out to be a very noisy circuit.

This is a good question and we've thought about including a bottom plate option, but given the fact that tube amps have almost as dangerous a top plate, we feel that Kit Builders need to take responsibility for safely installing the equipment. Years ago when I was young all of our tube gear was housed (with partially open backs) in consoles, now we've taken our tube gear out of those lovely consoles and left them exposed. So my recommendation to all tube amp owners that have children in the home (even occasional visits from grandchildren) is to place the gear on a higher shelf with no climbing access and use a large lockable breaker bar to shut the system down when you aren't in the room. For an average adult, tube gear that is well designed like our Kits is perfectly safe to operate as designed and we've never had a problem. This broader question would make a good Tube Lab.

@@tubelab194 A grill, even if it's plastic on the bottom would keep fingers safe just in case someone decides to turn the unit over whilst it's switched on. Remember, valve amplifiers use high voltages, exercise caution. If you are selling these units you will be liable if there is a design feature that contributes to unsafe operation. It wouldn't cost much to install a grill or mesh. I dont know of any electrical device that has free open access to the circuit boards and wiring, even if it’s at the bottom and needs to be turned over.

Hey John! If you look below the video the entire equipment chain is listed in order. So much of your capacitance is going to depend on your cables I doubt there's much point in even calculating the capacitance in the circuit. My suggestion would be to start with low capacitance cables with no additional capacitance added and see how you like it. Then add 100pF and see if things improve or not, if yes then possibly try 200pF. In our system we're running it without any additional capacitance and we're happy with the sound. Once the first production prototype is built we'll be able to easily experiment. And that was the point of designing in some switchable components, to allow experimenters the opportunity to tweak.

Plate grid capacitance of a given tube is very variable when measured cold rather than hot ...it also varies a lot beween tubes of different brands ........first always try without extra capcitor....sounds often ( not always ) so much better without .. your ears will tell you . .....TL designer does it right and he uses his ears.

@@frankgeeraerts6243 thanks for your reply.i have a few small value polystyrene and polypropylene caps to experiment with.im definitely going to start with no additional capacitance on the input and slowly add and listen to change.based on your experience I may have to repeat if using a different tube.

@@johnstuchlik5828 You can also install a selector ....with No capacitance and some small different values and comparing by listening to the balance and harmony as well as detail and air in the music , use more than one album to evaluate as recordings differ ...