Korg 'Nutube' - next generation vacuum tubes

[BubbaFet]

In Korg's application circuit schematic, I like how they use JFETs to give their device "tube" tone. Cool!

[rezzonics]

Isn't it just a buffer that let pass everything?

[BubbaFet]

Isn't it just a buffer that let pass everything?

Yes, but it's a MAGIC buffer, like found on a Boss MT-2 Metal Zone input, or on  Rat pedal output.

[teemuk]

They're just buffers, nothing magical in them. And yes, they are pretty much "transparent" unless you try to drive too low-impedance loads (e.g. less than 1K) or simply drive the bufferl with too hot signal. Both would cause the buffer to distort but that should not happen in usual operating conditions.

What folks need to realize about these NuTubes is that they are not your generic 12AX7's. The transconductance is low, B+ voltages failry lowish, and overall impeance levels moderately high-ish, for a vacuum tube circuit. These things are also pretty much operated in grid current region, which is why buffering is even needed in the first place. That alone will change operating characteristics into something different than those of circuits that are not operated in grid current region. IMO, I don't see a gigantic difference to "starved plate" circuits in general. I'm quite sure a plain gain stage is charcteristically closer to Butler "Tube Driver" circuit (and alike) than generic single-ended 12AZ7-based gain stages.

Not that it wouldn't mean one could't squeeze some nice clipping distortion out of these circuits.

Also, the directly-heated cathode can be a nuisance considering schemes such as cathode followers or schemes that utilize cathode loads.

[BubbaFet]

They're just buffers, nothing magical in them. ...

No magic allowed, eh? 

[aron]

Bummer. I wanted to believe the hype!

[EBK]

JFETs are magical, and if you use a TL07x for your buffer, you get TWO per op amp! 

You also get something called an epi-FET, whatever the hell that is (must be amazing)!


(Not trying to be mean. Just goofing off.)

[GGBB]

You also get something called an epi-FET, whatever the hell that is (must be amazing)!

I believe that is the antidote for sudden allergic reactions to BJT or op-amp based pedals.

[BubbaFet]

With very few exceptions, every circuit that I love and use has a JFET in it.   ... the JFET rules for me.

Bummer. I wanted to believe the hype!

If you are refering to jfet buffers then perhaps you are conflicted about "Magic".    Or did the "Ministry" get to you too? (Harry Potter reference.)   

If you are referring to Nutubes then I agree about the hyperbole.

[rezzonics]

Does anybody know of some liable, proven "starved tube" schematics used for overdrive or distortion? Valve Caster?
I've read that Spice models are not usable for starved tube circuits, but Koren proposed a method to approximate tube model parameters from real measured curves. Could not these parameters be recalculated on the starved region?
Mi idea is to compare simulations of NuTube with "an equivalent" starved tube circuit.

[BubbaFet]

There are some here...
http://revolutiondeux.blogspot.com/2012/04/dean-markley-overlord-to-b-k-butler.html

[rezzonics]

They're just buffers, nothing magical in them. And yes, they are pretty much "transparent" unless you try to drive too low-impedance loads (e.g. less than 1K) or simply drive the bufferl with too hot signal. Both would cause the buffer to distort but that should not happen in usual operating conditions.

What folks need to realize about these NuTubes is that they are not your generic 12AX7's. The transconductance is low, B+ voltages failry lowish, and overall impeance levels moderately high-ish, for a vacuum tube circuit. These things are also pretty much operated in grid current region, which is why buffering is even needed in the first place. That alone will change operating characteristics into something different than those of circuits that are not operated in grid current region. IMO, I don't see a gigantic difference to "starved plate" circuits in general. I'm quite sure a plain gain stage is charcteristically closer to Butler "Tube Driver" circuit (and alike) than generic single-ended 12AZ7-based gain stages.

Not that it wouldn't mean one could't squeeze some nice clipping distortion out of these circuits.

Also, the directly-heated cathode can be a nuisance considering schemes such as cathode followers or schemes that utilize cathode loads.

Very interesting. I will try to generate SPICE models from NuTube 6P1 datasheet curves and starved 12AX7 curves from this study and compare the response of both.
I will post the results here.
Stay tuned!

[frequencycentral]

Vox appear to like nutubes.

http://www.voxamps.com/MV50

[sajy_ho]

Is there any Pentodes too? Cause having an output Pentode tone before a SS PA would be amazing...

[rezzonics]

I have published on my blog the calculated Spice models with curves from a starved 12AX7 and a Korg Nutube:
Korg NuTube 6P1 vs 12AX7 starved tube / valve: SPICE models

Starved 12AX7:


Korg Nutube 6P1:


I will try to make a comparison of frequency response with a basic circuit.

[rezzonics]

I have published on my blog the gain and frequency response comparison at +24VDC of a NuTube 6P1 triode and a starved 12AX7 tube:
Korg NuTube 6P1 vs 12AX7 starved tube / valve: Gain and frequency response

I am a bit surprised by the results because, if the simulation models are correct, the starved 12AX7 tube has more gain than the NuTube 6P1 triode, 24.6 dB of gain for the 12AX7 compared to 17 dB of the NuTube.

The NuTube requires plate resistors 8 times higher than 12AX7, NuTube maximizes gain with 400K load and 12AX7 with 50K load (in starved mode). NuTube requires bypass capacitors 15 times higher than 12AX7 in order to have a similar response at low frequencies:

The following figure shows the gain and frequency response with bypass capacitors from 10nF to 10uF for 12AX7 (in green) and for NuTube from 150nF to 150uF for NuTube (in blue)

[PRR]

> I am a bit surprised

Why?

12AX7 is a HIGH-gain device.

NuTube is a display panel. Never meant to have electronic gain at all.

Are you modeling the HIGH grid current the NuTube needs in most of its "good" operating range?

[thermionix]

Vox appear to like nutubes.

http://www.voxamps.com/MV50

So they're selling a 50W amplifier paired up with a 25W speaker?

Recipe for success!

[BubbaFet]

So they're selling a 50W amplifier paired up with a 25W speaker?

Recipe for success!

It's a solid state class d amp that puts out 25W into an 8 ohm speaker, and 50W into 4 ohms. So the speaker in question is most likely 8 ohms.

[rezzonics]

> I am a bit surprised

Why?

12AX7 is a HIGH-gain device.

NuTube is a display panel. Never meant to have electronic gain at all.

Are you modeling the HIGH grid current the NuTube needs in most of its "good" operating range?

I am surprised because 12AX7 is a high gain device when operated at +350VDC, but I was expecting a huge degradation at +24VDC and NuTube is based on a display technology but it has been designed to work as an amplifier.
I added a current source in the simulations to model grid current, typical grid current is 6uA, max is 30uA.