diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   Hybrid "SE Tube Amp Sound Style" Chip Amp idea (http://www.diyaudio.com/forums/chip-amps/9348-hybrid-se-tube-amp-sound-style-chip-amp-idea.html)

ThorstenL 2nd January 2003 11:53 AM

Hybrid "SE Tube Amp Sound Style" Chip Amp idea
 
Hi Folks,

Here something for those that like to experiment a lot. Now everyone has been gaincloning away like they are going out of fashion. Great. But maybe here is the seasoning for those that like "toobz sound".

There are several key ingrediences to "SE Amp sound". I am sure some can be "assembled" by using a hybrid topology. Given that most who have tried agreed that the "inverted gainclone" (being inverted it clearly is NOT a Clone of the Gaincard BTW) sounds very transparent, how about combining this with a suitable Valve preamp circuit and using a circuit that allows the Output Impedance to be matched closely to an SE valve Amp?

So here my take:

Use the linestage of the Arthur Loesch Preamp. If a 5687 is used the gain of this linestage into a 15k load is around 20 and sound is good. For those who lack a schematic, the simplified circuit would use 1/2 5687 with a 10k Anode Load resistor, a 220 Ohm cathode (Bias) resistor with a 470uF Cathode Bypass Capacitor, 250V +B and a suitable output coupling capacitor, I'd say 2u2 should do.

We will be DELIBERATLY running the Valve at high signal levels on the Anode, so we get around 20V RMS to feed our "gainclone" circuit. Of course, such a level causes notable Distortion, 2nd harmonics dominant which allows us to cancel some of the Speaker Drivers (2nd Harmonic) distortion. For this it is best to use a Speaker with no X-Over on the woofer.

So, our "inverted gainclone" Circuit would be fed via a 15..22k series resistor. Let's call it 22kOhm, this with a 2u2 coupling cap gives a -3db point of 3Hz, about the same as the cathode combo.

Now with our valve stage giving already 20V RMS with 1V RMS on the Input we really do not want much gain in the Solid State output stage.

In fact, I'd say we want no gain at all, so the first step would be a 22k feedback resistor and a 2k4 resistor from the inverting input to ground to ensure stability (minimum closed loop gain = 10 for the LM3875). The resistor on the positive input to minimise offset should in theory be 2k4//22k - I think 2k2 is close enough.

So far so good, this is a simple inverted non-clone with a preceeding "sound shaping" Valve Stage where the Valve will dominate the sound and transfer function (for those interested in the why, read the Aleph vs Lamm Thread).

The next step would be to introduce mixed feedback which allows us a defined output impedance, which in most SET Amplifiers will be around 3 Ohm on the 8 Ohm tap and 1.5 Ohm on the 4 ohm tap (and obviously 6 Ohm on the 16 Ohm tap). For this we take the schematic and calculations from Rod Elliotts Site (he is on rare occasions useful instead of irritating to annoying) here:

http://sound.westhost.com/project56.htm

From his schematic (Figure 2) and comments if we still use our 0.22 Ohm resistor in the Output Circuit but this time in the ground leg would suggest a 1k5 resistor between the top of the 0.22 Ohm resistor and the negative input of the Chip.

With this the DCR on the negative input drops further and we need to adjust the Value of the resistor in the positive input or probably can just bridge it out.

I would not claim that the resulting Amplifier will match a really good SE Amplifier, but it will give a lot of chances to "shape" the sound of the Amplifier by selecting components and probably a good dash of "SE Style sound" and that with up to around 50 Watt power.

It would likely make a nice little project for someone with too much time on her or his hands and a decent Metalworkshop. The result would likely be something I would not mind at all to play around with on an extended evaluation (and if you come to collect don't be surprised if the door bell does not work and the phone is disconnected)....

Sayonara

halojoy 2nd January 2003 12:18 PM

Thanks Kwei yung tan
 
Thanks Kwuei.
As I always follows gurus blindly,
I am already on my way to the Electronic component supplier
here in Sweden.
To get all that is required to do Exactly! Everything
you say.

I bow before you my GOD (see profile)
--------------------

This idea is so superb.
SE tube + Inverted Gainclone.
Use Elliot's way to get "Variable Impedance amp"
to Get perfect Output.
-----------------------------------

Please post the Coimplete Schematic,
so I do not have to do any thinking by myself,
with all component values, please.

Thank you, thank you again! ;)

/halojoy - has found his new master
- will follow him to the end

Banfi T. 2nd January 2003 12:56 PM

Something happening in the kitchen! I can already smell it. A bit from tubes, a bit from IC's just like if we were putting different tastes together to get a harmony at the end...Will it work? I always wondered about this. Since I am a minimalist in many ways this aproach is a bit strange to me. Regarding this short description for what reason do you use the LM in a stage which is definitly not optimal for it? Just to "shape" the sound? How about neutrality? I think one of the reasons SE stages sound good is that they do not try to depend on symmetry and they have usually fairly simple overall design thus they provide a simple path to the speaker. But building an SE amplifier wich has an another power amplifier at the middle of the circuit just for - well I do not why - is definitly unique. :scratch1:

However, no one has the holy grall and it may happen that circuit comes alive and has a good sound.

best regards

bt

argo 2nd January 2003 03:07 PM

Mr Wang,

Actually this is quite interesting, except the mixed feedback for defined output impedance, which I donít grasp why one needs 3ohm Zout.

Argo

fcserei 2nd January 2003 03:54 PM

Quote:

Originally posted by argo
Mr Wang,

Actually this is quite interesting, except the mixed feedback for defined output impedance, which I donít grasp why one needs 3ohm Zout.

Argo

Why would you want ->0 output impedance? To model an ideal crossover with ideal driving impedance - which has nothing to do with reality?

ThorstenL 2nd January 2003 04:13 PM

Hi,

Quote:



Actually this is quite interesting, except the mixed feedback for defined output impedance, which I donít grasp why one needs 3ohm Zout.


Well, much depends upon ones view what is best to drive a loudspeaker. You can drive it from a low impedance to maximise certain of the speakers distortion and compression while providing maximum electrical damping to cone/dome resonances. You can drive it from a very high Impedance thus minimising certain distortion and compressions but failing to damp any mechanical resonances electrically.

Or you can choose a middleway that still works with many speakers and nevertheless improves distortion and compression. At any extent, the fairly high output impedance of a feedback less Valve Amp will influence the sound of the Speaker/Amplifier system and if it is not matched by the "SE Valve Amp soundalike hybrid" it will not even remotely sound alike....

Anyway, I cannot give all secrets here. But I suggest you only, always and ever think "system" not "subsystems". If you view Amplifier, Speaker and so on as seperate enteties that exist in a "pure" form where they are not influenced by the other you already lost.

Sayonara

halojoy 2nd January 2003 04:38 PM

Think System - not details
 
Quote:

Originally posted by Kuei Yang Wang
Hi,
-----------------------------------
But I suggest you only, always and ever think "system" not "subsystems". If you view Amplifier, Speaker and so on as seperate enteties that exist in a "pure" form where they are not influenced by the other you already lost.
Sayonara

This is very true!
Think system - the different parts should fit together
as good as possible.

If you have very good amplifier, but it does fit your loudspeakers,
neither the amplifier nor the loudspeaker
can perform up to its best potential.

Ciao!

halojoy - better known as gromanswe, who is the most "gromanized" person ever :D

analog_sa 2nd January 2003 04:50 PM

I instinctively dislike the idea. Too much unnecessary manipulation like tone controls. It may appeal if you like sound effects, but the virtues of the gainclone are not similarities to SE sound. Reminds me of these 'valve buffers' (like Musical Fidelity) which people use to reduce the resolution of digital and make it halfway listenable. Not my cup of tea.

cheers

peter

dhaen 3rd January 2003 03:23 AM

Actually Thorsten, that's a damn good idea... hic
Whether it works depends on more variables than you and I can think of simultaneously:xeye:

I am MUCH too busy with my SET at the moment.
So please...continue;)

ThorstenL 3rd January 2003 10:23 AM

Hi,

Quote:

Actually Thorsten, that's a damn good idea... hic
Whether it works depends on more variables than you and I can think of simultaneously:xeye:

Well, i have studies the variables a good while and I think i have the MAJOR ones covered, but then WTFDIK....

Quote:

I am MUCH too busy with my SET at the moment.

Me too, it's more fun too and carries a higher status....:D

I just thought I'll toss out the idea into the wide sea to see what comes back, since I had it.

Sayonara


All times are GMT. The time now is 02:49 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2