Tube Sound Processor

[ja2dhc]

What is the TUBE SOUND ?
I think the biggest contribution comes from small distortion
such as 0.5% THD.

To generate 0.5% THD, I propose to install a pre processor before
solid state main amp. I did simulation with the attached circuit
schematic using 12B4A on TINA-TI simulator. The results are as follow:

Voltage Gain = 4
THD = 0.512% @0.4Vrms output

Circuit Description:

Power supply: Laptop PC's AC adapter, DC 12V out put for both
plate and heater power supply.

I eliminated the Cathode By-pass capacitor as it affects to
degrade the sonic. 300mA of 12B4A's heater current is
merged into a cathode resistor with the 12B4A's plate current.
Resulting. only 2.2 ohm of the cathode resistor is enough and
the negative feed back due to cathode resistor is minimized

I have not yet assembled the actual amp of this circuitry.
Simulation only at the moment.

'73 de JA2DHC

 

[Robert Kesh]

does the sim include grid current? some do, some don't. it will affect things at very low voltage. and you've no grid leak resistor, i would sim that, because grid current will put a voltage across it.

 

[ja2dhc]

Hi,Robert Kesh, Thanks for a quick response.

The SPICE model of 12B4A used in my simulation employees
grid current (see quoted SPICE model at the bottom). However
This SPICE model assumes only positive grid domain.
don't care negative domain of the grid (ie; don't care
the thermal velocity effect) .

'73 de JA2DHC

*
* Generic triode model: 12B4A
* Copyright 2003--2008 by Ayumi Nakabayashi, All rights reserved.
* Version 3.10, Generated on Sat Mar 8 22:41:12 2008
* Plate
* | Grid
* | | Cathode
* | | |
.SUBCKT 12B4A A G K
.PARAM X1=1 X2=0.087065329 X3=-1.2036947
.PARAM X4=0.55479636 X5=5.1134435 X6=2.7036947
.PARAM X7=0.0047680512 X8=9.2167935 X9=0.0036584566
.PARAM Y1=0.0023840256 Y2=0.0031806909
BK IK 0 V=U(V(G,K)+X1)*X7*URAMP(V(G,K)+X1+URAMP(V(A,K))/X8)^1.5+(1-U(V(G,K)+X1))*X9*(X2*URAMP(V(A,K)))^X3*(X4*URAMP(V(G,K)+X1+URAMP(V(A,K))/X5))^X6
BA A K I=URAMP((Y2*URAMP(V(A,K))^1.5)-URAMP((Y2*URAMP(V(A,K))^1.5)-V(IK)+Y1*URAMP(V(G,K))^1.5*(URAMP(V(G,K))/(URAMP(V(A,K))+URAMP(V(G,K)))*1.2+.4)))+1E-10*V(A,K)
BG G K I=Y1*URAMP(V(G,K))^1.5*(URAMP(V(G,K))/(URAMP(V(A,K))+URAMP(V(G,K)))*1.2+.4)
* CAPS
CGA G A 4.8p
CGK G K 5p
CAK A K 1.5p
.ENDS

 

[DF96]

Some people like distortion and may call it 'tube sound'. They often deny this. They may obtain the sound they prefer by misusing valves, but possibly not realising that this is what they are doing.

Others like valves for other reasons, and know that when used properly their distortion is sufficiently low that it will be inaudible and so can be ignored. Mostly, they dislike 'tube sound' as a phrase.

 

[12B4A]

I think you're going to find the sim is very unpredictive of actual performance. You're going to need 12 volts plate-cathode just to get any conduction going.

 

[Wavebourn]

Actually, "tube sound" means absence of specific distortions generated by poorly designed solid state amplifiers. You can't add absence. But you can design and build solid state gear that does not have that specific distortions, so sounds like tubey.

 

[yangshau]

tube sound

I think another important aspect of "tube sound" comes from the output transformer drive. Instead of using voltage drive a speaker in solid state design, the transformer acts more like an energy drive. This in turn can achieve better dynamics when driving a nonlinear load such as speaker. So I don't think just introducing the distortion can emulate the "tube sound".

 

[tinitus]

What is the TUBE SOUND ?

tube sound originates from a time when most speakers lacked midbass BSC, simply because they didnt know better

tube amps could work like magic on that 'problem'
you might say it relates mostly to high output impedance, due to the output transformer
but its only partly true
a very simple tube preamp could also do wonders in this respect

but Im sure its much different today
speakers are much better designed
and knowledge about transistor based design have also improved some

 

[DF96]

I think another important aspect of "tube sound" comes from the output transformer drive. Instead of using voltage drive a speaker in solid state design, the transformer acts more like an energy drive. This in turn can achieve better dynamics when driving a nonlinear load such as speaker. So I don't think just introducing the distortion can emulate the "tube sound".

This appears to be an attempt to turn a weakness into a virtue, by redefining reality. The correct way to deal with this is to deal with the weakness by using feedback; not try to woffle it away.

 

[ja2dhc]

I have finished to assemble my sound processor (pre-amp)
to generate "Tube Sound"

The final schematic and external view are shown in the
attached photos.

Focus Points of circuit design

*Single LAP TOP PC use DC15V power supply to cover everything,
Plate Voltage, Heater, Power Amp (Class-D).

*Heater currents are merged into a cathode resister of 1.1 ohm
so that By-Pass capacitor can be eliminated since current feedback
is very low due to low resistance of 1.1 ohm. Channel separation is
about 50dB (measured).

*In order to avoid effect of grid current, I used input
transformer whose resistance of the secondary winding is 240ohm.

* Plate current is about 4mA (measured).

From now on, I will measure. Now I am typing this post with listening
music, good sound !

'73 de JA2DHC

 

[tomchr]

There's a fair amount of tube emulation circuits showing up in guitar amps and such. Some of them are quite good.

There seems to be two different approaches:
- DSP modeling
- Solid state circuits that create a harmonic structure similar to that of a tube.

These models/emulators seem to do a good job of modeling the effects of the tube transfer characteristic, but often not such a good job of modeling the effects of OPT saturation, etc. For hifi purposes, it wouldn't surprise me if one could design a tube emulator circuit that would get you 90 % of the "tube sound" for 10 % of the cost of a tube amp.

A tube emulator still wouldn't be as pretty as a well-excecuted tube amp, though.

~Tom

 

[ja2dhc]

I measured the distortion characteristics as shown in the
attached photo.

I use Class-D amplifier for the final power amp. This amp uses
Tripath's TA2024 with DC15V power supply.

THD =2%@1W is the worst data among my home brew amplifiers
as is a successful result of the objective of Tube Sound Processor.

'73 de JA2DHC[/QUOTE]

 

[Sprags]

What is the TUBE SOUND ?

That would be a shame if some lame developer thought they could develop an app to simulate 'tube sound' for an MP3 tune to be filtered through on an iPhone or iPad. And you could select what type of tube sound. Push-Pull KT88 or 6550, SE 300B or 2A3....hmmm. Don't get them started.

 

[Rainwulf]

Some people like distortion and may call it 'tube sound'. They often deny this. They may obtain the sound they prefer by misusing valves, but possibly not realising that this is what they are doing.

Others like valves for other reasons, and know that when used properly their distortion is sufficiently low that it will be inaudible and so can be ignored. Mostly, they dislike 'tube sound' as a phrase.

Lets have two people who need glasses.
They are looking at a fine art painting.

One of them wears thick distorted glasses. This person is the tube sound.
The other person wears thin high tech contacts.

I would rather be the person with the high tech contacts that allows me to see the picture more the way its viewed with pure eyes, then with some form of distortion thats "Warmer" that technically, and logically, distorts the image.

Why would a distorted image be a better thing then an undistorted image?
You want accuracy, extremely low distortion, high response, hi fidelity.

 

[lcsaszar]

For the same reason as Monet's Water Lilies generates more emotion than the photograph of the same nature scene...

 

[Rainwulf]

Exactly. You want to look at the image with as little possible distortion between you and the image.
Adding distortion, even it it sounds more pleasing to the ear, is STILL ADDING DISTORTION.

 

[ja2dhc]

The relationship between Distortion and Sound is described
in the following article which is written by Pete Millett.

the Sound of Distortion

My most favorite Tube Sound comes from my 211 SET amp
with Class-A2. This amp was designed by me.
see: 211 SET
(sorry written in Japanese).

Today's sound processor aims to close to 211SET with sound.

 

[DF96]

One of them wears thick distorted glasses. This person is the tube sound.

Yes. Not me, though. Read what I said.

extremely low distortion

Unnecessary, given the distortion from transducers. Low, yes; extremely low, mere specmanship. Pursuing one parameter to the nth degree will usually, in real engineering, mean too much compromise on other parameters.

 

[Rainwulf]

Sorry i generalised the parameters that a hi-fi person wants.

There is a whole whack of numbers in the parameters of an audio amp. You either want zero, 1, or infinity for each one of those parameters.

I understand that transducers cause distortion. You cant escape that, but thats no reason to accept distortion in an amplifier, then call it "tube sound".

 

[DF96]

There is a whole whack of numbers in the parameters of an audio amp. You either want zero, 1, or infinity for each one of those parameters.

Gain? Typically around 20.

Bandwidth? Must be restricted at both LF and HF ends to avoid problems - only a fool wants a 'DC to daylight' amplifier, unless he is running off batteries in the middle of a desert.

Output impedance? Zero could increase RF pickup on cables by making them a better resonator.

No, your approach (if you really believe what you said) is naive and oversimplistic.